Professional Documents
Culture Documents
Lc249 Liability For Psychiatric Illness
Lc249 Liability For Psychiatric Illness
PART I: INTRODUCTION
2.1-2.66
2.12-2.51 2.13-2.46
13 13
2.13-2.15 2.16-2.18
13 14
2.19-2.33
16
19 20 21 22 24
(f) The plaintiff is an employee who is directly involved in an incident caused by his or her employers negligence which results in the injury or imperilment of another (g) Can the plaintiff who is a mere bystander ever recover? (2) Cases where the defendant has neither imperilled nor caused physical injury to any person (a) The plaintiff who suffers reasonably foreseeable psychiatric illness induced through stress at work (b) The plaintiff who suffers reasonably foreseeable psychiatric illness as a result of the defendant causing damage to property (c) Miscellaneous 3. THE CLASSIFICATION OF PRIMARY AND SECONDARY VICTIMS 4. TWO ADDITIONAL RESTRICTIONS ON RECOVERY (1) The shock requirement (2) The defendant as the immediate victim
2.41-2.44
25
26 27 27 28 29 30
33 33 36
SECTION C: REFORM
PART IV: REFORM I: INTRODUCTION PART V: REFORM II: FIVE GENERAL ISSUES
1. A RECOGNISABLE PSYCHIATRIC ILLNESS 2. REASONABLE FORESEEABILITY OF PSYCHIATRIC ILLNESS AND THE TEST OF REASONABLE FORTITUDE 4.1-4.7 5.1-5.54 5.1-5.6 5.7-5.27 55 57 57 59
ii
3. THE SHOCK REQUIREMENT (1) Arguments for abandoning the shock requirement (2) Arguments for retaining the shock requirement 4. RECOVERY WHERE THE IMMEDIATE VICTIM IS THE DEFENDANT 5. THE CLASSIFICATION OF PRIMARY AND SECONDARY VICTIMS
67 67 69 72 76
80 80 81 84 89 90
90 92 94 95 95 98 99 103
7.1-7.34 7.1-7.16
109 109
IMPERILMENT, BUT THE PLAINTIFF DOES NOT HAVE A CLOSE TIE OF LOVE AND AFFECTION WITH THE PERSON INJURED OR IMPERILLED (1) Rescuers (2) Involuntary Participants (3) Employees (4) Bystanders 2. WHERE THE PLAINTIFF SUFFERS PSYCHIATRIC ILLNESS OTHERWISE THAN AS A RESULT OF THE DEFENDANT CAUSING HIS OR HER OWN OR ANOTHER PERSONS PHYSICAL INJURY OR IMPERILMENT (1) Where the plaintiff suffers psychiatric illness through stress at work (2) Where the plaintiff suffers psychiatric illness as a result of the defendant negligently causing damage or danger to property (3) Where the plaintiff suffers psychiatric illness as a result of the negligent communication of distressing news 7.1-7.4 7.5-7.8 7.9-7.10 7.11-7.16 7.17-7.34 109 110 111 112 113
PART VIII: REFORM V: THE RELATIONSHIP BETWEEN OUR RECOMMENDED LEGISLATION AND THE COMMON LAW
1. THE CONTINUED EXISTENCE AND DEVELOPMENT OF THE COMMON LAW DUTY OF CARE 2. REPLACEMENT OF THE COMMON LAW DUTY OF CARE WHERE THE NEW STATUTORY DUTIES OF CARE APPLY 3. COMMENCEMENT OF THE LEGISLATION
8.1-8.9
118
8.1-8.3 8.4-8.7
118 119
8.8-8.9
120
SECTION D: SUMMARY
121
PART IX: SUMMARY OF RECOMMENDATIONS APPENDIX A: DRAFT BILL WITH EXPLANATORY NOTES APPENDIX B: LIST OF PERSONS AND ORGANISATIONS WHO COMMENTED ON CONSULTATION PAPER NO 137
127
135
iv
See, eg, articles in The Times 29 March 1995, The Guardian 29 March 1995, The Independent 29 March 1995 and the Daily Mail 30 March 1995 relating to our Consultation Paper on this topic; the correspondence published in response in The Independent 31 March, 1, 4, 5 and 7 April 1995; the coverage of the out-of-court settlements made to certain police officers who attended at the Hillsborough disaster in the Daily Mail 4 June 1996, The Guardian 4, 5 and 6 June 1996, The Daily Telegraph 5 June 1996 ; the coverage of the Court of Appeals decision in favour of the police officers in Frost v Chief Constable of South Yorkshire Police in The Times 1 November 1996, The Guardian 1 November 1996, The Independent 1 November 1996, Police Review 29 November 1996, and The Sunday Times 25 May 1997; and the articles reporting on the successful claim for psychiatric illness by a halfbrother of a Hillsborough victim in The Times 12 December 1996, The Guardian 12 December 1996, The Evening Standard 12 December 1996 and The Independent 19 February 1997. Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310. This was a test case brought by 16 relatives and friends of some of the victims. All but one were ultimately unsuccessful. Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194. The Court of Appeal (by a majority) found in favour of all but one of the officers. See para 1.5 n 14 below.
claims. The apparent injustice of this position has been acknowledged by judges,4 newspapers,5 MPs6 and legal commentators.7 1.2 The issue of liability for psychiatric illness provokes a range of strongly-held opinions. At one end of the scale are those who argue that the same principles that apply to liability for physical injury should be applied to liability for psychiatric illness, and that there is no legitimate reason to impose special restrictions in respect of claims for the latter.8 At the other extreme are those who argue that liability for psychiatric illness should be abandoned altogether. They say that the arbitrary rules which are required to control potential liability are so artificial that they bring the law into disrepute.9 We agree that, in some areas at least, the present law relating to liability for psychiatric illness is unsatisfactory. We do not accept, however, that all those whose sufferings have been described by the House of Lords as no less real and frequently no less painful and disabling10 than physical injuries should be denied the possibility of claiming compensation when their illness results from another persons negligence. Nor, at the other extreme, do we believe that liability for psychiatric illness should, in all cases, be equated with liability for physical injury. We set out in Part VI our reasons for believing that certain policy restrictions should remain in order to restrict the scope of potential liability. The aim of the proposals that we outline in this Report is to remove what we believe to be unnecessary constraints on claims for negligently inflicted psychiatric illness
1.3
I am aware that many people regard it as fundamentally unjust that the police should recover damages for PTSD sustained on that terrible day while the relatives claiming in Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310 failed: Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194, 1220, per Henry LJ. An editorial, Putting safety first - at last, published in The Guardian on 5 June 1996 suggested that the relatives of Hillsborough victims had every reason to feel affronted following substantial out-of-court settlements in favour of police who entered the affected pens, when the relatives had received only paltry sums or nothing at all. An early day motion was signed by several Members of the House of Commons stating that this House recognises the compensation awards made to police officers traumatised in the Hillsborough disaster; and believes that natural justice demands similar levels of compensation for traumatised family members of those killed and injured in the Hillsborough tragedy: EDM 121 1996/1997. N J Mullany and P R Handford, Hillsborough Replayed (1997) 113 LQR 410, 417: Lurking not far beneath the surface of the judgments [in Frost] is an uneasiness that if it is acceptable to compensate the police officers as a result of their involvement at Hillsborough, the spectators whose claims were denied by the House of Lords in Alcock have had a rough deal. ... The only satisfactory solution is to admit that the House of Lords decision in Alcock brought about the wrong result and that the law should move much closer to the first instance decision of Hidden J in that case. This view is put forward most forcefully by N J Mullany and P R Handford in their seminal book on this subject, Tort Liability for Psychiatric Damage (1993). This view is cogently expressed by Dr J Stapleton, In Restraint of Tort in P Birks (ed), The Frontiers of Liability (1994) vol 2, pp 94-96. She describes the law relating to liability for psychiatric illness as the area where the silliest rules now exist and where criticism is almost universal. McLoughlin v OBrian [1983] 1 AC 410, 433, per Lord Bridge.
10
thereby alleviating the arbitrariness of the current law, but without giving rise to fears of uncontrolled liability. 1.4
11 In March 1995 we published a Consultation Paper on this subject. We received a very large number of responses (150), and we have derived enormous assistance from them. A list of those who responded to the Consultation Paper is set out in Appendix B. We are very grateful for the time and effort spent by consultees. Of particular importance to us were the responses from those with medical expertise who were able to provide us with guidance on the causes, effects and prevalence of psychiatric illness.
1.5
In considering the need for legislation, we have been conscious of the fact that the common law in relation to many areas of liability for psychiatric illness is still developing. Even since the publication of our Consultation Paper there have been a number of relevant decisions.12 Two are of particular importance. In Page v Smith,13 the House of Lords (Lords Keith and Jauncey dissenting) held that where a person suffered psychiatric illness as a result of his or her direct involvement in an accident, general principles of negligence applied so that the rules relating to liability were no different from those which would have applied if the plaintiff had suffered a physical injury. In Frost v Chief Constable of South Yorkshire Police,14 referred to above, the Court of Appeal (Judge LJ dissenting) awarded compensation to police officers who had suffered psychiatric illness as a result of carrying out their professional duties at the scene of the Hillsborough football stadium disaster. We have no desire to restrict judicial activity in this area. We feel that it is important that the law should be able to develop incrementally as relevant experts learn more about psychiatric illness and society further recognises its debilitating consequences. We do not think that medical knowledge has advanced to a sufficiently mature stage for the complete codification of liability for psychiatric illness to be a sensible option. We have therefore adopted an approach of minimalist intervention, proposing legislative reform only in those areas where the
1.6
11
Liability for Psychiatric Illness Consultation Paper No 137. Page v Smith [1996] AC 155; Page v Smith (No 2) [1996] 1 WLR 855; Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194; Duncan v British Coal Corporation [1997] 1 All ER 540; Vernon v Bosley (No 1) [1997] 1 All ER 577; McCarthy v Chief Constable of South Yorkshire Police (unreported, 11th December 1996); Hegarty v EE Caledonia Ltd [1997] 2 Lloyds Rep 259; Young v Charles Church (Southern) Ltd, The Times 1 May 1997. [1996] AC 155. [1997] 3 WLR 1194. The Chief Constable is appealing to the House of Lords. While we recognise that the issues raised in the appeal will be relevant to some of the issues that we discuss in this Report, we have concluded that it would not be sensible for us to delay publication of the Report until after the decision of the House of Lords. The case primarily concerns liability to rescuers, about which we make no specific legislative recommendations. In any event, our overall strategy, as will be seen, is to recommend minimal legislative intervention while allowing the common law to develop. Furthermore, we hope that this Report will be of assistance to the House of Lords.
12
13
14
present law is clearly unsatisfactory, but in all other cases leaving the common law to develop.15 1.7 With this in mind - and in line with the central provisional proposals in our Consultation Paper (which, as we shall see, were supported by the overwhelming majority of consultees) - our recommendations for legislative reform deal primarily with one particular class of plaintiff: those who suffer psychiatric illness as a result of the death, injury or imperilment of a loved one. It is in relation to this class of plaintiff that most criticism of the current position has been made16 and that judges have called for legislative intervention.17 We recommend that, where it is reasonably foreseeable that such a plaintiff might suffer psychiatric illness, the plaintiff s proximity to the scene of the accident, and the manner by which he or she learns of it, should not be used as criteria to restrict the claim.18 In addition, we make two recommendations that are of general application to psychiatric illness claims. First, the requirement that the psychiatric illness be induced by a shock should be abandoned.19 And secondly, where the plaintiff s psychiatric illness is suffered as a result of another persons death, injury or imperilment, it should not be an absolute bar to recovery that that person is the defendant him or herself.20 This Report (like our Consultation Paper) is not concerned with liability for psychiatric illness suffered as a consequence of physical injuries of the plaintiff. Compensation for such illness is available on the same terms as compensation for any physical injury,21 the courts seemingly being satisfied that any policy-based arguments which have been advanced to justify the imposition of limiting controls in cases of mere psychiatric illness22 are not relevant where the plaintiff has suffered a physical injury. Nor does this Report deal with liability outside the tort of negligence. Consultees (impliedly) confirmed our initial view that it is exclusively in relation to the tort of
1.8
1.9
15
See further paras 4.1 to 4.3 below. See, for example, F A Trindade, The Principles Governing the Recovery of Damages for Negligently Caused Nervous Shock [1986] CLJ 476, 485-495; A Unger, Undue Caution in the Lords (1991) 141 NLJ 1729; S Hedley, Hillsborough - Morbid Musings of a Reasonable Chief Constable [1992] CLJ 16; K J Nasir, Nervous Shock and Alcock: The Judicial Buck Stops Here (1992) 55 MLR 705; H Teff, Liability for Psychiatric Illness after Hillsborough [1992] OJLS 440; K Wheat, Nervous Shock: Proposals for Reform [1994] JPIL 207; P R Handford, Compensation for Psychiatric Injury: The Limits of Liability (1995) 2 Psychiatry, Psychology and Law 37; M A Jones, Liability for Psychiatric Illness - More Principle, Less Subtlety? [1995] 4 Web JCLI. Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 417, 419, per Lord Oliver; the Court of Appeal in Alcock [1992] 1 AC 310, 376, per Stocker LJ; and McLoughlin v OBrian [1983] 1 AC 410, 429-431, per Lord Scarman. See para 6.16 below. See para 5.33 below. See further para 5.43 below. We do, however, recommend that the courts should have scope to decide not to impose a duty of care where the defendant chose to cause his or her death, injury or imperilment. Page v Smith [1996] AC 155, 187, per Lord Lloyd; M A Jones, Textbook on Torts (5th ed 1996) p 125. But see Page v Smith [1996] AC 155, 172, per Lord Jauncey. See para 6.6 below.
16
17
18
19
20
21
22
negligence that special restrictions have been adopted by the courts in relation to 23 It would seem, therefore, that any legislative liability for psychiatric illness. intervention in relation to other torts is not only unnecessary at this stage but would impose a new restriction on the recovery of damages for psychiatric illness.24 1.10 Having said that, we recognise that one could argue that any special rules that apply in relation to recovery for psychiatric illness for negligence should extend to recovery for psychiatric illness for other torts, such as public or private nuisance, breach of statutory duty or liability under Rylands v Fletcher.25 If not, a plaintiff who is unable to recover damages in negligence because of the application of special restrictions, might be able to avoid their application and make a successful claim by framing the action for damages under another tort.26 However, not all the reasons advanced for restricting liability for psychiatric illness in negligence apply with equal force to other torts. One of the most frequently cited arguments for limiting liability for psychiatric illness in negligence claims is the fear of a proliferation of claims arising from a single event.27 The weight of this objection is considerably weaker in relation to claims for the tort of private nuisance,28 breach of statutory duty29 and possibly Rylands v Fletcher liability.30 Further reasons may
23
Our Consultation Paper asked consultees whether they had encountered special restrictions (ie restrictions not applied to liability for physical injury) in respect of claims for psychiatric illness not based on the tort of negligence: Consultation Paper No 137, para 1.6. None reported that they had, although two pointed out that special rules may apply where the rules of negligence have been replaced by liability under an international convention. See further paras 6.43 to 6.47 below. In the Consultation Paper we asked for views as to whether we were correct in our understanding that, in actions for breach of contract, no special restrictions are placed on the recovery of damages for psychiatric illness and whether, if that is so, consultees considered, as we do, that that is a justified approach for the law to take: Consultation Paper No 137, para 1.6. The vast majority of consultees who considered these issues confirmed our understanding and agreed that this is a justified approach. We do not intend to consider contractual liability in this Report. (1868) LR 3 HL 330. For example, in Hegarty v EE Caledonia Ltd [1997] 2 Lloyds Rep 259 the plaintiff, a workman on board a support vessel at the scene of the Piper Alpha disaster, claimed damages for psychiatric illness against the owners of the rig both in negligence and for breach of statutory duty imposed by regulations enacted under the Mineral Workings (Offshore Installations) Act 1971. Liability under the Act is strict (MacMillan v Wimpey Offshore Engineers and Constructors Ltd 1991 SLT 515) and the Court of Appeal held that the action for breach of statutory duty, if a good one, would with one bound free [the plaintiff] from the control mechanisms imposed on the claims of secondary victims by the House of Lords in Alcock [1997] 2 Lloyds Rep 259, 268, per Brooke LJ. See also, Dooley v Cammell Laird [1951] 1 Lloyds Rep 271; Young v Charles Church (Southern) Ltd, The Times 1 May 1997 and M Napier and K Wheat, Recovering Damages for Psychiatric Injury (1995) pp 16-17. The validity of this floodgates argument in relation to negligence claims is considered at para 6.8 below. Private nuisance protects interests in land and the potential class of plaintiff is therefore limited. An action can generally only be brought by a person with an interest in the land: Hunter v Canary Wharf [1997] AC 655 (HL). The potential class of plaintiff may also be limited in an action for breach of statutory duty. When considering whether to impose tortious liability for breach of statutory duty where the legislation is silent on the question of a civil remedy, the court may do so if it is apparent that the obligation or prohibition was imposed for the benefit or protection of a
24
25
26
27
28
29
be put forward for not extending any special restrictions on liability in negligence to other torts.31 For example, damages are recoverable without special restrictions in private nuisance for mental distress caused by the interference with the enjoyment of ones own land,32 so that to impose limitations on recovery for psychiatric illness would seem odd. In relation to breach of those statutory duties which Parliament has specifically provided will give rise to civil liability for injury, which includes impairment of mental condition,33 it is doubtful whether any special common law restrictions could be imposed, and all-embracing legislative reform would not seem appropriate. If reform were thought necessary, this would need to be considered on a statute by statute basis.34 While we have limited our considerations to the tort of negligence, we do not seek to deny that our proposed reforms could be adopted in relation to other torts if, in future, any special restrictions were deemed necessary.
particular class of individuals: Lonrho Ltd v Shell Petroleum Co Ltd (No 2) [1982] AC 173, 185, per Lord Diplock. Note, however, that civil liability may also be imposed where the statute creates a public right and an individual member of the public suffers particular, direct and substantial damage other and different from that which was common to all the rest of the public: Lonrho v Shell Petroleum Co Ltd (No 2) [1982] AC 173, 185, per Lord Diplock. The true ambit of this second class has been described as a matter of conjecture, with several factors pointing towards its having limited use: K M Stanton, Breach of Statutory Duty in Tort (1986) p 49-52.
30
It is not clear whether a plaintiff must have an interest in land affected by the escape in order to maintain an action under Rylands v Fletcher. See M A Jones, Textbook on Torts (5th ed 1996) p 309; Salmond & Heuston on the Law of Torts (21st ed 1996) pp 310-312; Street on Torts (9th ed 1993) p 388. Damages for psychiatric illness caused by an intentional tort appear to be available without special restriction (assuming that the tort is one that protects against personal injury rather than, eg, economic loss). In Wilkinson v Downton [1897] 2 QB 57 the defendant, as a practical joke, deliberately and falsely told the plaintiff that her husband had been injured in a road accident. The plaintiff suffered severe shock and became seriously ill. Wright J held that the plaintiff was entitled to recover in tort for the physical harm, in this case the psychiatric illness, which she suffered as a result of the defendants wilful act. We do not consider psychiatric illness caused by an intentional tort any further in this Report. See, eg, Pelmothe v Phillips (1899) 20 LR (NSW) 58. In St Helens Smelting Co v Tipping (1865) 11 H.L.C. 642, 650, Lord Westbury referred to an action for nuisance based on personal discomfort as being based on the personal inconvenience and interference with ones enjoyment, ones quiet, ones personal freedom, anything that discomposes or injuriously affects the senses or the nerves. Note that damages for economic loss (recovery for which, like damages for psychiatric illness, is subject to special restrictions in the tort of negligence) are available without special restrictions for breach of statutory duty provided that the plaintiff can show that financial damage is the type of damage that the statute is intended to protect: see, eg, Rickless v United States Artists Corporation [1988] QB 40. See also K M Stanton, Breach of Statutory Duty in Tort (1986) pp 26-27. See, eg, s 11 of the Animals Act 1971, s 45(1) of the Consumer Protection Act 1987, s 1(9) of the Occupiers Liability Act 1984 and s 11(7) of the Mineral Workings (Offshore Installations) Act 1971. The Occupiers Liability Act 1957 does not specifically refer to impairment of a mental condition, but rather imposes a duty on an occupier to see that his visitor will be reasonably safe. If this were construed as covering mental injury, then it would appear that, as with the Occupiers Liability Act 1984, no common law restrictions could be imposed limiting those entitled to recovery. See further para 6.48 below. It may be that the courts would feel able to act with a greater degree of flexibility and impose additional restrictions if deemed necessary in those cases where the legislation is
31
32
33
34
1.11
Critics of the present fault-based tort system will no doubt argue that enhancing rights to claim damages for psychiatric illness will favour the already privileged minority of accident victims, and that wholesale reform of the current 35 compensation system for personal injury would be fairer and more efficient. We should therefore make clear at the outset that our terms of reference require us to recommend improvements to the existing tort system, rather than advocating its replacement.36 We recognise that, in practice, the cost of the recommendations which we have made in this Report is likely to be borne by a large section of the public through higher insurance premiums. In order to assess how great that impact might be, we asked the Association of British Insurers (ABI) if they could provide us with an estimate of the increase in premiums that our proposals would have in one particular sector, the motor insurance market. We are most grateful for the help that the ABI has given us on this issue. It is clear that any estimate must be a very speculative one. In particular, whatever methodology is used, it is necessary to predict the increase in the number of personal injury claims that would be brought as a result of our proposals. Since there is no existing data on which to base this prediction, any figure chosen must be based on insurers general feel and understanding of the market. Our discussions with the ABI suggested that a reasonable assumption seems to be that our proposals would give rise to a ten per cent increase in the number of personal injury claims. Using this assumption, one might then calculate that an insurance company which currently spends 100 million on personal injury claims out of a total 500 million spent on motor claims altogether would see an increase in claims of 10 million if our proposals were to be implemented, representing a two per cent rise overall. Premiums would therefore have to be increased by at least two per cent to cover this rise. These figures, however, are based on two further assumptions that will not always be borne out. The first assumption is that the average cost of a claim made under our proposals would be the same as the average cost of a personal injury claim made today. If the average cost of a psychiatric illness claim under our proposals were in fact higher, then the percentage increase in cost would be greater. The second assumption relates to the ratio of spending on personal injury claims as opposed to vehicle damage claims. The proportion of a motor insurance premium which covers personal injury as opposed to vehicle damage is greater in a third party policy than in a comprehensive policy. So, if, in the example above, the insurance company held a larger proportion of third party policies and spent 100 million on personal injury claims out of a total of only 200 million on motor claims altogether, then the 10 million increase would represent a five per cent rise overall. It is because of the various assumptions that must be made in attempting to pinpoint any figure, that it is very difficult to reach any firm estimate. However, we understand from our discussions with the ABI, using the sort of
1.12
1.13
silent on the issue of civil liability for breach of statutory duty, but the courts have found it to exist (that is, in relation to the tort of breach of statutory duty in its classic sense).
35
A few consultees, including Professor J A Jolowicz, Professor R Lewis and Dr J Stapleton, were strongly of this view. See also P S Atiyah, The Damages Lottery (1997) p 32. See especially the description of Item 11 of the Fifth Programme of Law Reform (1991) Law Com No 200.
36
methodology we have here set out, that it is reasonable to estimate that our proposals would give rise to an increase in motor insurance premiums in the range 37 of two to five per cent. 1.14 The rest of this Report is arranged as follows. Part II sets out the present law. Part III contains a brief summary of the medical background. In Parts IV to VIII we set out our proposals for reform. Part IX contains a summary of our recommendations. A draft Bill to give effect to our recommendations is to be found in Appendix A. Appendix B contains a list of those who responded to the Consultation Paper. We gratefully acknowledge the invaluable assistance of:- Professor Simon Wessely of Kings College, London and Mr Miles Mandelson of St Helens & Knowsley Hospitals, Merseyside, both of whom read and commented on Section B, The Medical Background; the Association of British Insurers (ABI) for their assistance in assessing the practical impact of our proposals on insurance premiums; and Mrs Ann Smart of St Hughs College, Oxford, who carried out the analysis of consultation. We are also grateful to the following for their help: Lord Justice Phillips, Ben Hytner QC, Robert Webb QC, Adrian Whitfield QC, Charles Haddon-Cave, Patrick Griggs, Richard Williams, Mr J W Davies of Brasenose College, Oxford, Professor Michael Jones of the University of Liverpool, Professor Andrew Tettenborn of the University of Exeter, Professor Nicholas Gaskell of the University of Southampton, the Department of Environment, Transport and the Regions and the Treasury Solicitors Department.
1.15
37
We have not attempted to assess the likely impact of our proposals on other sectors of the insurance market. However, in those sectors where the bulk of the premium covers personal injury (as, for example, is the case with employers liability insurance) the percentage increase is likely to be higher than that estimated for motor insurance premiums, where the part of the premium covering vehicle damage would be unaffected.
2.2
Dulieu v White & Sons [1901] 2 KB 669. Ravenscroft v Rederiaktieblaget Transatlantic [1991] 3 All ER 73, 76, per Ward J. [1996] AC 155. The previous House of Lords decisions being Bourhill v Young [1943] AC 92, McLoughlin v OBrian [1983] 1 AC 410 and Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310. See Lord Macmillan in Bourhill v Young [1943] AC 92, 103: [I]n the case of mental shock there are elements of greater subtlety than in the case of an ordinary physical injury and these elements may give rise to debate as to the precise scope of legal liability. See Lord Lloyd in Page v Smith [1996] AC 155, 188: In an age when medical knowledge is expanding fast, and psychiatric knowledge with it, it would not be sensible to commit the law to a distinction between physical and psychiatric injury, which may already seem somewhat artificial, and may soon be altogether outmoded. McLoughlin v OBrian [1983] 1 AC 410, 431. Hinz v Berry [1970] 2 QB 40, 42, per Lord Denning MR. Hinz v Berry [1970] 2 QB 40. Brice v Brown [1984] 1 All ER 997.
10
11 12 post-traumatic stress disorder, pathological grief disorder and chronic fatigue syndrome (CFS).13 Expert medical evidence will generally be required to establish that the plaintiff has suffered a recognisable psychiatric illness. The ordinary emotions of anxiety, fear, grief14 or transient shock are not conditions for which the law gives compensation.15
(2) The Test of Reasonable Foreseeability (a) Reasonably foreseeable psychiatric illness 2.4 Dulieu v White & Sons16 was the first English case to recognise a cause of action for negligently inflicted psychiatric illness (not consequent on any physical impact). In that case, Kennedy J suggested that recovery should only be available when the plaintiff s illness arose from a reasonable fear of injury to him or herself. This limitation was, however, soon rejected by the Court of Appeal in Hambrook v Stokes Bros17and replaced with a more liberal test: the plaintiff was required to establish that his or her psychiatric illness was a reasonably foreseeable consequence of the defendants conduct. In Hambrook a mother saw a lorry careering out of control from the direction in which she had just left her children. She was so terrified by her fear for their safety that she became ill and later died. The court found it objectionable that a plaintiff should be able to recover damages for an illness suffered because of fear for him or herself, but not when it resulted from fear for another.18 Instead, in an action brought by the father under the fatal accidents legislation, the majority held that, had the mother survived, she would have been entitled to damages on the ground that the defendant should have
2.5
11
Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194. Vernon v Bosley (No1) [1997] 1 All ER 577. Page v Smith [1996] AC 155. Although there is a limited statutory right of action for bereavement: Fatal Accidents Act 1976, s1A. See Claims for Wrongful Death, Law Commission Consultation Paper No 148 (1997) paras 2.63 to 2.74. Page v Smith [1996] AC 155, 171, per Lord Jauncey; Hinz v Berry [1970] 2 QB 40 (no recovery for the inevitable grief and sorrow at losing a husband); Hicks v Chief Constable of the South Yorkshire Police [1992] 2 All ER 65 (fear of impending death suffered by victims of the Hillsborough disaster did not give rise to a cause of action which survived for the benefit of the victims estate); Nicholls v Rushton, The Times 19 June 1992 (shock and shaking up suffered after a road accident falling short of an identifiable psychiatric illness was not compensatable); Reilly and Reilly v Merseyside RHA [1995] 6 Med LR 246 (claustrophobia and fear suffered when trapped in a lift for over an hour did not give rise to a cause of action). See, however, the decision in Whitmore v Euroways Express Coaches Ltd, The Times 4 May 1984 where damages were awarded for ordinary shock which was more than emotional distress or worry but which did not amount to a psychiatric illness. The decision is generally regarded as an aberration: see M A Jones, Ordinary shock - thin skull rules OK? (1984-1985) 4 Lit 114. But see N J Mullany and P R Handford, Tort Liability for Psychiatric Damage (1993) pp 18-21 where it is suggested that the case may be part of a line of Commonwealth authority which questions the insistence on a recognisable psychiatric illness. [1901] 2 KB 669. [1925] 1 KB 141. [1925] 1 KB 141, 151, per Bankes LJ; 157, per Atkin LJ. Sargant LJ dissented.
12
13
14
15
16
17
18
10
anticipated that a mother in her position might be so terrified for her children that her health was injured. 2.6 However, [i]t is not every emotional disturbance or every shock which should have been foreseen.19 In Bourhill v Young20 the defendant had been speeding on his motor cycle, collided with a car and was killed. The plaintiff was some forty five feet from the accident on the far side of a tram. The crash occurred outside her range of sight, although she heard the collision and suffered fright resulting in nervous shock. In an action brought by the plaintiff against the defendants estate, the House of Lords held that the defendant did not owe her a duty of care since he could not reasonably have foreseen that a person in the position of the plaintiff would be affected by his negligent driving.21 Two particular points should be noted in relation to the application of the test of foreseeability of psychiatric illness where the plaintiff has suffered psychiatric illness as a result of an injury or fear of injury to another person. First, in assessing whether psychiatric illness is reasonably foreseeable, the defendant, unless he or she has special knowledge to the contrary, may assume that the plaintiff is a person of customary phlegm22 and has a normal standard of susceptibility.23 What danger of particular infirmity that would include must depend on all the circumstances.24 However, once the plaintiff has established that it is reasonably foreseeable that a person of reasonable fortitude would suffer some psychiatric illness, then the normal eggshell skull or thin skull rule of remoteness of damage applies, so that the susceptible plaintiff may recover for the full extent of the illness.25 Secondly, foreseeability of the psychiatric illness is considered ex post facto in the light of all that has happened.26 Unless hindsight is used, [t]he question ceases to be whether it is foreseeable that a reasonably robust person would have suffered psychiatric illness as a result of what actually happened and becomes instead whether it is foreseeable that such a person would have suffered psychiatric illness as a result of what might have happened but did not in fact do so.27
2.7
2.8
19
Bourhill v Young [1943] AC 92, 117, per Lord Porter. [1943] AC 92. Ibid, 98, per Lord Thankerton; 102, per Lord Russell; 105, per Lord Macmillan; 111, per Lord Wright; 117-119, per Lord Porter. Ibid, 117, per Lord Porter. Ibid, 110, per Lord Wright. See also, eg, Lord Russell of Killowen in McLoughlin v OBrian [1983] 1 AC 410, 429. Bourhill v Young [1943] AC 92, 110, per Lord Wright. Brice v Brown [1984] 1 All ER 997. See also M A Jones, Ordinary shock - thin skull rules OK? (1984-1985) 4 Lit 114, 116-118. The quantum of damages may be reduced to take into account the fact that the plaintiff might at some point have suffered the illness in any event: Page v Smith (No 2) [1996] 1 WLR 855, 857. Bourhill v Young [1943] AC 92, 110, per Lord Wright; McLoughlin v OBrian [1983] 1 AC 410, 420, per Lord Wilberforce; 432, per Lord Bridge; Page v Smith [1996] AC 155, 188, per Lord Lloyd. Page v Smith [1996] AC 155, 179, per Lord Jauncey.
20
21
22
23
24
25
26
27
11
2.9
In applying the test of reasonable foreseeability, the judge should treat himself as representative of the reasonable man. As Lord Bridge noted in McLoughlin v OBrian,28 the judge could either receive the evidence of psychiatrists as to the degree of probability that the particular cause would have a particular effect, or, relying on his own opinion of the operation of cause and effect in psychiatric medicine, as fairly representative of that of the educated layman, [the judge] should treat himself as the reasonable man and form his own view from the primary facts as to whether the proven chain of cause and effect was reasonably foreseeable. Lord Bridge said that the latter option had been the one always taken and in any event was to be preferred. The consensus of informed judicial opinion was the best yardstick by which to determine whether the harm was foreseeable in law. (b) The distinction between a primary and a secondary victim and the test of reasonably foreseeable personal injury (whether physical or psychiatric)
2.10
Until the House of Lords decision in Page v Smith29 it was generally assumed that the test of reasonable foreseeability of psychiatric illness applied in all cases where the plaintiff claimed damages for negligently inflicted psychiatric illness30 (except where the psychiatric illness was consequent on a physical injury).31 However, in Page, the House of Lords held (by a majority) that foreseeability of psychiatric illness was not the correct test where the plaintiff was directly involved in the accident and well within the range of foreseeable physical injury.32 In such circumstances the plaintiff was to be regarded as a primary, rather than a secondary, victim of the accident,33 and the duty of care owed by the defendant to him or her was the same duty of care not to cause personal injury which applies in cases of physical harm. Following the decision in Page, in cases involving primary victims (as described above) there is no need to distinguish between physical injury and psychiatric illness in applying the reasonable foreseeability test. It is sufficient to ask whether the defendant should reasonably have foreseen that the plaintiff might suffer some personal injury (physical or psychiatric) as a result of the defendants negligence. If so, the defendant is under a duty of care to avoid causing personal injury (physical or psychiatric). It is unnecessary to ask whether the defendant should
2.11
28
[1983] 1 AC 410, 432. [1996] AC 155. King v Phillips [1953] 1 QB 429, 441, per Denning LJ; F A Trindade, The Principles Governing the Recovery of Damages for Negligently Caused Nervous Shock [1986] CLJ 476, 482; N J Mullany and P R Handford, Tort Liability for Psychiatric Damage (1993) p 70; Liability for Psychiatric Illness Consultation Paper No 137, para 2.9. See para 1.8 above. [1996] AC 155, 184, per Lord Lloyd. On the distinction between a primary victim and a secondary victim see further paras 2.52 to 2.60 below.
29
30
31
32
33
12
reasonably have foreseen that the plaintiff might suffer psychiatric illness, and it is 34 irrelevant that the foreseeable physical injury did not in fact occur. 2. WHO MAY RECOVER? 2.12 Next, we examine which categories of plaintiffs (assuming that they can establish reasonable foreseeability and that they are suffering from a recognisable psychiatric illness) are entitled to recover damages for negligently inflicted psychiatric illness and the different rules that apply to them. Traditionally, claims have been brought by plaintiffs for psychiatric illness suffered as a result of a fear of physical injury to themselves, or as a result of the physical injury or imperilment of another. The bulk of the case law covers this area. However, more recently claims have been brought by plaintiffs who have suffered psychiatric illness in other situations, where the defendant has not caused, or put any person at risk of, physical injury. Liability in these other areas is developing on a case by case basis. (1) Cases where the Plaintiff suffers Psychiatric Illness as a result of his or her own Imperilment (or reasonable fear of danger) or as a result of the Physical Injury or Imperilment of Another caused by the Defendant (a) The plaintiff is within the area of reasonably foreseeable physical injury 2.13 After Page v Smith,35 no special rules apply in relation to liability for psychiatric illness suffered by a plaintiff who was within the area of reasonably foreseeable physical injury created by the defendants negligence. Provided that the plaintiff can show that the defendant should reasonably have foreseen that his or her conduct might cause the plaintiff any personal injury (physical or psychiatric), the plaintiff may recover for psychiatric illness suffered as a result of the defendants negligence under normal tort principles, whether or not the foreseeable physical injury does in fact occur. This is illustrated by the straightforward facts of Page. The plaintiff was involved in a car accident of moderate severity caused by the defendants negligent driving. Neither the plaintiff, the defendant, nor any passenger suffered any physical injury. The plaintiff gave no evidence that he was in fear of his own or others safety. However, following the accident he suffered a recurrence of chronic fatigue syndrome (CFS) of such severity that it was unlikely that he would work again. The House of Lords held (by a majority) that since it was reasonably foreseeable that the plaintiff might suffer physical injury as a result of the defendants negligent driving, the defendant owed him a duty of care to avoid causing personal injury without more. It was not relevant to ask whether injury by way of psychiatric
2.14
34
Page v Smith [1996] AC 155, 190, per Lord Lloyd. See also paras 2.13 to 2.15 below. Lord Keith and Lord Jauncey dissented: they held that in all claims for psychiatric illness the plaintiff must show that psychiatric illness was reasonably foreseeable and in this case it was not:[1996] AC 155, 168-170, per Lord Keith; 178-180, per Lord Jauncey. [1996] AC 155.
35
13
illness was reasonably foreseeable either to the plaintiff (who had a history of CFS) or to a person of normal fortitude.36 2.15 The Page decision was applied by the Court of Appeal in Young v Charles Church (Southern) Ltd.37 The plaintiff, a construction worker, suffered psychiatric illness after witnessing the death of his colleague from a distance of some six to ten feet. His colleague was killed instantly when a scaffold pole, which the plaintiff had just handed him, brushed against a live overhead electric power line. Another colleague standing nearby also suffered burns. Unlike the car driver in Page, the plaintiff s psychiatric illness was caused by the impact upon him of the dreadful injuries and death of his colleague. However, the Court of Appeal held that that distinction was not significant. The plaintiff was at risk of physical injury from an accident which could be foreseen, and his illness was caused by the accident which did occur as a result of the defendants negligence. He was therefore a primary victim under the Page test and able to recover damages for his psychiatric illness.38 (b) The plaintiff is not actually in danger but, because of the sudden and unexpected nature of events, reasonably fears that he or she is in danger 2.16 In McFarlane v EE Caledonia Ltd39 Stuart-Smith LJ identified three situations in which a plaintiff may recover damages for psychiatric illness sustained through fear of physical injury to him or herself. The first situation is where the plaintiff is in the actual area of danger created by the event, but escapes physical injury by chance or good fortune. A plaintiff in this situation would now be able to rely on the Page40 decision and would come within category (a) above. The third situation relates to rescuers, and is dealt with in category (d) below. The second situation is more difficult. This is where the plaintiff is not actually in danger, but because of the sudden or unexpected nature of events, reasonably believes that he or she is.41 Such a plaintiff can recover, but to do so must show not only that he or she genuinely feared for his or her safety, but that the defendant should reasonably have foreseen that a person of ordinary fortitude in the plaintiff s position would
36
The case was referred back to the Court of Appeal to consider the issue of causation. In Page v Smith (No 2) [1996] 1 WLR 855 the Court of Appeal confirmed that the test of causation in Bonnington Castings Ltd v Wardlaw [1956] AC 613 was the correct test to apply and that the plaintiff had established that the negligence of the defendant had materially contributed to the recrudescence of his CFS. The decision is noted by A Sprince, Page v Smith (No 2) - the saga ends but the questions remain (1996) 12 PN 80. The Times 1 May 1997. The Times 1 May 1997; Transcript No QBENF 96/0920/C: see the judgments of Evans LJ at pp 12-13 of the transcript and of Hutchison LJ at pp 34-35 of the transcript. But see Hobhouse LJ who thought that the plaintiff was a secondary victim under the Page test, since his position was that of an observer of a traumatic and fatal injury inflicted upon another (at p 22 of the transcript). On the difficulties which the courts have encountered in applying the primary/secondary victim distinction, see further paras 2.57 to 2.59 below. [1994] 2 All ER 1. [1996] AC 155. Stuart-Smith LJ, giving the judgment of the court, said that an example of this type of case was Dulieu v White & Sons [1901] 2 KB 669, where the plaintiff was not in fact at risk of physical injury, but she naturally was put in fear for her own safety. But see Page v Smith [1996] AC 155, 191 where Lord Lloyd categorises the plaintiff in Dulieu as being within the area of foreseeable physical injury, although not in fact injured.
37
38
39
40
41
14
have done so. In McFarlane the plaintiff was a workman on a support vessel which went to the aid of the crew on board the burning Piper Alpha oil platform. He spent about two hours on the support vessel before being evacuated by helicopter and claimed that he had suffered psychiatric illness as a result. He failed to recover under this heading because it could not be said that the defendants ought reasonably to have foreseen that a person of ordinary fortitude in his position 42 would suffer psychiatric illness. Nor did the Court of Appeal (differing from the trial judge) accept from his evidence that the plaintiff had genuinely been in fear of his safety.43 In a later case brought by another workman who had been on the same support vessel, Hegarty v EE Caledonia Ltd,44 the plaintiff claimed that he fell within this second category identified by Stuart-Smith LJ in McFarlane. In Hegarty, the Court of Appeal accepted that the plaintiff was a person of reasonable fortitude and that he had genuinely feared that he was in danger. However, the Court agreed with the trial judges finding that this fear was not a reasonable one.45 2.17 It is not clear from the case law how much, if at all, this category adds to category (a). In other words, would the plaintiff s genuine fear only have been a reasonable fear if the plaintiff were at risk of foreseeable physical injury, and therefore covered by category (a) in any event? By stating that the plaintiff in this category need not have been in actual danger,46 the decisions seem to expand upon category (a). Yet a more restrictive approach seems to have been taken in Hegarty. Brooke LJ said that, once it had been found that the rescue vessel had come close to danger but was not in fact ever in danger, it was almost inevitable that the plaintiff s fear for his life should be found to be an irrational one. The law would not accommodate people who are not directly threatened but who genuinely and irrationally believe that they are, because their dilemma is not reasonably foreseeable.47 It is possible that such a dilemma would be reasonably foreseeable where the plaintiff was abnormally susceptible to fear from the particular present danger and the defendant was aware of this fact. Whether such a plaintiff would be able to recover under this category is open to question. It would involve the test of reasonable fear being assessed on a more subjective basis than has previously been the case. If that were to be the law, there would presumably be no need for the plaintiff or a third party to have actually been physically injured or imperilled by the defendant.
2.18
42
[1994] 2 All ER 1, 11-12. [1994] 2 All ER 1, 12. [1997] 2 Lloyds Rep 259. [1997] 2 Lloyds Rep 259, 271. McFarlane v EE Caledonia Ltd [1994] 2 All ER 1, 10, per Stuart-Smith LJ; Hegarty v EE Caledonia Ltd [1997] 2 Lloyds Rep 259, 266, per Brooke LJ. [1997] 2 Lloyds Rep 259, 271.
43
44
45
46
47
15
(c) The defendant causes the death, injury or imperilment of a person other than the plaintiff, and the plaintiff can establish sufficient proximity in terms of: (i) his or her tie of love and affection with the immediate victim; (ii) his or her closeness in time and space to the incident or its aftermath; and (iii) the means by which he or she learns of the incident48 2.19 While early cases dealing with plaintiffs who suffered psychiatric illness pursuant to another persons death, injury or imperilment established that the plaintiff must show that his or her psychiatric illness was reasonably foreseeable,49 it became clear that certain factors, such as the plaintiff s closeness in time and space to the scene of the accident50 and the plaintiff s relationship to the immediate victim51 were particularly important to the finding of liability. It was not initially clear, however, whether these were factors relevant to the test of foreseeability, or whether they were additional hurdles over and above foreseeability that the plaintiff must surmount in order to establish a duty of care. In McLoughlin v OBrian52 Lord Wilberforce (with whom Lord Edmund-Davies agreed) thought that reasonable foreseeability was not the sole test. The risk of opening the door to a limitless number of claims required that the law should impose additional proximity tests in terms of the class of persons whose claims may be recognised, the proximity of such persons to the accident, and the means by which the shock was caused.53 Lord Scarman and Lord Bridge thought that these three factors were to be weighed in applying the reasonable foreseeability test, but were not limitations on it.54 This issue was decisively dealt with in the decision of the House of Lords in Alcock v Chief Constable of South Yorkshire Police.55 This was a test case brought by a number of relatives and friends of spectators involved in the Hillsborough disaster. It was admitted that the death and injuries of the fans at the stadium occurred as a result of the negligence of the police and it was assumed for the purposes of the trial that each of the plaintiffs had proved the infliction of psychiatric illness.56
2.20
48
For the position where the person injured or imperilled is the defendant him or herself, see para 2.66 below. See para 2.5 above. Bourhill v Young [1943] AC 92 (plaintiff some 45 to 50 feet from the accident scene and out of visual range failed to recover); King v Phillips [1953] 1 QB 429 (a mother who heard her child scream from some 70 to 80 yards distance when a taxi backed into him failed to recover). Successful plaintiffs prior to Alcock included mothers (Hambrook v Stokes Bros [1925] 1 KB 141; Hinz v Berry [1970] 2 QB 40; McLoughlin v OBrian [1983] 1 AC 410; Brice v Brown [1984] 1 All ER 997), a father (Boardman v Sanderson [1964] 1 WLR 1317) and a spouse (McLoughlin v OBrian [1983] 1 AC 410). [1983] 1 AC 410. Ibid, 421-422. Ibid, 431 and 441-443 respectively. Lord Russells opinion on this point is not clear: ibid, 429. [1992] 1 AC 310. [1992] 1 AC 310, 318 (Hidden J); 351 (Parker LJ); 406 (Lord Oliver).
49
50
51
52
53
54
55
56
16
Sixteen plaintiffs claimed damages, and ten were successful at first instance. The Court of Appeal allowed the defendants appeal in respect of nine of these plaintiffs and denied the cross-appeals by the six unsuccessful plaintiffs. Ten of the fifteen plaintiffs appealed to the House of Lords. The relationship of these plaintiffs to the immediate victims ranged from parents to brother, sister, brotherin-law, fiance and grandfather. Two of the plaintiffs were present at the match, whilst the others had watched events on television either as the disaster unfolded on live broadcasts or subsequently on recorded bulletins. None of the plaintiffs were successful before the House of Lords. 2.21 The House of Lords unanimously adopted Lord Wilberforces view that liability for psychiatric illness was limited on policy grounds by the concept of proximity. This involved, in the words of Lord Oliver, not only an element of physical proximity to the event but a close temporal connection between the event and the plaintiff s perception of it combined with a close relationship of affection between 57 Therefore, even where the psychiatric the plaintiff and the primary victim. illness is a reasonably foreseeable consequence of the defendants conduct, if all three additional proximity requirements are not met, the claim will fail. These three proximity requirements are considered in turn below, after we have examined (in paragraphs 2.22 to 2.24) a preliminary point concerning the relationship between the defendant and the immediate victim. It has been suggested that in order to succeed under this category (c) the plaintiff must first establish that the defendant was in breach of a duty of care to the immediate victim.58 The principal reason for this appears to be that success by a plaintiff in a claim for psychiatric illness where the defendants conduct with regard to the immediate victim was not negligent would result in disparate legal standards of conduct being required from the defendant in the same circumstance.59 We do not consider that this is the position under the present law and think that such reasoning stems from the confusion created by the use of the terminology primary and secondary victims. It was Lord Oliver in Alcock v Chief Constable of South Yorkshire Police who first explicitly referred to the distinction between a primary and secondary victim in claims for psychiatric illness, and although he thought that it was useful terminology, he recognised the possibility of confusion
2.22
2.23
57
Ibid, 411. B J Rodger, Nervous Shock and Breach of Duty of Care Owed to Secondary Victims 1997 SLT 22. In Dillon v Legg, a decision of the Californian Supreme Court allowing recovery for psychiatric illness suffered by the mother of a girl killed in a car accident in front of her, Tobriner J stated: In the absence of the primary liability of the tortfeasor for the death of the child, we see no ground for an independent and a secondary liability for claims for injuries by third parties. The basis for such claims must be the adjudicated liability and fault of the defendant; that liability and fault must be the foundation for the tortfeasors duty of care to third parties who, as a consequence of such negligence, sustain emotional trauma: 29 ALR 3d 1316, 1320-1321 (1968). But see P G Heffey, The Negligent Infliction of Nervous Shock in Road and Industrial Accidents (1974) 48 ALJ 240, 251-254 and Lord Wright in Bourhill v Young: If, however, the appellant has a cause of action it is because of a wrong to herself. She cannot build on a wrong to someone else: [1943] AC 92, 108. Ibid, 23.
58
59
17
when he said: Although it is convenient to describe the plaintiff ... as a secondary victim, that description must not be permitted to obscure the absolute 60 essentiality of establishing a duty owed by the defendant directly to him. The plaintiff must therefore show that an independent duty of care is owed by the defendant to him or her, and there is no suggestion that such claim is parasitic on any claim that the immediate victim might have in respect of his or her personal injuries.61 Indeed it is clear that the plaintiff may be successful where he or she fears that another has been injured even though they are in fact unharmed.62 This point was clearly recognised by Lord Oliver when he said: There may, indeed, be no primary victim in fact. It is, for instance, readily conceivable that a parent may suffer injury, whether physical or psychiatric, as a result of witnessing a negligent act which places his or her child in extreme jeopardy but from which, in the event, the child escapes unharmed.63 Indeed, one of the plaintiffs in Alcock suffered illness consequent upon his fear for the safety of his nephew, who in fact escaped unharmed from the tragedy. Although ultimately unsuccessful,64 there was no suggestion that this plaintiff should fail because his relative was not injured. 2.24 We consider that this approach is justifiable, even where it results in the plaintiff being able to recover damages for psychiatric illness suffered pursuant to the injury of a loved one caused by the defendant in circumstances where the defendant would not be liable in negligence to the physically injured person.65 For example, the plaintiff may suffer psychiatric illness as a result of injuries inflicted by the defendant on a person who has agreed an exclusion clause exempting the defendant from liability for the injuries.66 This does not necessarily mean, however, that the defendant should be able to ignore the claims of any others who might foreseeably be injured by his or her acts, including those with a close tie of love and affection to the injured person. Likewise, the defendant may be able to rely on the defence of ex turpi causa to defeat an injured persons claim for damages, whereas there may be no similar public policy justification to deny the claim of a loved one who suffers psychiatric illness as a result.
60
Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 411. The position may therefore be contrasted with that of the dependant claiming under the Fatal Accidents Act 1976 whose success depends on whether the deceased would have been able to maintain an action at the moment of death had death not taken place: Fatal Accidents Act 1976, s1(1). For example, in Dooley v Cammell Laird [1951] 1 Lloyds Rep 271 and in Galt v British Railways Board (1983) 133 NLJ 870 the plaintiff recovered damages for psychiatric illness suffered after he mistakenly feared that his work colleagues had been injured. Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 412. The Court of Appeal held that he had failed to satisfy the proximity of relationship test: Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 361, per Parker LJ; 380, per Stocker LJ; 385, per Nolan LJ. He did not appeal to the House of Lords. Although see para 6.37 below in relation to policy issues that may negate both duties of care. Such an exclusion clause could only be valid in relation to non-business liability: Unfair Contract Terms Act 1977, ss 1(3), 2(1).
61
62
63
64
65
66
18
(i) a close tie of love and affection 2.25 The plaintiff must establish a close tie of love and affection to the immediate 67 Such a tie may be present in family relationships or those of close victim. friendship.68 In Alcock, all their Lordships were agreed that the closeness of the tie must be proved in each case by the plaintiff, although it may be rebuttably presumed in the case of a spouse, parent or child69 and possibly fianc(e).70 More distant relatives and friends are required to show that their relationship is so close and intimate that their love and affection for the victim is comparable to that of the normal spouse, parent or child,71 but it is the closeness of the care as opposed to the nature of the relationship which is important.72 In Alcock one plaintiff had been present at the ground and had witnessed the accident in which his two brothers were killed. His claim failed because he produced no evidence of a close tie of love and affection with his brothers and no presumption of such a tie was to be made in the case of siblings.73 However, in a subsequent action, McCarthy v Chief Constable of South Yorkshire Police,74 a plaintiff whose half-brother had died at Hillsborough successfully recovered damages for the psychiatric illness which he suffered. He adduced evidence from relatives and friends that his family was very close, and the two half-brothers particularly so.75 There was general agreement amongst their Lordships in Alcock that the issue of proximity by relationship should be decided on a case by case basis. Lord Oliver stated that creating a list of categories within which claims may succeed and without which they are doomed to failure would work great injustice and could not be rationally justified. Lord Jauncey thought that any such dividing line would be arbitrary and lacking in logic.76
2.26
2.27
67
Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 397, per Lord Keith. Ibid, 397, per Lord Keith. Ibid, 398, per Lord Keith; 403, per Lord Ackner; 422, per Lord Jauncey. Ibid, 398, per Lord Keith. Ibid, 403, per Lord Ackner; 422, per Lord Jauncey. McLoughlin v OBrian [1983] 1 AC 410, 422, per Lord Wilberforce. Lord Oliver thought that the claim also failed on the degree of perception. Although present at the ground, the perception was a gradual as opposed to a sudden process: [1992] 1 AC 310, 417. See further paras 2.61 to 2.65 below. Unreported, 11th December 1996. In Scotland, where the rules relating to liability for negligently inflicted psychiatric illness are similar to those in England, the courts have adopted a strict interpretation of the requirement for a close tie of love and affection. In Robertson v Forth Road Bridge Joint Board 1996 SLT 263, Lord President Hope, supporting the trial judge, held that the plaintiffs, one of whom had spent the greater part of his working life with the deceased and had socialised with him on a weekly basis, failed to show that they had the necessary close tie of love and affection required by Alcock. See M J M Bogie, A Shocking Future?: Liability for Negligently Inflicted Psychiatric Illness in Scotland [1997] Jur Rev 39, 46. [1992] 1 AC 310, 415-416, per Lord Oliver; 422, per Lord Jauncey.
68
69
70
71
72
73
74
75
76
19
2.29
An attempt by the plaintiffs in Alcock v Chief Constable of South Yorkshire Police80 to extend the concept beyond the immediate aftermath failed. Several plaintiffs who had not been present at the ground when the tragedy occurred went there subsequently in order to identify the bodies of relatives. The earliest such plaintiff arrived at the scene between eight and nine hours after the accident, as opposed to the hour or so after the accident that Mrs McLoughlin had arrived at the hospital. Lord Ackner thought that, while the identification process might correctly be described as part of the aftermath, it was not part of the immediate aftermath.81 Lord Jauncey agreed, but also went further. He thought that the purpose for which the plaintiff comes upon the immediate aftermath was also relevant in testing proximity. The plaintiffs in Alcock went to the scene for the purpose of identifying the bodies. This, he said, was a very different situation from that in which the plaintiff goes within a short time of the accident to provide comfort and care.82 (iii) the means of perception
2.30
In McLoughlin, Lord Wilberforce noted that there was no case in which the law had compensated shock brought about by communication by a third party, and
77
McLoughlin v OBrian [1983] 1 AC 410, 422, per Lord Wilberforce. [1983] 1 AC 410. Ibid, 422. See also the Australian case, Jaensh v Coffey (1984) 155 CLR 549, where the plaintiff saw her injured husband at the hospital to which he had been taken in severe pain before and between his undergoing a series of emergency operations. The aftermath was held to include the hospital to which the injured victim was taken and persisted for so long as he remained in the state produced by the accident up to and including immediate post accident treatment. [1992] 1 AC 310. Ibid, 405. Ibid, 424. This reasoning has been criticised. In Compensation for Psychiatric Injury: The Limits of Liability (1995) 2 Psychiatry, Psychology and Law 37, 43-44 P R Handford points out that there is all the difference in the world between a relative who arrives to identify a body knowing that the person concerned is already dead, and one who is viewing rows of bodies, as in the Hillsborough case, hoping against hope that the person they seek will not be one of them.
78
79
80
81
82
20
said that this was surely right. The shock, he said, must come through sight or hearing of the event or its immediate aftermath. He left open the question whether some equivalent of sight or hearing, such as simultaneous television, would suffice.83 2.31 Perception by television was considered in Alcock. Several plaintiffs had watched the disaster unfold on live broadcasts. This was held not to create the necessary degree of proximity for two reasons. First, none of the scenes depicted the suffering of recognisable individuals, this being excluded, as the defendant was aware, by the broadcasting code of ethics. The cameras from their different vantage points showed different scenes, all of which one person could not have seen, with edited pictures and a superimposed commentary. This was not the equivalent of actual sight and hearing.84 Secondly, the pictures did not provide the degree of immediacy required to found a claim for psychiatric illness.85 The trauma arose not from seeing the original viewing of the pictures, but in part from the confirmation that the death of a loved one had occurred and in part from the linkage of this confirmation to the images seen earlier.86 However, simultaneous broadcasts were not ruled out as being equivalent to actual sight and hearing of the event or its immediate aftermath in every case. Lord Ackner cited the example given by Nolan LJ in the Court of Appeal of the televising of a special event in which children were travelling in a hot air balloon when it suddenly burst into flames. The impact of such simultaneous television pictures might, he thought, be as great, if not greater than, actual sight of the accident.87 Following Alcock, a plaintiff who suffers psychiatric illness after the communication of distressing news by a third party will not be able to recover.88 The House of
2.32
2.33
83
McLoughlin v OBrian [1983] 1 AC 410, 422-423. [1992] 1 AC 310, 398, per Lord Keith; 405, per Lord Ackner; 417, per Lord Oliver; 423, per Lord Jauncey. The plaintiffs accepted that had the broadcasting code of ethics not been followed, this would have constituted a novus actus interveniens breaking the chain of causation between the defendants negligence and their psychiatric illness: [1992] 1 AC 310, 405. See paras 2.61 to 2.62 below. [1992] 1 AC 310, 417, per Lord Oliver. Ibid, 405. And see [1992] AC 310, 417, per Lord Oliver. Different issues are raised by the question whether a person may be liable for negligently communicating distressing news, either true news in an insensitive way or news that turns out to be false. There is no English authority directly on point, although in AB & Others v Tameside & Glossop Health Authority and Trafford Health Authority [1997] 8 Med LR 91 (CA) the defendants admitted a duty of care not to break news in an insensitive manner. The case concerned claims by women who had suffered psychiatric illness after being told by letter from their health authorities that they had received medical treatment from a health worker who was HIV positive. Brooke LJ made it clear that, the defendants having admitted the existence of a duty of care, this case breaks no new ground, so far as the law is concerned: [1997] 8 Med LR 91, 93. In an earlier case, Allin v City & Hackney Health Authority [1996] 7 Med LR 167 (CC), the plaintiff successfully claimed damages for PTSD suffered after being misinformed that her baby had died and then learning, six hours later, that in fact it had survived. Again, the defendant did not dispute that a duty of care was owed. See S Dziobon and A Tettenborn, When the truth hurts: the incompetent
84
85
86
87
88
21
Lords cast doubt89 on the High Court decisions in Hevican v Ruane90 and Ravenscroft v Rederiaktieblaget Transatlantic.91 In both cases the plaintiffs had suffered psychiatric illness after being told about the death of their child in an accident caused by the defendants negligence. The decision in Ravenscroft v Rederiaktieblaget Transatlantic has since been reversed by the Court of Appeal92 as inconsistent with Alcock. (d) The plaintiff is a rescuer 2.34 Rescuers form a special category of plaintiff entitled to recovery for psychiatric illness.93 In Chadwick v British Railways Board94 Mr Chadwicks estate recovered damages for psychiatric illness suffered by Mr Chadwick as a result of the horror of assisting at the scene of a railway disaster which occurred near his home in which 90 people were killed and many more were injured. The defendant owed Mr Chadwick a duty of care since it was reasonably foreseeable that somebody might try to rescue the passengers and suffer injury in the process.95 A plaintiff may recover as a rescuer even if the rescue attempts were made in the course of his or her professional duties. In Frost v Chief Constable of South Yorkshire Police96 several police officers who had provided first aid at the scene of the Hillsborough disaster and had attempted to resuscitate victims were able to recover damages for post-traumatic stress disorder suffered as a consequence of their involvement. The suggestion that the officers should not be able to recover because it was part of their professional duties to deal with such situations was rejected.97
2.35
transmission of distressing news (1997) 13 PN 70 and M A Jones, Negligently inflicted psychiatric harm: is the word mightier than the deed? (1997) 13 PN 111.
89
[1992] 1 AC 310, 398, per Lord Keith; 401, per Lord Ackner; and 418, per Lord Oliver. [1991] 3 All ER 65. [1991] 3 All ER 73. [1992] 2 All ER 470 (Note). Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194, 1202, per Rose LJ; Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 408, per Lord Oliver and 420-421, per Lord Jauncey; and McLoughlin v OBrian [1983] 1 AC 410, 419, per Lord Wilberforce. [1967] 1 WLR 912. [1967] 1 WLR 912, 921. In giving his judgment, Waller J quoted Cardozo J in Wagner v International Railway Company 232 NY Rep 176, 180 (1921): Danger invites rescue. The cry of distress is the summons to relief. The law does not ignore these reactions of the mind in tracing conduct to its consequences. It recognises them as normal. It places their effect within the range of the natural and probable. The wrong that imperils life is a wrong to the imperilled victim; it is a wrong also to his rescuer. [1997] 3 WLR 1194. See also paras 2.37 to 2.38 below. Rose LJ accepted the plaintiffs argument that, because a professional rescuer is able to recover damages for any physical injury that is suffered during the course of a rescue attempt (Ogwo v Taylor [1988] AC 431), by analogy the professional rescuer ought also to be able to recover damages for any psychiatric illness: [1997] 3 WLR 1194, 1203. In an earlier decision, Hale v London Underground Ltd [1993] PIQR Q30, liability had been
90
91
92
93
94
95
96
97
22
2.36
Not every involvement at the scene of a disaster will qualify as a rescue attempt. 98 For example, in McFarlane v EE Caledonia Ltd the plaintiff was on board a support vessel going to the aid of the victims of the Piper Alpha disaster. However, he was classified as non-essential personnel and his involvement in the rescue effort was limited. He moved blankets with a view to preparing a helihanger to receive casualties and encountered and perhaps assisted two walking injured as they arrived on the rescue vessel. This was not sufficient to bring him within the rescuer category. In Duncan v British Coal Corporation99 a pit deputy attempted to resuscitate one of the men for whom he was responsible who had been crushed by machinery. He failed to recover damages for his consequential psychiatric illness as a rescuer. While his actions were proximate in time to the deceaseds injury, he was not geographically proximate when the incident occurred. When he arrived at the scene there was no danger to him or to the deceased. The first aid which he administered was within the normal scope of his employment duties and there were no unusually distressing features. Likewise in Frost100 three of the police officers at the Hillsborough stadium failed to qualify as rescuers (although the majority of the Court of Appeal held that two of them were entitled to recover as employees).101 It is worth considering the different activities that each carried out. The position of six police officers was considered by the Court of Appeal.102 Janet Smith was not at the ground when the incident occurred, but reported about one hour later to the local hospital, where she was asked to strip bodies and complete casualty forms in the mortuary. Later she took personal effects to the makeshift mortuary set up in the gymnasium at the ground. DC Hallam was in the gymnasium when the disaster occurred. He attempted to revive an apparently dead boy. He saw the mayhem created as more casualties, relatives and police arrived in the gymnasium. PC Glave was also in the gymnasium but was ordered to go to the end of the ground far from the pens where the tragedy occurred. He helped to move bodies and obtained first aid. None of these officers was held to qualify as a rescuer on the grounds that they were not sufficiently closely involved in the incident or its immediate aftermath.103 In contrast, the majority of the Court of Appeal (Judge LJ dissenting) held that the remaining three officers all qualified as rescuers. Anthony Bevis was summoned to the gymnasium from his patrol elsewhere. He approached to within 20 yards of the pens fencing and came across corpses. He tried to resuscitate victims and later formed part of a line of officers controlling the crowd. Mark Bairstow arrived
2.37
admitted in the case of a professional fireman who sustained post-traumatic stress disorder in the Kings Cross fire.
98
[1994] 2 All ER 1. [1997] 1 All ER 540. The Court of Appeals judgment in this case was given at the same time as judgment in Frost v Chief Constable of South Yorkshire Police. [1997] 3 WLR 1194. See para 2.41 below. Judge LJ dissented. One of the six officers in fact withdrew his appeal from the High Court decision, but the Court of Appeal still considered his position as there were several pending cases in which the officers had performed similar duties to him. [1997] 3 WLR 1194, 1204-1205, per Rose LJ; 1220, per Henry LJ. See also 1228-1229, per Judge LJ.
99
100
101
102
103
23
at the ground some time after the incident and went straight to the pens end. He saw bodies lying on the ground and checked to see if they were dead. He assisted in giving heart massage to a boy, and helped to carry out the bodies. Inspector White was initially outside the ground. He saw congestion at the gate and pulled people out of the pens fearing for their safety. He then entered the ground and could see dead bodies. He joined a line of officers passing the dead and injured away from the pens. Later he helped in the identification of bodies. 2.38 In reaching his decision in relation to each officer, Rose LJ (with whom Henry LJ agreed) said that factors to be considered in deciding whether a particular plaintiff is a rescuer include: the character and extent of the initial incident caused by the tortfeasor; whether that incident has finished or is continuing; whether there is any danger, continuing or otherwise, to the victim or to the plaintiff; the character of the plaintiff s conduct, in itself and in relation to the victim; and how proximate, in time and place, the plaintiff s conduct is to the incident.104 (e) The plaintiff is an involuntary participant 2.39 The category of involuntary participant was first recognised by Lord Oliver in Alcock. It includes those cases where the negligent act of the defendant has put the plaintiff in the position of being, or of thinking that he is about to be or has been, the involuntary cause of anothers death or injury and the illness complained of stems from the shock to the plaintiff of the consciousness of this supposed fact.105 He included in this category the case of Dooley v Cammell Laird106 where a crane-driver suffered psychiatric illness after seeing a defective rope on his crane snap thereby causing the crane to drop its load onto the hold of a ship where he knew his fellow employees were working. Although no-one in fact was injured, he feared for their safety. Also included were Galt v British Railways Board,107 where a train driver came upon two workmen as he rounded a bend, and being unable to stop, feared that he had killed them; and Wigg v British Railways Board,108 where a train driver came upon the body of a dead person who had been struck by the door of the train he was driving very soon after it pulled out of the station. However, in Frost v Chief Constable of South Yorkshire Police Henry LJ suggested a different categorisation for these cases. He described them as being master and servant cases where the employers negligence in imperilling a fellow workmate makes the plaintiff an unwilling participant in the event.109 If this were the
2.40
104
[1997] 3 WLR 1194, 1203. [1992] 1 AC 310, 408. [1951] 1 Lloyds Rep 271. (1983) 133 NLJ 870. The Times 4 February 1986. [1997] 3 WLR 1194, 1212. In reaching their decision in Frost, Henry and Rose LJJ distinguished the decision of the Inner House of the Court of Session in Scotland in Robertson & Rough v Forth Road Bridge Joint Board 1996 SLT 263. Three employees were asked by their employer to remove a sheet of metal from the Forth Bridge in windy conditions. A gust of wind caught the sheet of metal and one employee was blown over the side of the bridge and killed. The accident was witnessed by the second employee and heard by the third, both of whom subsequently suffered psychiatric illness. The Court ruled that the plaintiffs could not recover since there was no suggestion that they believed that
105
106
107
108
109
24
110 rationale for the decisions, then the involuntary participant would not have a claim against any tortfeasor, but only against a tortfeasor who is his or her employer. This would effectively rule out recovery by, for example, a car driver who ran over a pedestrian having been unable to stop in time due to the negligent manufacture of his of her car brakes. The category of involuntary participant would be subsumed by category (f) below.
(f) The plaintiff is an employee who is directly involved in an incident caused by his or her employers negligence which results in the injury or imperilment of another 2.41 An employee may be able to recover damages from his or her employer in respect of reasonably foreseeable psychiatric illness suffered in the course of employment as a result of the employers negligence. Just as an employer has a duty of care to his or her employees not to expose them to the risk of physical injury, so also an employer has a duty of care not to expose them to the risk of psychiatric illness.111 In Frost, two of the police officers who had failed to qualify as rescuers, DC Hallam and PC Glave, recovered damages from the Chief Constable (who for these purposes was treated as their employer) on the basis that the carrying out of their duties had brought them within the range of foreseeable psychiatric injury created by their employers negligence. An employer is not, however, in every circumstance liable for psychiatric illness suffered by an employee who carries out employment duties following his or her employers negligence. Where the illness is suffered pursuant to another persons physical injury or imperilment, employees must show that they were directly involved, in the course of their employment, in the consequences flowing from their employers negligence112 and that the carrying out of employment duties brought them within the area of risk of physical or psychiatric injury.113 In Frost it was held that Janet Smith114 was not within the category of those officers to whom a duty of care was owed by virtue of their employment relationship, since she was not at the ground when the negligent loss of control occurred. Everything
2.42
they had been the cause of the accident. Without this, they were mere bystanders. In Frost, Rose LJ doubted that the English courts would reach the same decision, while Henry LJ thought that even though the Robertson decision might be correct on the facts, it had not shut the door to claims by employee plaintiffs who had participated in an accident caused by the negligence of their employer: [1997] 3 WLR 1194, 1203-1204 and 1214-1215 respectively.
110
This cannot have been the reasoning behind the decision in Dooley v Cammell Laird [1951] 1 Lloyds Rep 271 since the plaintiff was awarded judgment not only against his employer (for breach of statutory duty as the occupier of the shipbuilding yard) but also against the owner of the defective rope (for negligence). Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194, 1204, per Rose LJ. Ibid, 1203, per Rose LJ. Ibid, 1205, per Rose LJ. For the facts of the case see para 2.37 above.
111
112
113
114
25
that she subsequently did was no more than could properly be asked of a police officer in the ordinary carrying out of police duties after a serious incident.115 2.43 On the basis that in Frost, DC Hallam and PC Glave were able to recover as employees but not as rescuers, it might at first appear that the required involvement of an employee is not as great as that of a rescuer. However, this seems not to be the case since the court also held that Mark Bairstow, who qualified as a rescuer, would not have qualified as an employee because he was not within the area of risk of physical and psychiatric injury when the incident occurred.116 As can be seen, the extent of the employee category following Frost is by no means clear. There is no obvious distinction between the activities carried out by the various officers to explain why some fell within the area of risk of foreseeable physical or psychiatric illness and others did not, although their success or failure seems to be based on whether they were present at the ground when the incident first occurred, rather than summoned to it later. Furthermore, as discussed below,117 an employers general duty to provide a reasonably safe system of work relates not only to any physical injury that an employee may suffer, but also to any psychiatric injuries. In a case where an employee suffered a nervous breakdown caused by stress at work, the test that has been applied is simply whether the breakdown was reasonably foreseeable.118 Why, in cases of psychiatric illness consequent upon injuries to a third person, the employee should have to prove not only that his or her illness was reasonably foreseeable but also that he or she was in some sense directly involved is not clear. (g) Can the plaintiff who is a mere bystander ever recover? 2.45 Whether a mere bystander (that is a person who witnesses the death, injury or imperilment of the immediate victim but has no close tie of love and affection with him or her) may in any circumstances recover damages for psychiatric injury is not certain. In McLoughlin v OBrian, Lord Wilberforce thought that such claims would be denied, either on the basis that bystanders must be assumed to be possessed of sufficient fortitude to enable them to endure the calamities of modern life, or on the basis that defendants cannot be expected to compensate the world at
2.44
115
The pit deputy in Duncan v British Coal Corporation [1997] 1 All ER 540 also failed to recover under this category. (For the facts of the case see para 2.36 above). The Court of Appeal held that the first aid that the plaintiff rendered was clearly within the scope of his duties and there were no unusually distressing features. He was outside the area of risk of physical or psychiatric injury when the accident took place and had not been exposed to any unnecessary risk of injury when he attended the scene. The defendant employer was therefore not in breach of any duty owed as an employer to the pit deputy. [1997] 3 WLR 1194, 1206. The judgment of Henry LJ is, with respect, difficult to follow on this point. He says, at 1220A, that he agrees with Rose LJs conclusions in relation to the individual officers, but later says that where the plaintiff is a direct victim because of the duty that his employer owes to him, that should be the first head of recovery to be considered because it might be wider and would not, as far as he could foresee, be narrower than any entitlement as a rescuer: [1997] 3 WLR 1194, 1220C. See para 2.47. Walker v Northumberland County Council [1995] 1 All ER 737. See further para 2.47 below.
116
117
118
26
119 The question was left open in Alcock v Chief Constable of South Yorkshire large. Police. Lord Keith thought that whilst it was not ordinarily foreseeable that a bystander would suffer psychiatric illness as a result of witnessing an accident, if the circumstances of a catastrophe occurring very close by were particularly horrific, then this might be the case. Lord Ackner provided as an example of such a horrific event the sight of a petrol tanker careering out of control into a school in session and bursting into flames. Lord Oliver could not exclude the possibility of successful claims by bystanders who witnessed circumstances of such horror that would be likely to traumatise even the most phlegmatic spectator.120
2.46
However, the Court of Appeal in McFarlane v EE Caledonia Ltd121 thought that as a matter of both principle and policy the court should not extend the duty of care to those who are mere bystanders or witnesses of horrific events. To extend liability to those who do not have a close tie of love and affection to an immediate victim would be to base the test of liability on foreseeability alone, something which was ruled out by the decision in Alcock. Moreover, there would be great practical problems in deciding which accidents were sufficiently horrific, since reactions to horrific events are entirely subjective.122 (2) Cases where the Defendant has neither Imperilled nor caused Physical Injury to any Person (a) The plaintiff who suffers reasonably foreseeable psychiatric illness induced through stress at work
2.47
In Walker v Northumberland County Council,123 Colman J held that a social worker was entitled to damages for a second nervous breakdown caused by stress at work. The case was decided on well-established general principles of negligence124 and the judgment does not refer to Alcock or any other nervous shock cases. Colman J said that it was clear law that an employer has a duty to provide his employee
119
[1983] 1 AC 410, 422. [1992] 1 AC 310, 397, per Lord Keith; 403, per Lord Ackner; 416, per Lord Oliver. In Post Traumatic Stress Disorder: turning the tide without opening the floodgates (1995) 35 Med Sci Law 287, M McCulloch et al say that although English courts have consistently refused to allow claims by mere bystanders, from a psychological point of view it is entirely foreseeable that such people may experience nervous shock. [1994] 2 All ER 1. Ibid, 14. [1995] 1 All ER 737. J Messham, A flood of claims? (1995) 139 SJ 732. In an earlier case, Francis Aston v Imperial Chemical Industries Group (unreported, 21st May 1992), Rose J had awarded damages to a plaintiff for the depressive illness which he suffered as a result of his anxiety for his health following exposure to carcinogenic fumes in his workplace. The fumes could cause angiosarcoma of the liver, which the plaintiff was told has a latency period of about 15 years, but is usually fatal within six months of the symptoms appearing. In rejecting the employers argument that he owed no duty of care in respect of psychiatric illness which was not consequential on a physical injury, Rose J said: It furthermore seems to me that the employer whose system of work negligently induces psychiatric injury without any physical injury by, for example, excessive noise or flickering lights or psychological pressures is just as liable as one who causes physical injury because the duty of care exists and the necessary proximity exists by reason of the master and servant relationship.
120
121
122
123
124
27
with a reasonably safe system of work and to take reasonable steps to protect him from risks which are reasonably foreseeable. While the law had developed almost exclusively in cases involving physical injury, there was no logical reason why risk of psychiatric damage should be excluded from the scope of an employers duty of care or from the co-extensive implied term in the contract of employment. Colman J recognised that this would give rise to difficult questions in relation to foreseeability and causation, particularly for professional employees who may be ambitious and dedicated, determined to succeed in a career that is known to be demanding and who may have a degree of discretion as to how and when and for how long they work. On the facts of this case, the defendant was found not liable in respect of the plaintiff s first breakdown which was not reasonably foreseeable. However, having been made aware of his susceptibility and the stress which he felt under, the Council should have reduced the plaintiff s work-load on his return, as had been agreed. For failing to do this, the Council was in breach of its duty of care in respect of the second breakdown. (b) The plaintiff who suffers reasonably foreseeable psychiatric illness as a result of the defendant causing damage to property 2.48
125 On a preliminary The most important decision here is Attia v British Gas Plc. issue the Court of Appeal held that a plaintiff could be entitled to damages for a psychiatric illness which she suffered as a result of witnessing a fire caused by the negligence of the workmen whom she had employed to install central heating in her home. The case was allowed to proceed to trial on the facts. The court regarded the issue as raising questions of remoteness only. Dillon LJ said that the defendants unquestionably owed a duty of care to the plaintiff not to set fire to her home, and indeed her claim for property damage had already been settled. The court held that psychiatric illness caused by property damage could be reasonably foreseeable and was not prepared to accept that as a matter of policy all such claims should be ruled out. Bingham LJ suggested examples to show why he thought recovery could not be denied: Suppose, for example, that a scholars lifes work of research or composition were destroyed before his eyes as a result of a defendants careless conduct, causing the scholar to suffer reasonably foreseeable psychiatric damage. Or suppose that a householder returned home to find that his most cherished possessions had been destroyed through the carelessness of an intruder in starting a fire or leaving a tap running, causing reasonably foreseeable psychiatric damage to the owner.126
2.49
However, the criteria for liability are not clear. The facts of Attia may be regarded as particularly strong in that the property which was damaged belonged to the plaintiff, a home and furniture might be regarded as having particularly sentimental value, and the plaintiff witnessed the destruction herself.127
125
[1988] QB 304. Ibid, 320. This area of the law is further complicated by the need to take account of the decision in Perry v Sidney Phillips & Son [1982] 1 WLR 1297. In an action for negligence, the plaintiff was awarded damages for the worry and distress he suffered as a result of the physical inconvenience of living in a house with severe defects which the defendant surveyors had failed to report. In the Australian case, Campbelltown City Council v Mackay (1989) 15
126
127
28
2.50
128 Owens v Liverpool Corporation may also be regarded as a case involving property damage. The defendant had negligently driven into a hearse, causing damage to the hearse and overturning the coffin so that it appeared that the coffin might fall out. Relatives of the deceased man saw the damage from their position in the funeral procession and suffered nervous shock as a result. The Court of Appeal held that the right to recover damages for psychiatric illness was not restricted to cases in which apprehension as to human safety was involved, although the fact [t]hat alleged shock results from apprehension as to a less important matter may well be material in considering whether the allegation be proved.129 It is doubtful how far the case, which was disapproved by three members of the House of Lords in Bourhill v Young,130 may still be relied upon.131
(c) Miscellaneous 2.51 There are other miscellaneous situations in which recovery may be available for a negligently inflicted psychiatric illness (assuming that the standard elements of the tort of negligence can be made out). These include: where a patient suffers a psychiatric illness because of negligent treatment by his or her psychiatrist;132 where a prisoner foreseeably suffers a psychiatric illness as a result of ill-treatment by prison officers;133 and where recipients of distressing news suffer reasonably foreseeable psychiatric illness as a result of the news being broken in an insensitive manner.134
NSWLR 501, plaintiffs who conceded that they could not claim damages for psychiatric illness pursuant to the destruction of their dream home because their illness was not shockinduced, nevertheless successfully recovered the same level of damages under their alternative claim based on Perry v Sidney Phillips & Son.
128
[1939] 1 KB 394. Ibid, 400. [1943] AC 92, 100, per Lord Thankerton; 110, per Lord Wright; 116, per Lord Porter. Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 412, per Lord Oliver. Cf M v Newham LBC [1995] 2 AC 633 (although in that case the psychiatrist had been retained by the local authority, under its statutory duties to protect children, to discover whether the child had been sexually abused). Cf R v Deputy Governor of Parkhurst Prison, ex p Hague [1992] 1 AC 58, 165-166, per Lord Bridge; Racz v Home Office [1994] 2 AC 45 (but note that in that case mental distress alone had been suffered and the central cause of action was misfeasance in public office, not negligence). AB & Others v Tameside & Glossop Health Authority and Trafford Health Authority [1997] 8 Med LR 91 and Allin v City & Hackney Health Authority [1996] 7 Med LR 167. See para 2.33 n 88 above. N J Mullany notes the increasing number of claims being pursued through the US and Australian courts for psychiatric illness induced through fear of what might happen in the future. This might be the case where, say, the plaintiff has been exposed to toxic chemicals and fears that he or she may contract cancer at some future date: N J Mullany, Compensation for Fear and Worry-Induced Psychiatric Illness: The Australian Position (1997) 4 Psychiatry, Psychology and Law 147 and Fear for the Future: Liability for Infliction of Psychiatric Disorder in N J Mullany (ed), Torts in the Nineties (1997) ch 5.
129
130
131
132
133
134
29
3. THE CLASSIFICATION OF PRIMARY AND SECONDARY VICTIMS 2.52 As we explained above,135 the decision in Page v Smith drew a distinction between the primary victim and the secondary victim of an accident. Lord Lloyd said that it is essential in all claims relating to psychiatric illness to make this classification,136 and recent decisions have attempted to do so.137 However, the cases display a great deal of confusion as to which categories of plaintiff should be regarded as primary and which as secondary victims. Furthermore, the reason for and effect of making the distinction is not always clear.138 We consider both these issues below. It was Lord Oliver, in his judgment in Alcock v Chief Constable of South Yorkshire Police, who first drew attention to the distinction that may be drawn between different categories of potential claimants for psychiatric illness. He said that although it was customary to group together all cases in which damages were claimed for psychiatric illness, in fact they could be divided into two categories: those cases in which the injured plaintiff was involved, either mediately or immediately, as a participant, and those in which the plaintiff was no more than the passive and unwilling witness of injury caused to others.139 Lord Oliver gave three examples of plaintiffs whom he would classify as within the former category, that is as being involved as participants. These were: (i) plaintiffs who feared for their own safety; (ii) rescuers; and (iii) involuntary participants. Lord Lloyd, in his judgment in Page v Smith, developed this categorisation further. The factual distinction between a primary and a secondary victim of an accident, he said, is obvious and has important legal consequences.140 However, it is not clear that he was adopting the same division as Lord Oliver took in Alcock. Lord Lloyd referred to primary victims only in terms of those who are directly involved in the accident and well within the range of foreseeable physical injury.141 Lord Lloyds category of primary victim may therefore be narrower than Lord Olivers, excluding, for example, rescuers and involuntary participants unless they come within the range of foreseeable physical injury. 142
2.53
2.54
135
See para 2.10. [1996] AC 155, 197. See, for example, Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194, 1203, per Rose LJ; 1209-1215, per Henry LJ; 1224-1229, per Judge LJ; Hegarty v EE Caledonia Ltd [1997] 2 Lloyds Rep 259, 265-266; Young v Charles Church (Southern) Ltd, The Times 1 May 1997; Transcript No QBENF 96/0920/C at pp 9-13, per Evans LJ; at pp 21-22, per Hobhouse LJ; at p 34, per Hutchison LJ. See M A Jones, Liability for Psychiatric Illness - More Principle, Less Subtlety? [1995] 4 Web JCLI where it is argued that the distinction between a primary and secondary victim does not stand up to analysis and that, although it may be a rationalisation of the present state of the law, it does not provide any justification for the different liability rules that are applied. See also para 5.51 n 96 below. [1992] 1 AC 310, 407. [1996] AC 155, 184. [1996] AC 155, 184, per Lord Lloyd; and see 182, per Lord Browne-Wilkinson. On the difficulty of reconciling Lord Lloyds and Lord Olivers categorisations see N J Mullany and P R Handford, Hillsborough Replayed (1997) 113 LQR 410, 416 and Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194, 1213, per Henry LJ.
136
137
138
139
140
141
142
30
2.55
One of the reasons for this discrepancy is probably the different purposes for which Lord Oliver and Lord Lloyd were drawing the distinction. Lord Oliver clearly thought that it was relevant for issues of proximity. The primary victim cases, according to Lord Oliver, illustrate a directness of relationship (and thus a duty) which is almost self-evident from a mere recital of the facts.143 However, a secondary victim is not, without more, in a relationship of sufficient proximity with the defendant as to give rise to a duty of care. Secondary victims must establish a duty of care based on proximity in terms of their relationship with the immediate victim, their closeness to and their perception of the accident.144 In Page v Smith, Lord Lloyd used the primary/secondary victim classification for different purposes. He thought that the law should not commit itself to a distinction between physical injury and psychiatric illness which, he said, may already be somewhat artificial, and may soon be altogether outmoded. Nothing would be gained by treating them as different kinds of personal injury, so as to require the application of different tests in law.145 Whether the plaintiff suffered a physical injury or psychiatric illness the test of liability is therefore the same: whether the defendant could reasonably foresee that his conduct would expose the plaintiff to risk of personal injury. In the case of a primary victim the question will almost always turn on whether the foreseeable injury is physical, whereas in the case of a secondary victim the question will usually turn on whether the foreseeable injury is psychiatric. But it is the same test in both cases, with different applications.146 However, in the case of secondary victims the law requires not only foreseeability but also imposes certain control mechanisms in order as a matter of policy to limit the number of potential claimants. Lord Lloyd gave two examples of the control factors to which he was referring. First, that the psychiatric illness should have been reasonably foreseeable in a person of normal fortitude and, secondly, that hindsight may be used in applying the foreseeability test.147 Several cases highlight the difficulties which the courts have had in putting the primary/secondary victim distinction into practice. In Frost v Chief Constable of South Yorkshire Police,148 the majority classified five of the police officers as primary victims following Lord Olivers broad categorisation in Alcock, although Henry LJ recognised that this was difficult to reconcile with Lord Lloyds narrower
2.56
2.57
143
Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 407. Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 411. Page v Smith [1996] AC 155, 188. Lord Browne-Wilkinson endorsed Lord Lloyds remarks about the dangers of the court seeking to draw hard and fast lines between physical illness and its causes on the one hand and psychiatric illness and its causes on the other: Although medical science has not as yet progressed very far in elucidating the processes whereby psychiatric disorders come about, recent developments suggest a much closer relationship between physical and mental processes than had previously been thought: [1996] AC 155, 182. [1996] AC 155, 190, per Lord Lloyd. [1996] AC 155, 197. [1997] 3 WLR 1194. See paras 2.37 to 2.38 and 2.40 to 2.44 above.
144
145
146
147
148
31
classification in Page.149 These officers were held to be primary victims because they were directly involved150 and were participants in the accident.151 However, Henry LJ thought that the correct labelling of the plaintiffs did not really matter since the distinction went only to the question of whether proximity could be presumed, and in the case of the officers, their proximity as employees acting in the course of their employment needed to be critically examined in any event.152 2.58 In Hegarty v EE Caledonia Ltd,153 the Court of Appeal held that a primary victim was either (i) as laid down in Page v Smith, a person directly involved in the accident in question and well within the range of foreseeable physical injury; or (ii) one who was involved as a participant and who feared for his or her own safety. This second category, although said to be based on Lord Olivers classification in Alcock v Chief Constable of South Yorkshire Police, has a gloss added on to it: it requires that the plaintiff should have feared for his or her safety. This would exclude the officers in Frost v Chief Constable of South Yorkshire Police154 and the involuntary participants in Dooley v Cammell Laird,155 Galt v British Railways Board156 and Wigg v British Railways Board.157 What, if anything, turned on whether the plaintiff in Hegarty was a primary victim is not clear from the judgment. It was clear that even if he fell within the second category of primary victim, the plaintiff had to show that a person of ordinary fortitude in his position would reasonably be in such fear of personal injury so as to suffer a psychiatric illness,158 a test which Lord Lloyd in Page v Smith had said only applied to secondary victims. Further difficulties arose in Young v Charles Church (Southern) Ltd.159 The plaintiff, though himself within the area of foreseeable physical injury created by the defendants negligence, suffered psychiatric illness as a result of witnessing the traumatic death of his work colleague. According to Evans and Hutchison LJJ, this
2.59
149
[1997] 3 WLR 1194, 1213. Judge LJ dissented. He thought that the officers were secondary victims. He said that he did not regard Lord Olivers broad categorisation of primary victims in Alcock as supporting the conclusion that everyone falling within the rescue principle must be regarded as a primary victim. If it did, then it would require reassessment in the light of Page v Smith. Likewise, he said that an employee who establishes that his employers breach of duty has caused him psychiatric illness is not automatically to be categorised as a primary victim of his employers negligence. In both cases, depending on their involvement in the incident, some will be primary victims and others will be secondary victims: [1997] 3 WLR 1194, 1224 and 1227. [1997] 3 WLR 1194, 1203, per Rose LJ. Ibid, 1213, per Henry LJ. Ibid, 1213-1214. [1997] 2 Lloyds Rep 259. [1997] 3 WLR 1194. [1951] 1 Lloyds Rep 271. See para 2.39 above. (1983) 133 NLJ 870. See para 2.39 above. The Times 4 February 1986. See para 2.39 above. [1997] 2 Lloyds Rep 259, 266. The Times 1 May 1997.
150
151
152
153
154
155
156
157
158
159
32
was irrelevant, and following Page v Smith, the plaintiff was a primary victim. Hobhouse LJ disagreed: the plaintiff was a secondary victim since he was an observer of a traumatic and fatal injury inflicted upon another.161 2.60 There is therefore a confusing inconsistency of approach by the courts to the classification of primary and secondary victims. Nor is it clear how the distinction will be drawn in non-accident cases:162 for example, Lord Lloyds approach assumes that the plaintiff is in danger.163 We return to this issue in Part V.164 There, we consider how one might cut through the confusion, whether one approach to the primary/secondary distinction is better than another, and whether statutory reform on this distinction is merited. 4. TWO ADDITIONAL RESTRICTIONS ON RECOVERY (1) The Shock Requirement 2.61 The plaintiff may be required to show that his or her psychiatric illness was induced by a shock. According to Lord Ackner, what is required is the sudden appreciation by sight or sound of a horrifying event, which violently agitates the mind.165 Lord Keith required a sudden assault on the nervous system,166 while Lord Oliver referred to a sudden and unexpected shock to the plaintiff s nervous system.167 The principle is described most fully by Brennan J in Jaensch v Coffey,168 a decision of the High Court of Australia. He said: A plaintiff may recover only if the psychiatric illness is the result of physical injury negligently inflicted on him by the defendant or if it is induced by shock. Psychiatric illness caused in other ways attracts no damages, though it is reasonably foreseeable that psychiatric illness might be a consequence of the defendants carelessness.169 He gave two examples where no recovery would be available: The spouse who has been worn down by caring for a tortiously injured husband or wife and who suffers psychiatric illness as a result goes without compensation; a parent made distraught
160
160
The Times 1 May 1997; Transcript No QBENF 96/0920/C at pp 12-13 and 34 respectively. The Times 1 May 1997; Transcript No QBENF 96/0920/C at p 22. Hobhouse LJ agreed with the outcome of the case, since he held that the plaintiff could recover as an employee who suffered psychiatric illness as a result of his employers negligence in accordance with the principles laid down in Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194. That is, those referred to in paras 2.47 to 2.51 above. In AB & Others v Tameside & Glossop Health Authority and Trafford Health Authority [1997] 8 Med LR 91, 99 Brooke LJ noted that on a future occasion the question whether the plaintiff in a case involving the negligent communication of accurate but distressing information should correctly be regarded as a primary victim might have to be carefully considered by the courts. See paras 5.45 to 5.54 below. Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 401. Ibid, 398. Ibid, 411. (1984) 155 CLR 549. Ibid, 565.
161
162
163
164
165
166
167
168
169
33
by the wayward conduct of a brain-damaged child and who suffers psychiatric illness as a result has no claim against the tortfeasor liable to the child.170 2.62 The requirement for a shock was not explicitly referred to in the English case law until the decision in Alcock v Chief Constable of South Yorkshire Police. In that case, neither Lord Keith nor Lord Oliver thought that the viewing of live television broadcasts of the Hillsborough disaster could be regarded as giving rise to a shock. Lord Keith said that while the pictures were capable of giving rise to anxiety for the safety of relatives known or believed to be in the area affected, this was very different from seeing the fate of a relative shortly after the event.171 Lord Oliver considered that the images would have given rise to grave concern and worry, followed by a dawning consciousness over an extended period that the feared consequence had occurred, with final confirmation being provided by news of the death or the identification of the body.172 In Alcock the court was only concerned with plaintiffs who had suffered psychiatric illness as a result of the death, injury or imperilment of another. However, the shock test has subsequently been adopted in relation to other categories of plaintiff. In Frost v Chief Constable of South Yorkshire Police, Henry LJ seemed to consider that the test was relevant to the categories of rescuers and employees, whether they be correctly classified as primary or secondary victims.173 In Hegarty v EE Caledonia Ltd, Brooke LJ referred to the requirement for shock-induced psychiatric illness in the case of a plaintiff who suffered as a result of fear for his or her own safety.174 By analogy, it seems likely that the shock test applies in cases of psychiatric illness caused by property damage. The point has not been raised in the sparse relevant English case law175 but would appear to be a requirement under Australian case law.176 It has yet to be decided whether a plaintiff who falls within the Page v Smith test of being within the area of reasonably foreseeable physical injury must also satisfy a shock requirement.177 A shock is clearly not required in cases of psychiatric illness induced through stress at work.178
170
Ibid, 565. These examples were cited with approval by Lord Ackner in Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 400. [1992] 1 AC 310, 398. [1992] 1 AC 310, 417. Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194, 1208. [1997] 2 Lloyds Rep 259, 266. See paras 2.48 to 2.50 above. See Campbelltown City Council v Mackay (1989) 15 NSWLR 501, 503. The parties agreed that the plaintiffs could only recover damages for the psychiatric illness suffered as a result of the damage to their dream home if the illness resulted from a sudden affliction to their respective psyches. The majority in Page v Smith held that, where the plaintiff was within the area of risk of foreseeable physical injury, the courts ought not to distinguish as a matter of principle between physical injury and psychiatric illness. Clearly the law does not impose any sudden impact or onset test in claims relating to physical injury, so one could argue that by analogy no shock should be required in claims relating to psychiatric illness. In M v Newham London Borough Council [1995] 2 AC 633, the plaintiffs, a child and her mother, allegedly suffered anxiety neurosis as a result of having been separated for a year on the basis of a false psychiatric report that the child was being abused by the mothers boyfriend. In the Court of Appeal, Sir Thomas Bingham MR, dissenting, thought that the childs claim
171
172
173
174
175
176
177
34
2.63
The shock requirement has led to some harsh and seemingly arbitrary decisions. 179 For example, in Sion v Hampstead Health Authority the plaintiff s son was injured in a motor-cycle accident and admitted to hospital. The plaintiff stayed at his sons bedside and watched him slowly deteriorate in health, suffer from a heart attack, fall into a coma and eventually die. The plaintiff alleged that the hospital was negligent in its treatment and sought damages for the abnormal grief reaction that he suffered pursuant to his sons death. The hospital was successful in striking out the claim on the basis that the plaintiff s medical report did not describe the plaintiff as suffering from a shock. Rather, the report described a process continuing for some time from the first arrival in the hospital to the appreciation of medical negligence after the inquest. When death came, it was not surprising but expected. Similarly, the plaintiffs in Taylorson v Shieldness Produce Ltd180 failed in their claim for damages for psychiatric illness suffered after the death of their son. The boy had been crushed by a reversing vehicle and taken directly to hospital. The plaintiffs were told of the accident soon after it occurred, drove to the hospital and then followed an ambulance which transferred him to a second hospital. The parents caught very brief glimpses of their child as he was taken from the ambulance into intensive care. His father went to his bedside as soon as he was allowed after treatment, but his mother was dissuaded from going until the next day. They stayed with him during the following two days until his life support machine was switched off three days after the accident. The Court of Appeal felt bound by the shock requirement to deny recovery. The plaintiffs illnesses could not be attributed to one shocking event but rather grew out of a sequence of events that extended over an appreciable period of time.181 There may have been some recent relaxation of the suddenness that is required to meet the shock test. While in Alcock v Chief Constable of South Yorkshire Police Lord Oliver suggested that the plaintiffs claim should be denied, because their
2.64
2.65
should not be struck out even though the psychiatric illness was not the result of a shock: [1995] 2 AC 633, 663-664. Staughton and Peter Gibson LJJ struck out the childs claim without expressing any view on whether psychiatric illness which is not the result of a sudden occurrence is compensatable. The plaintiff in Sion v Hampstead Health Authority [1994] 5 Med LR 170 (see para 2.63 below) argued, on the basis of the remarks made by Sir Thomas Bingham MR in Newham, that it was unnecessary to prove a shock. The Court of Appeal disagreed, and interpreted Sir Thomas Bingham MRs comments as being intended to apply only to primary victims such as the child in Newham and not to secondary victims such as the plaintiff in Sion. The House of Lords in Newham [1995] 2 AC 633 upheld the Court of Appeals decision without considering this issue. More recent cases, however, still refer to shock-induced injury in relation to plaintiffs falling within the Page test: for example, Young v Charles Church (Southern) Ltd, The Times 1 May 1997.
178
See para 2.47 above. [1994] 5 Med LR 170 (CA). [1994] PIQR P329 (CA). The decision in Tredget v Bexley Health Authority [1994] 5 Med LR 178 (CC) may be contrasted to these cases. Parents recovered damages for the psychiatric illness which they both suffered as a result of the mothers negligent treatment during childbirth which resulted in the death of their child two days later. Judge White considered that the shock requirement was satisfied. He said: Although lasting for over 48 hours from the onset of labour to the death, this effectively was one event.
179
180
181
35
discovery of the death of loved ones was gradual rather than sudden, and even though the period of time before the plaintiffs discovered for certain the fate of their loved ones was in no case more than twenty hours, in Frost v Chief Constable of South Yorkshire Police182 Henry LJ found that it was the length of the exposure and the circumstances of the trauma that caused the police officers to suffer psychiatric illness, rather than any sudden and immediate shock. He did not think that this necessarily conflicted with the dicta in Alcock, but if it did, then he was prepared to allow the expansion, which he said was required as a result of better medical understanding of what may trigger a psychiatric illness. He said: [W]hat matters is not the label on the trigger for psychiatric damage, but the fact and foreseeability of psychiatric damage, by whatever process. ... Clearly the law should accept PTSD rather than exclude it whether it is caused by sudden shock (properly defined) or not.183 (2) The Defendant as the Immediate Victim 2.66 Dicta by Deane J in the Australian case, Jaensch v Coffey,184 suggest that, where the plaintiff suffers psychiatric illness pursuant to the death, injury or imperilment of another, damages are not recoverable where that other was the defendant him or herself.185 Dicta by Lord Robertson in his judgment in the Court of Session in Bourhill v Young186 are to the same effect. He suggests that the rationale for the exclusion is that there must be some end to the legal consequences of a careless act. An alternative more persuasive argument may be that imposing an obligation on people not to harm themselves, so as not to cause others psychiatric illness, places an undesirable restriction on their self-determination.187 Lord Oliver in
182
[1997] 3 WLR 1194. [1997] 3 WLR 1194, 1208-1209. (1984) 155 CLR 549, 604. But see Churchill v Motor Accidents Insurance Board (unreported, 2nd December 1993) where Green CJ of the Supreme Court of Tasmania doubts Deane Js dicta saying: It is worth observing that the policy reasons for denying liability in cases where psychiatric injury arises out of the death of the negligent person are not self-evident. 1941 CS 395, 399. Consultation Paper No 137, para 5.47. See further paras 5.34 to 5.43 below.
183
184
185
186
187
36
Alcock v Chief Constable of South Yorkshire Police cited Deane J and suspected that 188 an English court would reach the same conclusion.
188
37
3.2
Consultation Paper No 137, Part III. In Hegarty v EE Caledonia Ltd [1997] 2 Lloyds Rep 259, 263 Brooke LJ said that this Part of the Consultation Paper should be compulsory reading for judges and legal practitioners ... because it contains a clear, dispassionate account of PTSD and the sort of events that have been known to trigger off this kind of illness. We specifically asked those consultees with medical expertise whether they considered that our summary was fair and accurate. Those that expressed an opinion said that they did. We are grateful to Professor Simon Wessely and Mr Miles Mandelson for reading and commenting on this Part of the Report. We would like to emphasise, however, that as lawyers and not medical experts, we pass no opinion on the accuracy of the research referred to in this Part. The development of this area of the law and psychiatry without consultation between each other is criticised in M Trimble, Medicine and the Law: Conflict or Debate J Psychosom Res 1995;39:671-674. For other surveys of the medical background, see N J Mullany and P R Handford, Tort Liability for Psychiatric Damage (1993) pp 24-42 and M Napier and K Wheat, Recovering Damages for Psychiatric Injury (1995) pp 31-57. One medical consultee complained that in several compensation cases in which he had been involved, the lawyers for the plaintiff tended to push him towards a diagnosis of PTSD, when he felt that the primary condition from which the plaintiff suffered was a depressive illness. DSM-III, 1980, p 237, revised in 1987 and known as DSM-III-R. Now replaced by DSMIV, where PTSD is referred to at para 309.81, p 424.
38
6 Classification of Diseases, ICD-10, PTSD has been widely accepted as a recognisable psychiatric illness.7 Secondly, since an essential criterion for a diagnosis of the disorder is that characteristic symptoms develop following exposure to an extreme traumatic stressor,8 a plaintiff who satisfies the court that he or she is suffering from PTSD after exposure to the defendants negligent act, may have little difficulty proving legal causation. Thirdly, the diagnostic criteria for PTSD under DSM-IV specifically require that the persons reaction to the stressor should have involved intense fear, helplessness or horror.9 A plaintiff who is suffering from PTSD diagnosed under DSM-IV will therefore have little trouble satisfying the court that his or her illness was shock-induced.
3.3
However, we now believe that, in considering the medical background, it is preferable not to concentrate exclusively on PTSD. It has been claimed that this diagnosis has wrongly upstaged other conditions, such as anxiety disorder and depression, which are also consequent on trauma.10 Several consultees commented that in their experience PTSD is not the most prevalent diagnosis of psychiatric illness following trauma.11 In addition, it is generally agreed that PTSD is often found together with another psychiatric illness - in medical terms it has a high level of comorbidity.12 We are concerned that if the criteria for allowing
ICD-10, 1992, para F43.1, p 147. Although see paragraph 3.4 below. The Diagnostic and Statistical Manual of Mental Disorders, 1994 (DSM-IV) published by the American Psychiatric Association and the ICD-10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines, 1992 published by the World Health Organisation represent the two main diagnostic classificatory systems used by the psychiatric profession. See paragraph 3.5 below. See further paragraph 3.5 below. D A Alexander, Trauma Research: A New Era J Psychosom Res 1996;41:1-5. See also R J Ursano, C S Fullerton and B G McCaughey, Trauma and Disaster in R J Ursano, B G McCaughey and C S Fullerton (eds), Individual and Community Responses to Trauma and Disaster (1994) ch 1. In their response to the Consultation Paper, Dr Mayou and Dr Bryant commented: We would argue that interest in PTSD has often unreasonably overshadowed the significance of anxiety and depressive disorder. R Telford, A Rowlands and J Wright said: We believe it would be misleading to focus too heavily on post-traumatic stress disorder. Professor Alexander said that it would be inappropriate if PTSD were to usurp other diagnoses, such as depression, as major consequences of trauma. See also the decision of Thorpe LJ in Vernon v Bosley (No 1) [1997] 1 All ER 577, 610: [PTSD] is but one of many relevant psychiatric diagnoses. The fact that it is a diagnosis that necessarily involves a shock is not a reason for elevating it to the exclusion of other psychiatric illnesses that may be shock-induced. This point was made by several of the medical consultees including Professor Wessely (depressive disorders and phobic anxiety may be more common psychiatric sequelae to trauma); Dr Mayou (importance of phobic anxiety after road traffic accidents); Professor Alexander (other diagnoses, such as depression, can be the primary diagnosis after trauma and are no less worthy of compensation) and Professor Rosser (pathological grief is at least as common a sequelae after a traumatic event as PTSD). In a survey of survivors of a mass shooting spree, PTSD was found to be the most prevalent psychiatric disorder suffered, but considerable levels of comorbidity were found. Approximately 30% of those who developed PTSD were also found positive for major depression: C S North et al, Posttraumatic Stress Disorder in Survivors of a Mass Shooting Am J Psychiatry 1994;151:82-88. A higher level of comorbid major depression (53%) was found in a survey of motor vehicle accident victims: E Blanchard et al, Psychiatric Morbidity Associated with Motor Vehicle Accidents J Nerv Ment Dis
10
11
12
39
claims for compensation are developed with only one particular diagnosis, PTSD, in mind, difficulties might arise in applying those criteria more generally to 13 negligently caused psychiatric illness. Moreover, since we conclude in Part V that the requirement that the psychiatric illness be shock-induced should be abandoned, we also think it appropriate to consider the cause and prevalence of other psychiatric illnesses that do not require sudden reaction to a stressor in their diagnosis. With this in mind, we look first at recent literature on PTSD and then consider other psychiatric illnesses that may also be compensatable in a negligence action. 2. POST-TRAUMATIC STRESS DISORDER 3.4 The phrase post-traumatic stress disorder was coined in the 1970s and was officially recognised with the publication of DSM-III in 1980. Veterans returning from the Vietnam War were found to be suffering from severe stress and in need of treatment, yet there was no diagnosis to fit their syndrome. PTSD was a concept created to meet that need.14 However, the acceptance of PTSD among psychiatrists has not been universal and the diagnosis remains controversial.15 One
1995;183:495-504. In a random sample of mothers, lifetime comorbidity of PTSD with major depression, anxiety or substance abuse was 73%: N Breslau et al, Psychiatric Sequelae of Posttraumatic Stress Disorder in Women Arch Gen Psychiatry 1997;54:81-87. In a study of an urban population of young adults, 82% of persons with PTSD had one or more other psychiatric disorder: N Breslau et al, Traumatic Events and Posttraumatic Stress Disorder in an Urban Population of Young Adults Arch Gen Psychiatry 1991;48:216-222. A review of the literature on comorbidity of PTSD is included in B L Green, Psychological Research in Traumatic Stress: An Update J Traumatic Stress 1994;7:341-362. In his response, M Mandelson, a clinical psychologist, wrote: It should be noted that secondary psychological or psychiatric conditions may arise from or alongside PTSD that are directly attributable to the traumatic event.
13
This point was made to us by Dr Sally Lloyd-Bostock in commenting on a draft of our Consultation Paper. A brief history of the concept of PTSD is given by N C Andreasen in Posttraumatic Stress Disorder: Psychology, Biology, and the Manichaean Warfare Between False Dichotomies Am J Psychiatry 1995;152:963-965 and by J D Kinzie and R R Goetz in A Century of Controversy Surrounding Posttraumatic Stress-Spectrum Syndromes: The Impact on DSM-III and DSM-IV J Traumatic Stress 1996;9:159-179. [T]he premature introduction and widespread acceptance of an operational definition [PTSD] which is still of uncertain validity has restricted understanding and research, overemphasising certain clinical features at the expense of a neglect of major psychiatric comorbidity: R Mayou, Accident Neurosis Revisited Br J Psychiatry 1996;168:399-403. In their response to our Consultation Paper, Dr Mayou and Dr Bryant noted: [T]here are still considerable uncertainties about the precise definition of a syndrome of PTSD and the extent to which it is reasonable to see it as a separate disorder as compared to being a part of the depressive and anxiety syndromes which follow stress in general. We do not doubt the reality and significance of various symptoms included in the syndrome, but it would be unwise to indicate that there is a generally accepted and final view about the definition. R Telford, A Rowlands and J Wright wrote: Despite the rapid growth of studies in this area, there is still only limited theoretical understanding of the complex interactions between the stressor event, the individual and his/her reactions, biopsychosocial mechanisms and cultural responses to the traumatised person. See also, S Wessely, Liability for Psychiatric Illness J Psychosom Res 1995;39:659-669: It is still questionable, and indeed doubtful, whether a single post trauma syndrome exists across all cultures, and in response to all traumas. The boundaries between PTSD and other more familiar psychiatric syndromes that become manifest after adversity are blurred. An article reviewing the conceptual origins of PTSD and recent research findings concludes that PTSD has a controversial
14
15
40
commentator refers to the ability of mankind to survive natural disasters and plagues since the beginning of time with the majority becoming neither partially nor totally disabled. He suggests that even when a person reacts catastrophically to a shattering experience the tendency is towards recovery and that any prolonged disability must depend on other factors such as personal issues and background.16 3.5 The diagnostic criteria for PTSD in DSM-IV require that the person develop characteristic symptoms17 following exposure to a traumatic event (frequently referred to as the stressor) in which (i) the person experienced, witnessed, or was confronted with an event or events that involved actual or threatened death or serious injury, or a threat to the physical integrity of self or others, and (ii) the persons response involved intense fear, helplessness, or horror.18 A diagnosis of PTSD under ICD-10 requires that the individual have been exposed to a stressful event or situation (either short- or long- lasting) of an exceptionally threatening or catastrophic nature, which is likely to cause pervasive distress in almost anyone.19 DSM-IV and ICD-10 provide examples of the types of stressor that may give rise to PTSD. The stressors referred to in DSM-IV are direct personal experience of an event that involves actual or threatened death or serious injury, or other threat to ones physical integrity; or witnessing an event that involves death, injury, or a threat to the physical integrity of another person; or, significantly, learning about unexpected or violent death, serious harm, or threat of death or injury experienced by a family member or other close associate.20 However, it is clear that many people do not develop PTSD in response to even the most serious traumatic event.21 In the Consultation Paper we referred to some
3.6
3.7
diagnosis which is at a vulnerable point: R Yehuda et al, Conflict Between Current Knowledge About Posttraumatic Stress Disorder and Its Original Conceptual Basis Am J Psychiatry 1995;152:1705-1713.
16
M Sim, Compensation Claims (1992) pp 187-200. See further para 3.7 n 21 below. DSM-IV, para 309.81, p 424. ICD-10, para F43.1, p 147. DSM-IV, para 309.81, p 424. ICD-10 gives as examples natural or man-made disaster, combat, serious accident, witnessing the violent death of others, or being the victim of torture, terrorism, rape, or other crime: ICD-10, para F43.1, p 147. For example, in a study of survivors of a tornado that swept through northern Florida, rates of psychiatric disorders were found to be very low one month after the event. Of the 42 participants in the study, only one was found to be suffering from PTSD: C S North et al, Acute Postdisaster Coping and Adjustment J Traumatic Stress 1989;2:353-360. For a diagnosis of PTSD under DSM-IV a person must suffer from (i) persistent reexperiencing of the traumatic event; (ii) persistent avoidance of stimuli associated with the event; and (iii) persistent symptoms of increased arousal. The symptoms must last for more than a month, and must cause clinically significant distress or impairment in social, occupational, or other important areas of functioning: DSM-IV, para 309.81, p 424. ICD-10 contains a similar list of characteristic symptoms, although it does not specify the number of symptoms which must be displayed nor their minimum duration before a diagnosis can be made: ICD-10, para F43.1, p 147. Several studies have been undertaken into the prevalence of psychiatric illness in the general population. By identifying cases of particular illnesses, these studies have been able to estimate the number of people who could expect to develop a particular disorder during the course of their lifetime (referred to as the lifetime prevalence of the
17
18
19
20
21
41
of the empirical research that has been undertaken in order to assess the prevalence of PTSD in people who have been exposed to trauma.22 A recent article has undertaken an overview of the current research into PTSD. It concludes that, on average, about one quarter of individuals who are exposed to an extreme stressor go on to develop PTSD.23 However, certain kinds of trauma appear to be more strongly associated with PTSD than others. In a large survey24 designed to study the prevalence of psychiatric disorders in the United States, the National Comorbidity Survey, it was found that rape and combat exposure were more likely to cause PTSD than exposure to other stressors.25 3.8 Much of the recent empirical research into the prevalence of PTSD has continued to focus on violent trauma. In one study, 28 per cent of the survivors of a mass murder spree by a gunman in a cafeteria in Texas were found to meet the criteria for PTSD,26 and in another, four of eleven survivors of a shooting attack on a van of Hasidic students were diagnosed with PTSD.27 A study of US soldiers involved in the peace-keeping mission in Somalia reported that approximately five months after their return, eight per cent were found to meet diagnostic criteria for PTSD.28
disorder). The lifetime prevalence of PTSD in the general United States population has been estimated to be 9.2% by N Breslau et al, Traumatic Events and Posttraumatic Stress Disorder in an Urban Population of Young Adults Arch Gen Psychiatry 1991;48:216-222 and 7.8% by R C Kessler et al, Posttraumatic Stress Disorder in the National Comorbidity Survey Arch Gen Psychiatry 1995;52:1048-1060. These are much higher figures than were produced in the two earliest community lifetime prevalence studies of PTSD which found lifetime prevalence of 1.0% in St. Louis and 1.3% in North Carolina: see J Davidson and J Fairbank, The Epidemiology of Posttraumatic Stress Disorder in J Davidson and E Foa (eds), Posttraumatic Stress Disorder: DSM-IV and Beyond (1993) p 148. It has been suggested that multiple factors were involved in the recent much higher prevalences found, including differences in the diagnostic criteria, assessment procedures, and other sample characteristics: R C Kessler et al, Posttraumatic Stress Disorder in the National Comorbidity Survey Arch Gen Psychiatry 1995;52:1048-1060.
22
Consultation Paper No 137, paras 3.10 to 3.14. B L Green, Psychological Research in Traumatic Stress: An Update 1994;7:341-362. 8,098 respondents took part in the survey. 65% of men and 46% of women who reported rape as their most upsetting trauma developed PTSD. 39% of those who reported combat exposure as their most upsetting trauma developed PTSD: R C Kessler et al, Posttraumatic Stress Disorder in the National Comorbidity Survey Arch Gen Psychiatry 1995;52:1048-1060. In a study on the prevalence of PTSD in an urban population of young adults, the rate of PTSD across types of stressors was assessed as follows: sudden injury or serious accident (11.6%); physical assault (22.6%); seeing someone killed or seriously hurt (23.6%); news of sudden death or accident of a close relative or friend (21.1%); threat to life (24%); rape reported by women (80%): N Breslau et al, Traumatic Events and Posttraumatic Stress Disorder in an Urban Population of Young Adults Arch Gen Psychiatry 1991;48:216-222. C S North et al, Posttraumatic Stress Disorder in Survivors of a Mass Shooting Am J Psychiatry 1994;151:82-88. B Trappler and S Friedman, Posttraumatic Stress Disorder in Survivors of the Brooklyn Bridge Shooting Am J Psychiatry 1996;153:705-707. It was thought that the relatively high frequency of PTSD found may be attributable to the fact that the group was closely cohesive, so associating it with an extremely intense experience of grief and loss as a result of the one fatality and the critical injury of another colleague. B T Litz et al, Posttraumatic Stress Disorder Associated with Peacekeeping Duty in Somalia for US Military Personnel Am J Psychiatry 1997;154:178-184.
23
24
25
26
27
28
42
3.9
Attention has also been given, however, to more commonly occurring and possibly 29 less overwhelming trauma. In a study of non-head-injured patients presenting themselves at a hospital accident department following a road traffic accident, 11 per cent were found to be suffering from PTSD during the year following the accident.30 A more recent study of people who had sought medical treatment following a motor vehicle accident found a much higher prevalence rate: 39 per cent were suffering from PTSD under DSM-III-R criteria one to four months after the accident.31 Empirical research has also been undertaken into the prevalence of PTSD following a traumatic experience that does not involve personal danger. One study of young adults in an urban population reported that approximately 25 per cent of those who witness others killed or badly injured develop PTSD.32 One year after a major volcanic eruption in Columbia that killed approximately twenty-four thousand people, a survey of adult patients attending a primary health care clinic showed that 24 per cent were suffering from PTSD. Some of the patients had been caught up in the catastrophe itself, whereas others lived in nearby areas and had been indirectly affected, either through the loss of family members, friends, property or business, or through the social disorganisation that followed the disaster as these communities had to absorb the influx of survivors. Eleven per cent of these indirect victims were found to be suffering from PTSD one year after the disaster.33 Several studies have shown that rescue workers, volunteer or professional, may suffer from PTSD. This may be the case, whether or not their work threatened their own personal safety. For example, a study of post-traumatic stress in volunteer firefighters found that a sense of helplessness over a traumatic situation was often critical in terms of the firefighters emotional response. Many reported events in which their physical safety was not threatened, but they had felt threatened by their inability to manage the physical or emotional trauma being suffered by the victim.34 Attention has now also been given to those who deal with
3.10
3.11
29
The diagnosis of PTSD in DSM-III-R, the predecessor of DSM-IV, required that the stressor should be a psychologically traumatic event that is generally outside the range of usual human experience. This requirement was dropped from the diagnosis in DSM-IV. R Mayou, B Bryant and R Duthie, Psychiatric consequences of road traffic accidents Br Med J 1993;307:647-651. E Blanchard et al, Psychiatric Morbidity Associated with Motor Vehicle Accidents J Nerv Ment Dis 1995;183:495-504. Reasons given for the higher prevalence include the possibility of selection bias among the referral sources or a self-selection bias among victims volunteering for the study and the case finding instrument used in the later study may have been more sensitive to PTSD than that used previously. For a review of the recent studies into psychological morbidity following road traffic accidents, see A D Gallo and W ParryJones, Psychological Sequelae of Road Traffic Accidents: An Inadequately Addressed Problem Br J Psychiatry 1996;169:405-407. N Breslau et al, Traumatic Events and Posttraumatic Stress Disorder in an Urban Population of Young Adults Arch Gen Psychiatry 1991;48:216-222. B R Lima et al, Psychiatric Disorders in Primary Health Care Clinics One Year After A Major Latin American Disaster Stress Med 1991;7:25-32. R A Bryant and A G Harvey, Posttraumatic Stress Reactions in Volunteer Firefighters J Traumatic Stress 1996;9:51-62.
30
31
32
33
34
43
the aftermath of a disaster, but who are never themselves exposed to any personal danger. Researchers studied a group of volunteer disaster workers who assisted professional mortuary staff following an explosion on a US naval ship which killed 47 of the crew. Probable PTSD was found to be present in 11 per cent after 1 month, reducing to ten per cent at 4 months and two per cent at 13 months.35 A study of 31 dentists who had identified the dead from the fire at the Branch Davidian compound in Waco, Texas, showed that post-traumatic stress symptoms were present, and that distress was significantly related to the hours of exposure to the remains, prior experience handling remains, age, and the support received from spouses and co-workers during the identifications.36 3.12 Little research appears yet to have been undertaken into the prevalence of PTSD suffered as a result of learning about the death of a loved one.37 However, one study found that of those relatives and friends, who experienced traumatic bereavements, 25 per cent developed PTSD.38 Some research has also been undertaken into the prevalence not of PTSD suffered but of exposure to traumatic events that would qualify as a stressor for PTSD purposes. It is difficult to analyse this research for our purposes, since the categories that it uses do not coincide with possible grounds of action in negligence. But in a study which estimated the lifetime prevalence of exposure to various trauma, two categories may be of interest to us: the estimated lifetime prevalence of witnessing another person being seriously injured or hurt was found to be 36 per cent for men and 15 per cent for women; while the lifetime
3.13
35
R Ursano et al, Longitudinal Assessment of Posttraumatic Stress Disorder and Depression after Exposure to Traumatic Death J Nerv Ment Dis 1995;183:36-42. In contrast to this finding, very little distress was found in a small group of police body handlers following the Piper Alpha disaster. The good outcome was attributed to organisational management and the sense of doing a good job: D A Alexander, Stress among police body handlers: a long term follow-up Br J Psychiatry 1993;163:806-808. In a study on the anticipatory stress felt by military mortuary workers prior to the arrival of human remains from the Gulf War, discomfort with mutilation and the grotesque were found to be significant predictors of intrusive thoughts and avoidance, two of the primary symptoms of PTSD: J McCarroll et al, Anticipatory Stress of Handling Human Remains from the Persian Gulf War J Nerv Ment Dis 1995;183:698-703. A study of individuals involved in the recovery of war dead from the Gulf War found that after more than one year, individuals who had handled human remains were at higher risk for PTSD symptoms than those who had not: J E McCarroll et al, Symptoms of PTSD Following Recovery of War Dead: 13-15 Month Follow-Up Am J Psychiatry 1995;152:939-941. J E McCarroll et al, Posttraumatic Stress Symptoms Following Forensic Dental Identification: Mt. Carmel, Waco, Texas Am J Psychiatry 1996;153:778-782. Identification and emotional involvement play an important role in the experience of disaster workers, rescue workers and families of victims. This mechanism is an important avenue through which the trauma of disaster is propagated to wider and wider circles. Our understanding of this mechanism is in its infancy: R J Ursano, B G McCaughey and C S Fullerton, The structure of human chaos in R J Ursano, B G McCaughey and C S Fullerton (eds), Individual and Community Responses to Trauma and Disaster (1994) p 403 at p 405. However, the study was concerned with PTSD following criminal victimisation, and therefore only looked at traumatic bereavement following homicide or drunk driving: D G Kilpatrick and H S Resnick, Posttraumatic Stress Disorder Associated with Exposure to Criminal Victimisation in Clinical and Community Populations in J Davidson and E Foa (eds), Posttraumatic Stress Disorder: DSM-IV and Beyond (1993) ch 7.
36
37
38
44
prevalence of suffering a great shock because a traumatic event has happened to somebody close was estimated to be 11 per cent for men and 12 per cent for women. However, since there is no indication of how the trauma to the close 39 In a survey of person occurred, these statistics may be of limited relevance. young adults (aged 21 to 30), it was found that seven per cent had witnessed someone being seriously hurt or killed and six per cent had received news of the sudden death or injury of a close relative or friend.40 3.14 Although it now appears to be generally accepted that there is a relationship between the extremity of the stressor and the prevalence of PTSD (a doseresponse relationship),41 research has also been carried out into whether other factors, beside the stressor, may play a role in the development of PTSD.42 These studies have identified prior vulnerability by reference to gender, education, socioeconomic status, prior psychiatric disorder43 and prior trauma.44 Factors after the
39
R C Kessler et al, Posttraumatic Stress Disorder in the National Comorbidity Survey Arch Gen Psychiatry 1995;52:1048-1060. N Breslau et al, Traumatic Events and Posttraumatic Stress Disorder in an Urban Population of Young Adults Arch Gen Psychiatry 1991;48:216-222. A review of various studies that quantified stressors suffered during combat, disaster, illness, injury and crime found that in 16 of the 19 studies examining the question of stressor intensity, a dose-response relationship between the stressor intensity and outcome was found: J S March, What Constitutes a Stressor? The Criterion A Issue in J Davidson and E Foa (eds), Posttraumatic Stress Disorder: DSM-IV and Beyond (1993) ch 3. This is not, however, universally accepted. The studies carried out by A McFarlane on a group of firefighters who were exposed to a devastating Australian bushfire has cast doubt on the suggested central aetiological role given to the stressor. He found neither the severity of the exposure to the disaster nor the losses sustained to be the major determinant of posttraumatic morbidity in the firefighters. Other factors, such as neuroticism and a past history of treatment for psychological disorder, were found to be better predictors of PTSD than the degree of exposure: A C McFarlane, The Aetiology of Post-traumatic Morbidity: Predisposing, Precipitating and Perpetuating Factors Br J Psychiatry 1989;154:221-228. A prospective study of inpatients who had suffered injury to a lower limb found that distress post injury, rather than the severity of the stressor, was the best predictor of PTSD: A Feinstein and R Dolan, Predictors of post-traumatic stress disorder following physical trauma: an examination of the stressor criterion Psychological Med 1991;21:85-91. The relationship between severity of exposure and the development of PTSD is also doubted in S M Southwick et al, Consistency of Memory for Combat-Related Traumatic Events in Veterans of Operation Desert Storm Am J Psychiatry 1997;154:173-177 where a study of veterans of the Gulf War found that a majority displayed inconsistent recall for specific features of combat trauma. A detailed review of studies into prestressor and poststressor factors in the development of PTSD is contained in J D Bremner, S M Southwick and D S Charney, Etiological Factors in the Development of Posttraumatic Stress Disorder in C M Mazure (ed), Does Stress Cause Psychiatric Illness? (1995) p149. Several such studies are also reviewed in B L Green, Psychological Research in Traumatic Stress: An Update J Traumatic Stress 1994;7:341362. Research into this area is, however, continuing, and a review of much of the literature on the role that factors, other than the stressor, may play in the development of PTSD concludes: The understanding of the role of personality and vulnerability factors in determining the psychological consequences of traumatic stress remains an issue of controversy: A C McFarlane, Vulnerability to Posttraumatic Stress Disorder in M E Wolf and A D Mosniam (eds), Posttraumatic Stress Disorder: Etiology, Phenomenology and Treatment (1990) ch 1. Most studies show that those with psychiatric illness prior to a traumatic event or disaster are at increased risk: R J Ursano, C S Fullerton and B G McCaughey, Trauma and Disaster in R J Ursano, B G McCaughey and C S Fullerton (eds), Individual and
40
41
42
43
45
stressor may also affect the development of PTSD,45 either by reducing the chance of onset, for example family and professional support,46 or by increasing the risk of the disorder, for example ongoing stressors such as unemployment or loss of a
Community Responses to Trauma and Disaster (1994) ch 1. See for example, H L Chubb and J I Bisson, Early Psychological Reactions in a Group of Individuals with Pre-Existing and Enduring Mental Health Difficulties Following a Major Coach Accident Br J Psychiatry 1996;169:430-433. This study of 21 survivors of a coach crash who were patients of two local consultant psychiatrists found a high incidence of PTSD (50%), although without a control group of individuals without pre-existing mental health difficulties one could argue that a sample of the general population would have similarly suffered. Prior psychiatric disorder has not invariably been found to be a significant predictor: C S North et al, Posttraumatic Stress Disorder in Survivors of a Mass Shooting Am J Psychiatry 1994;151:82-88, R A Bryant and A G Harvey, Posttraumatic Stress in Volunteer Firefighters: Predictors of Distress J Nerv Ment Dis 1995;183:267-271 and A E Skodol et al, PTSD Symptoms and Comorbid Mental Disorders in Israeli War Veterans Br J Psychiatry 1996;169:717-725.
44
At one time it was thought that prior trauma would increase the resilience of an individual to subsequent traumas, thereby reducing the risk of PTSD with subsequent reexposure. More recent research does not however appear to be consistent with this theory, but rather supports the stress sensitisation theory, which holds that exposure to repeated stress makes an individual more susceptible to the effects of stress: J D Bremner, S M Southwick and D S Charney, Etiological Factors in the Development of Posttraumatic Stress Disorder in C M Mazure (ed), Does Stress Cause Psychiatric Illness? (1995) ch 6. It has also been suggested that certain vulnerability factors do not only influence the likelihood of disorder in persons exposed to trauma, but they may also influence disorder in part by increasing the likelihood of exposure to traumatic events. In a study of an urban population of young adults, it was found that risk factors for exposure to traumatic events included low education, male sex, early conduct problems, extroversion, and family history of psychiatric disorder or substance problems. Risk factors for PTSD following exposure included early separation from parents, neuroticism, pre-existing anxiety or depression, and family history of anxiety: N Breslau et al, Traumatic Events and Posttraumatic Stress Disorder in an Urban Population of Young Adults Arch Gen Psychiatry 1991;48:216-222. See also N Breslau et al, Risk Factors for PTSD-Related Traumatic Events: A Prospective Analysis Am J Psychiatry 1995;152:529-535. The contribution of the recovery environment to individual and community responses to traumatic events cannot be overemphasised: R J Ursano, B G McCaughey and C S Fullerton, The structure of human chaos in R J Ursano, B G McCaughey and C S Fullerton (eds), Individual and Community Responses to Trauma and Disaster (1994) p 403 at p 407. Support from spouses during the identification by dentists of remains from the fire at the Branch Davidian compound in Waco, Texas was reported to be extremely important and was associated with a lower level of self-reported stress: J E McCarroll et al, Posttraumatic Stress Symptoms Following Forensic Dental Identification: Mt. Carmel, Waco, Texas Am J Psychiatry 1996;153:778-782.
45
46
46
47 loved one. One study produced results that suggest that a trauma survivors risk of PTSD may be related to his or her familys history of PTSD-like behaviour.48
3. OTHER PSYCHIATRIC ILLNESSES 3.15 In this brief outline of the medical background, it is not possible for us to survey all the psychiatric illnesses that could conceivably be compensatable in a negligence action. Nor is it possible for us to do justice to the abundant research that has been carried out in relation to these disorders. Instead, we attempt to outline what appear to be the most salient points relevant to the laws possible development in this area. There are many psychiatric illnesses which do not require exposure to a catastrophic event for a diagnosis to be made, but which studies have shown to be frequently suffered by those affected by trauma. We will consider three such disorders: depressive disorders, adjustment disorders and anxiety disorders. (1) Depressive Disorders 3.17 It is part of normal human experience to feel unhappy at times of adversity but, as we explained in our review of the present law, mere grief or distress is not compensatable at law.49 Depressive disorders are distinguished from more ordinary fluctuations in mood by the severity and duration of symptoms, and by the effect of these symptoms on nearly all activities.50 In broad terms, a depressive disorder is characterised by a cluster of symptoms including: depressed mood most of the day, nearly every day, for at least 2 weeks; loss of interest or pleasure in activities that are normally pleasurable; decreased energy; change in appetite; sleep disturbance; difficulties in concentrating; feelings of worthlessness; and recurrent thoughts of death.51 Several studies have shown that depressive disorders often follow severe stressors, such as the loss of a loved one.52 For example, in a study of survivors of a shooting
3.16
3.18
47
R A Bryant and A G Harvey, Posttraumatic Stress in Volunteer Firefighters: Predictors of Distress J Nerv Ment Dis 1995;183:267-271. A study of war veterans into the relationship between PTSD and stressful life events found that the more severe the PTSD, the greater the number of negative events that are likely to occur the following year. This supports the theory that the presence of a mental disorder leads to stressful events, which in turn exacerbate the disorder: Z Solomon, M Mikulincer and H Flum, Negative Life Events, Coping Responses, and Combat-Related Psychopathology: A Prospective Study J Abnormal Psychology 1988;97:302-307. C S Watson et al, Posttraumatic Stress Disorder (PTSD) Symptoms in PTSD Patients Families of Origin J Nerv Ment Dis 1995;183:633-638. See para 2.3 above. C T Kaelber, D E Moul and M E Farmer, Epidemiology of Depression in E E Beckham and W R Leber (eds), Handbook of Depression (2nd ed 1995) p 3 at p 5. See the diagnostic criteria for Major Depressive Disorder in DSM-IV, para 296.2, p 344 and for Depressive episode in ICD-10, para F32, p 119. DSM-IV, p 342. Symptoms will not qualify for a diagnosis of Major Depressive Episode under DSM-IV unless they are not better accounted for by Bereavement, i.e., after the loss of a loved one, the symptoms persist for longer than 2 months or are characterised by marked functional impairment, morbid preoccupation with worthlessness, suicidal ideation, psychotic symptoms or psychomotor retardation: DSM-IV, p 327. The validity of requiring
48
49
50
51
52
47
attack on a van of Hasidic students, five out of the eleven participants were found 53 to be suffering from major depressive disorder. A study of 23 households that survived an earthquake in Western India that killed more than ten thousand people found that 23 per cent were suffering from PTSD, 21 per cent from major depression, and nine per cent from adjustment disorder; the death of a relative was found to be associated with major depression, although not with other disorders.54 3.19 However, bereavement is not the only stressor associated with depressive disorder.55 For example, a study of 31 individuals who had been admitted to hospital after mountain accidents found that a large minority experienced at least mild depression, with four suffering from at least a mild diagnosable depression immediately after the accident.56 Several studies suggest that stressors may play a more significant role in the precipitation of the first or second episodes of depressive disorder and may play less of a role in the onset of subsequent episodes.57 While it is generally accepted that stressful life events play a critical role in bringing about depressive disorders, many people do not suffer depression even following the most severe of losses. It has been suggested that certain vulnerability factors (such as low self-esteem) increase the likelihood of depression in the presence of a stressor.58 Several studies have shown that depressive disorders are approximately
3.20
depressive symptoms which otherwise satisfy all the criteria of major depression to cross a threshold of severity in the case of bereavement is questioned by E G Karam, The Nosological Status of Bereavement-Related Depressions Br J Psychiatry 1994;165:48-52. The distinction between normal grief and a psychiatric illness is discussed in C M Parkes, Bereavement Br J Psychiatry 1985;146:11-17; P Burnett et al, Concepts of Normal Bereavement J Traumatic Stress 1994;7:123-128 and K Kim and S Jacobs, Stress of Bereavement and Consequent Psychiatric Illness in C M Mazure (ed), Does Stress Cause Psychiatric Illness? (1995) ch 7. See also M Gelder, D Gath, R Mayou and P Cowen, Oxford Textbook of Psychiatry (3rd ed 1996) pp 151-154 and G W Brown and T O Harris, Depression in G W Brown and T O Harris (eds), Life Events and Illness (1989) p 49.
53
B Trappler and S Friedman, Posttraumatic Stress Disorder in Survivors of the Brooklyn Bridge Shooting Am J Psychiatry 1996;153:705-707. P Sharan et al, Preliminary Report of Psychiatric Disorders in Survivors of a Severe Earthquake Am J Psychiatry 1996;153:556-558. One study found that while the risk for onset of major depression in the month following the death of a close relative was increased 20-fold, in the month following the serious illness of a close relative, the risk was increased 4-fold: K S Kendler et al, Stressful Life Events, Genetic Liability, and Onset of an Episode of Major Depression in Women Am J Psychiatry 1995;152:833-842. D F Peck, A Robertson and S Zeffert, Psychological Sequelae of Mountain Accidents: A Preliminary Study J Psychosom Res 1996;41:55-63. DSM-IV, p 342. See also D Pardoen et al, Life events and Primary Affective Disorders: A One Year Prospective Study Br J Psychiatry 1996;169:160-166. For an overview of the role that stressors or life events play in the onset and course of depression, see R C Cronkite and R H Moos, Life Context, Coping Processes, and Depression in E E Beckham and W R Leber (eds), Handbook of Depression (2nd ed 1995) ch 22. G W Brown and T O Harris, Depression in G W Brown and T O Harris (eds), Life Events and Illness (1989) ch 2 and G W Brown and T O Harris, Aetiology of anxiety and depressive disorders in an inner-city population. 1. Early adversity Psychological Med 1993;23:143-154. However, not all commentators accept these findings: see M Gelder, D Gath, R Mayou and P Cowen, Oxford Textbook of Psychiatry (3rd ed 1996) pp 216-217. In a
54
55
56
57
58
48
59 twice as common in females than in males. In the National Comorbidity Survey referred to earlier,60 lifetime prevalence of major depressive episode was estimated to be 17 per cent.61
(2) Adjustment Disorders 3.21 The essential feature of an adjustment disorder is the development of clinically significant emotional or behavioural symptoms in response to an identifiable stressor. For a diagnosis under DSM-IV, the symptoms must occur within 3 months of the onset of the stressor, which may be of any severity (ranging from exposure to a natural disaster to the termination of a relationship).62 The diagnosis under ICD-10 can only be given if the symptoms occur within 1 month of exposure to the stressor, which must not be of an unusual or catastrophic type if this diagnosis is to be given.63 Under both diagnoses the symptoms must not persist for more than 6 months after the cessation of the stressor. Individual predisposition or vulnerability is thought to play a greater role in the risk of occurrence and the shaping of the manifestations of an adjustment disorder than it does in PTSD, but it is still assumed that the condition would not have arisen without the stressor.64 In the study of survivors of a shooting attack on a van of Hasidic students, referred to earlier, two of the eleven participants were found to be suffering from
3.22
3.23
random study of women, it was found that women with pre-existing anxiety and PTSD had significantly increased risk for first onset major depression. However, women exposed to trauma who did not develop PTSD, had no increased risk of depression. It was suggested that either PTSD causes the depression, or PTSD is the marker of a pre-existing vulnerability to major depression: N Breslau et al, Psychiatric Sequelae of Posttraumatic Stress Disorder in Women Arch Gen Psychiatry 1997;54:81-87. One study found that genetic factors influenced the risk of onset of major depression in part by altering the sensitivity of individuals to the depression-influencing effect of stressful life events: K S Kendler et al, Stressful Life Events, Genetic Liability, and Onset of an Episode of Major Depression in Women Am J Psychiatry 1995;152:833-842. The strongest predictors of liability to major depression have been found to be, in descending order, stressful life events, genetic factors, previous history of major depression and neuroticism: K S Kendler et al, The Prediction of Major Depression in Women: Toward an Integrated Etiologic Model Am J Psychiatry 1993;150:1139-1148.
59
C T Kaelber, D E Moul and M E Farmer, Epidemiology of Depression in E E Beckham and W R Leber (eds), Handbook of Depression (1995) ch 1. See para 3.7 above. R C Kessler et al, Lifetime and 12-Month Prevalence of DSM-III-R Psychiatric Disorders in the United States Arch Gen Psychiatry 1994;51:8-19. DSM-IV, para 309, p 623. ICD-10, para F43.2, p 149. ICD-10, para F43.2, p 149.
60
61
62
63
64
49
65 adjustment disorder. In the study of survivors of an earthquake in India, also referred to earlier, nine per cent suffered from an adjustment disorder.66
(3) Anxiety Disorders67 3.24 Anxiety disorders may be phobic disorders, where anxiety is evoked only (or predominantly) by certain well-defined situations which are not currently dangerous, or may be generalised disorders, where the essential feature is a generalised and persistent anxiety which is not restricted to any particular environmental circumstances. Coexistent adjustment and anxiety disorders68 are common, as are coexistent depressive and anxiety disorders.69 Both phobic and generalised anxiety disorders have been found to be prevalent after trauma. For example, in the study mentioned above of patients presenting themselves at a hospital emergency department following road traffic accidents, 18 per cent were diagnosed as suffering phobic anxiety about travel at 1 year after the accident.70 Similarly, in the study mentioned above of patients attending a primary health care clinic one year after a major volcanic eruption in Columbia, although PTSD was the most prevalent diagnosis among the patients as a whole, when looking only at the group of indirect victims (those who had not experienced the disaster first hand, but who suffered as a result of the loss of a relative, friend, property, business or disruption to their community), the most prevalent diagnosis (22 per cent) was generalised anxiety disorder.71 Indeed, the prevalence rate for anxiety disorders in society is high. In a mass population survey of adults living in private households in Great Britain,72 7.7 per cent were diagnosed as having mixed anxiety and depressive disorder and 3.1 per cent with generalised anxiety disorder in the week before interview. The prevalence of all neurotic disorder was higher among women than men.73 In the National Comorbidity Study, the lifetime prevalence of an anxiety disorder was estimated to be 24.9 per cent, the second most prevalent disorder after substance
3.25
3.26
65
See para 3.18. B Trappler and S Friedman, Posttraumatic Stress Disorder in Survivors of the Brooklyn Bridge Shooting Am J Psychiatry 1996;153:705-707. See para 3.18. P Sharan et al, Preliminary Report of Psychiatric Disorders in Survivors of a Severe Earthquake Am J Psychiatry 1996;153:556-558. PTSD is included as an anxiety disorder in DSM-IV, whereas ICD-10 categorises PTSD as a reaction to severe stress. See the diagnosis for Adjustment Disorder with Anxiety DSM-IV, para 309.24, p 624 and Adjustment disorder with mixed anxiety and depressive reaction ICD-10, para F43.22, p 151. See the diagnosis for Mixed anxiety-depressive disorder DSM-IV, para 300.00, p 444 and Mixed anxiety and depressive disorder ICD-10, para F41.2, p 141. R Mayou, B Bryant and R Duthie, Psychiatric consequences of road traffic accidents Br Med J 1993;307:647-651. B R Lima et al, Psychiatric Disorders in Primary Health Care Clinics One Year After A Major Latin American Disaster Stress Med 1991;7:25-32. 10,108 adults took part in the study. P Mason and G Wilkinson, The Prevalence of Psychiatric Morbidity: OPCS Survey of Psychiatric Morbidity in Great Britain Br J Psychiatry 1996;168:1-3.
66
67
68
69
70
71
72
73
50
abuse.74 Because of these high figures, it has been suggested that the congruence of one particular stressor, such as a stressful job, and a neurotic disorder will not be synonymous with causation.75 Furthermore, anxiety disorders can be caused by a wide combination of factors. Potential risk factors include predisposition, early environment, social support, family situation and recent life events.76 Continual stress has an eroding effect and the cumulative effects of stressful life events may ultimately exhaust a persons ability to cope.77 4. THE DISTINCTION BETWEEN PSYCHIATRIC ILLNESS AND MERE MENTAL DISTRESS 3.27 The distinction between what constitutes mere mental distress and symptoms that amount to a recognisable psychiatric illness is not clear.78 Several medical consultees referred to this difficulty in their responses,79 one going so far as to suggest that the overlap between mental health and illness is so large a grey area that it is not suitable for the legal purpose to which the diagnosis is being put.80 Reliance on the diagnostic criteria contained in ICD-10 and DSM-IV is not always sufficient to distinguish those with the greatest impairment of functioning. A recent study of individuals who experienced disaster at close range demonstrated that there was a large number of people who failed to meet the full criteria for PTSD, but who were suffering from a number of symptoms that would require more or less the same level of care. The survey group consisted of individuals who had survived the Bijlmermeer disaster when an aircraft crashed into two apartment buildings in a Netherlands suburb. The post-traumatic stress reaction in individuals was divided into two categories: PTSD and partial PTSD (in partial PTSD, there are insufficient symptoms and/or insufficient symptom groups to
3.28
74
R C Kessler et al, Lifetime and 12-Month Prevalence of DSM-III-R Psychiatric Disorders in the United States Arch Gen Psychiatry 1994;51:8-19. S Wessely, Liability for Psychiatric Illness J Psychosom Res 1995;39:659-669. S Wessely, Liability for Psychiatric Illness J Psychosom Res 1995;39:659-669. This point was made by Dr Weller in his response to our Consultation Paper. The introduction to DSM-IV states that, although this manual provides a classification of mental disorders, it must be admitted that no definition adequately specifies precise boundaries for the concept of mental disorder: DSM-IV, Introduction, p xxi. See also, R Mayou, Psychological, Quality of Life and Legal Consequences of Road Traffic Accident Injury [1995] JPIL 277, 280: [D]espite their very considerable advantages, [the standard classifications] remain arbitrary with an overlap between categories and without clear justifications for distinctions between normality and mental disorder. For example, Professor Wessely referred to the ambiguity of psychiatric diagnosis; R Telford, A Rowlands and J Wright said: There is no general agreement inside or outside psychiatry about the definition of a psychiatric case. The British Medical Association stated that: [T]here is no sudden cut-off point where grief and other distresses suddenly become psychiatric illnesses. Professor Trimble said that the distinction between normal and abnormal grief is a very slender one. See also B Mahendra, Nothing but the whole tort (1996) 146 NLJ 1022: The distinction between normal grief and pathological psychiatric illness following bereavement is clearer in the eyes of the law than to doctors. And M McCulloch et al, Post Traumatic Stress Disorder: turning the tide without opening the floodgates (1995) 35 Med Sci Law 287: [S]adness and unhappiness shade into reactive depression and illness. Dr Parkes.
75
76
77
78
79
80
51
qualify for the diagnosis of PTSD). The results of the study showed that 26 per cent had full-blown PTSD, while 44 per cent had partial PTSD.81 3.29 Several consultees commented that it would be unjust to rely on the DSM-IV or ICD-10 criteria to distinguish psychiatric illness from mere mental distress. These categories were said not to reflect the complexities of the psychological impact of trauma82 and to exclude some diagnoses that are generally accepted.83 DSM-IV itself specifically cautions that it was developed for clinical, educational and research purposes and in many cases the clinical diagnosis of a DSM-IV disorder is not sufficient to establish the existence of a mental disorder for legal purposes, because of the imperfect fit between the questions of ultimate concern to the law and the information contained in a clinical diagnosis. Furthermore, it states that it includes, for research and clinical purposes, diagnostic categories such as pathological gambling and paedophilia, but that this should not imply that these conditions meet legal or other nonmedical criteria for what constitutes mental disorder. The clinical and scientific considerations involved in the categorisation of these conditions as mental disorders may not be relevant to legal judgments which take into account such issues as individual responsibility, level of disability and competency.84 5. THE POSSIBILITY OF FRAUD 3.30 The possibility of fraudulent or exaggerated claims is recognised by the psychiatric profession by the inclusion in DSM-IV of two diagnoses involving the intentional creation or feigning of physical or psychological symptoms. The first, factitious disorder,85 occurs where the feigning is carried out in the absence of external incentives for the behaviour, Munchhausens syndrome probably being one of the
81
I V E Carlier and B P R Gersons, Partial Posttraumatic Stress Disorder (PTSD): The issue of Psychological Scars and the Occurrence of PTSD Symptoms J Nerv Ment Dis 1995;183:107-109. Similarly, the results of a Canadian study suggested that partial PTSD carries with it a burden of disability that approaches - if not entirely matches - that produced by full PTSD: M B Stein et al Full and Partial Posttraumatic Stress Disorder: Findings From a Community Survey Am J Psychiatry 1997;154:1114-1119. This point was made by R Telford, A Rowlands and J Wright. For example, Dr Parkes said that although the diagnostic category of pathological grief has been well established, its omission from DSM-IV has resulted in its under use in compensation cases. Compensation was awarded for pathological grief disorder in Vernon v Bosley (No 1) [1997] 1 All ER 577. Recent research has shown that the symptoms of complicated grief are distinct from those of bereavement-related depression or anxiety: H G Prigerson et al, Complicated Grief as a Disorder Distinct from Bereavement-Related Depression and Anxiety: A Replication Study Am J Psychiatry 1996;153:1484-1486. In M J Horowitz at al, Diagnostic Criteria for Complicated Grief Disorder Am J Psychiatry 1997;154:904-910, it is argued that a new category of complicated grief disorder should be added to DSM-IV and that its absence has probably led to an underestimation of the frequency of increased psychopathology after the death of a loved one. See also, M J Horowitz et al, Pathological Grief: Diagnosis and Explanation Psychosom Med 1993;55:260-273. Another more controversial diagnosis, chronic fatigue syndrome (CFS), is also excluded from DSM-IV. CFS was, however, accepted as a recognisable psychiatric illness in Page v Smith [1996] AC 155. DSM-IV, Introduction, p xxiii and Cautionary Statement, p xxvii. DSM-IV, pp 471-475.
82
83
84
85
52
86 best known types of factitious disorder. The second, malingering, occurs where the patients feigning is specifically motivated by an obvious goal, such as financial compensation.87
3.31
But while some medical commentators consider that fraud or exaggeration are commonplace88 and while undoubtedly cases of feigned psychosis or actual fabrication do arise,89 the medical literature suggests that such fraud or exaggeration is not common.90 The majority of studies indicate that if subjects with disabilities for which no organic cause can be found have not gone back to work before their court case they are unlikely to go back afterwards. This is the case whether they receive large amounts of compensation, small amounts or none at all.91 Numerous tests have been developed which can help to ascertain whether an individual has faked or exaggerated psychological symptoms and whether he or
3.32
86
DSM-IV, para V65.2, p 683. For discussion on the problems of malingering, see M Napier and K Wheat, Recovering Damages for Psychiatric Injury (1995) pp 100-102. They conclude that: [T]he problem of malingering can be dealt with satisfactorily by competent doctors and lawyers, and is no more prevalent in psychiatric disorder cases than it is in those of a physical nature. See, for example, M Sim, Compensation Claims (1992) p 71: Malingering is not uncommon and in compensation cases there is, more often than not, an element of it ranging from slight exaggeration to frank simulation. See, for example, L A Neal and M C Rose, Factitious Post Traumatic Stress Disorder: a case report (1995) 35 Med Sci Law 352; L Sparr and L D Pankratz, Factitious Posttraumatic Stress Disorder Am J Psychiatry 1983;140:1016-1019; L D Pankratz, Continued Appearance of Factitious Posttraumatic Stress Disorder Am J Psychiatry 1990;147:811-812 and R B Lacoursiere, Diverse Motives for Fictious Post-traumatic Stress Disorder J Traumatic Stress 1993;6:141-149. R Mayou, Accident Neurosis Revisited Br J Psychiatry 1996;168:399-403 and Psychological, Quality of Life and Legal Consequences of Road Traffic Accident Injury [1995] JPIL 277, 287. In a six year study of 96 road accident victims who claimed compensation, behaviour designed to maximise financial compensation was found to be very uncommon. It was concluded that: [t]he prospect of greater financial reward did not appear to be a major factor leading to reports of increased disability or to delay in return to full activity, and the medical and psychosocial outcome for claimants and for those not able to claim compensation was very similar: B Bryant, R Mayou and S Lloyd-Bostock, Compensation Claims following Road Accidents: a six-year follow-up study (1997) 37 Med Sci Law 326. See also S Wessely, Liability for Psychiatric Illness J Psychosom Res 1995;39:659-669. Although note N R C Leng and A J Parkin, The Detection of Exaggerated or Simulated Memory Disorder by Neuropsychological Methods J Psychosom Res 1995;39:767-776 who report that the incidence of malingering is unknown, but has been estimated to lie anywhere within the range of 1-50%. D Healy, Images of Trauma (1993) p 108 and M J Tarsh and C Royston, A Follow-up Study of Accident Neurosis Br J Psychiatry 1985;146:18-25. In his response, Dr Weller referred to the body of literature which emphasises that settlement of the financial aspects will not affect the prognosis of psychiatric illness, but the litigation process may well aggravate distress. He argues that there is no scientific evidence to show that settlement will lead to substantial recovery. See also, G Mendelson, Compensation Neurosis Revisited: Outcome Studies of the Effects of Litigation J Psychosom Res 1995;39:695-706. This study of 760 litigants who had suffered physical and/or psychiatric injury after involvement in an accident found that of the 264 who were not working at the time of conclusion of litigation and who could be traced, 75% were not working after an average of 23 months following the conclusion of their cases.
87
88
89
90
91
53
she is a reliable informant.92 The tests are objective and are often scored by a computer. They should always be complemented by clinical evaluation, by an examination of the persons pre- and post-accident functioning, and by corroborative interviews with family members.93 6. CONCLUSIONS 3.33 We noted in the Consultation Paper that the studies drawn upon were open to more than one interpretation and that different studies produced different findings. Furthermore, the majority of the studies drawn on were compiled for treatment and medical research purposes, and they did not therefore consider how many of those who were identified as suffering from a psychiatric illness sought compensation or were likely to have been successful in doing so. However, in the light of our further research and the helpful responses we received from medical consultees, we feel able to expand upon the initial conclusions which we set out in the Consultation Paper.94 We suggest that five points of significance for the future development of the law in this area emerge from the medical evidence: (1) Psychiatry does recognise a distinction between mere mental distress and psychiatric illness, although the distinction between the two is a matter of degree rather than kind and, as medical knowledge advances, changes over time. A person may suffer psychiatric illness as a result of exposure to a traumatic event, including injury or risk of injury to him or herself, witnessing injury or risk of injury to another, or being told about the injury or unexpected death of a family member or close friend.
(2)
92
For a description of the structured interviews available and the scales used for scoring PTSD see DD Blake et al, The Development of a Clinician-Administered PTSD Scale J Traumatic Stress 1995;8:75-90 and J Shepherd, P Richmond and D Miers, Assessing General Damages: A Medical Model (1994) 144 NLJ 162, 163-164. A recent study criticised the methodology of earlier research which raised doubts about psychologists ability to detect malingering and, using improved methodology, found that 86.4% of subject psychologists were able to detect malingering from insanity: S Bourg, E J Connor and E E Landis, The Impact of Expertise and Sufficient Information on Psychologists Ability to Detect Malingering Behavioural Sciences and the Law 1995;13:505-515. Consultation Paper No 137, para 3.15.
93
94
54
(3)
Not all psychiatric illness that is suffered as a result of injury or risk of injury to oneself or another is shock-induced; and PTSD is not the only, or necessarily the most common, psychiatric illness that is suffered after exposure to trauma. Other factors, beside exposure to a traumatic event, can influence the onset and prognosis of psychiatric illness. There is a serious risk that the floodgates of litigation would be opened if the sole test for liability in negligence was whether it was reasonably foreseeable that psychiatric illness would be caused to the plaintiff.
(4)
(5)
55
4.2
4.3
Consultation Paper No 137, para 1.9. Consultation Paper No 137, para 5.69. 84% of consultees who responded on this point agreed that it would not be sensible at present to attempt complete codification. In the Consultation Paper we asked consultees whether they agreed with our provisional view that legislation is required to reform the law in the central area where the defendant has negligently injured or imperilled someone other than the plaintiff and the plaintiff, as a result, has foreseeably suffered a psychiatric illness: Consultation Paper No 137, para 5.69. 77% of the consultees who responded to this question agreed with our provisional view. [1992] 1 AC 310.
56
On the contrary, we recommend that, while legislation curing serious defects in the present law is appropriate, the law should otherwise be allowed to develop by judicial decision-making. 4.4 We have divided this Section into four Parts. The first (Part V) deals with difficulties that are thrown up by five issues which apply generally to liability for negligently inflicted psychiatric illness. These are: (i) the need to establish a recognisable psychiatric illness; (ii) the test of reasonable foreseeability; (iii) the shock requirement; (iv) recovery where the immediate victim is the defendant; and (v) the classification of primary and secondary victims. The second (Part VI) looks at the core area which we consider is in need of 6 legislative reform. This is the law relating to liability to a plaintiff who suffers psychiatric illness as a result of the injuring or imperilling of a person with whom the plaintiff has a close tie of love and affection. The third Part of this Section (Part VII) deals with the areas where we do not recommend that there should be legislative reform. First, we look at the law relating to liability to a plaintiff who suffers psychiatric illness pursuant to his or her involvement in, or sight of, an accident caused by the defendant which results in injury or imperilment, but where the plaintiff does not have a close tie of love and affection with the person who is injured or imperilled. Here we consider liability in relation to a rescuer, to an involuntary participant, to an employee and to a mere bystander. Secondly, we look at the law relating to liability to a plaintiff who has suffered psychiatric illness otherwise than as a result of his or her own or another persons injury or imperilment. Here we consider liability for psychiatric illness brought on by stress at work, pursuant to property damage and pursuant to the negligent communication of distressing news. In the final Part of this Section (Part VIII) we consider the relationship between our recommended legislation and the common law. In particular, we consider whether the legislation which we have recommended should provide rights which are in addition to, rather than in replacement of, a plaintiff s rights at common law.
4.5
4.6
4.7
Although in this Report and the draft legislation set out in Appendix A we continue to use the term plaintiff, we recognise that if the proposals recommended by Lord Woolf are implemented, it may be more appropriate to refer to the claimant: Access to Justice: Final Report to the Lord Chancellor on the Civil Justice System in England and Wales (1996) ch 12.
57
5.2
5.3
See para 2.3 above. See para 3.2 above. See further para 3.29 above. A need for caution against over-rigid adherence to the DSM criteria is expressed by L A Neal in The Pitfalls of Making a Categorical Diagnosis of Post Traumatic Stress Disorder in Personal Injury Litigation (1994) 34 Med Sci Law 117. Contrast our earlier suggestion in the Consultation Paper that all legitimate diagnoses of psychiatric conditions must meet the diagnostic criteria which are contained in the current versions of one or other of the two main diagnostic classificatory systems: Consultation Paper No 137, para 4.6. The responses we received from medical consultees have caused us to revise this view. DSM-IV, Introduction, p xxiii. See DSM-IV, Introduction, p xxiii: It must be noted that DSM-IV reflects a consensus about the classification and diagnosis of mental disorders derived at the time of its initial publication. New knowledge generated by research or clinical experience will undoubtedly lead to an increased understanding of the disorders included in DSM-IV, to the identification of new disorders, and to the removal of some disorders in future classifications. The text and criteria sets included in DSM-IV will require reconsideration in light of evolving new information.
58
6 reaction includes most reactions that would generally be considered normal responses to highly stressful events.7 Yet as we have seen,8 the law does not compensate mere grief or distress, and the phrase recognisable psychiatric illness has been adopted by the courts in order to distinguish compensatable suffering from these lesser reactions. We would anticipate, therefore, that faced with a diagnosis of acute stress reaction, the court would still require the plaintiff to establish that he or she was suffering more than mere mental distress.9
5.4
An alternative approach would be to set down in legislation that only certain specified illnesses should qualify for compensation. But this would appear to be unjust and to add a layer of complexity that is not required. Experts will frequently be in agreement that the plaintiff is suffering from a recognisable psychiatric illness, but disagree as to the most accurate diagnosis.10 Even providing a non-exhaustive list of psychiatric illnesses for which compensation may be claimed would be fraught with difficulty and could be misleading. The two diagnostic manuals use different terminology to refer to the same or similar symptoms, and any such list would require frequent updating if it were to reflect contemporary medical understanding. Such comment as there has been in the case law has recognised the impracticalities inherent in attempting a tight definition of particular psychiatric illnesses or of psychiatric illness in general.11
ICD-10, F 43.0, p 146. This diagnosis focuses on the development of anxiety, dissociative symptoms (such as numbing, detachment, disorientation, loss of awareness of identity) and other symptoms shortly after exposure to an extreme traumatic stressor. The diagnosis requires that symptoms should occur within 1 hour of exposure and diminish within 8 hours if the stressor is transient (within 48 hours if the stressor continues) and are usually minimal after about 3 days. DSM-IV includes a diagnosis, Acute Stress Disorder which requires an abnormal state defined as lasting for at least two days and up to four weeks: DSM-IV, para 308.3, p 429. This would exclude some conditions that would be included as acute stress reactions in ICD-10. M Gelder, D Gath, R Mayou and P Cowen, Oxford Textbook of Psychiatry (3rd ed 1996) p 137. See also L S OBrien, The Validity of the Diagnosis of Post Traumatic Stress Disorder [1994] JPIL 257, 273: These new disorders [acute stress reaction and acute stress disorder] would appear to provide a loop-hole which increases the number of victims who suffer psychiatric injury or nervous shock. Personally I have grave reservations about the significance and validity of these diagnoses which generally seem to represent what is a normal reaction to an adverse event and should be considered in the same category as normal rather than pathological grief. See para 2.3 above. For an exception, see para 2.49 n 127. But see M Napier and K Wheat, Recovering Damages for Psychiatric Injury (1995) pp 41-42. On consultation, the Association of Personal Injury Lawyers advocated a more radical approach, proposing that compensation should be available for mental distress. They argued that since the diagnoses, acute stress disorder and acute stress reaction, should already be regarded as within the current parameters of recovery, such a proposal would not dramatically extend the present position. L F Sparr and J K Boehnlein, Posttraumatic Stress Disorder in Tort Actions: Forensic Minefield Bull Am Acad Psychiatry Law 1990;18:283-302: [P]eople do not fit neatly into diagnostic categories. ... The battle of the categories is the direct result of the complexities of psychiatric diagnosis. In Vernon v Bosley (No 1) [1997] 1 All ER 577, 610 Thorpe LJ warned the courts against placing too heavy reliance on diagnostic classificatory systems. He said: DSM-III-R may provide the medical profession with a useful diagnostic tool but PTSD and its DSM-III-R classification should not ... be adopted in personal injury litigation as the yardstick by which
10
11
59
5.5
We have considered whether our proposed legislation should refer simply to a psychiatric illness rather than using the phrase recognisable psychiatric illness. However, we have concluded that since recognisable psychiatric illness has become a term of art with which the courts have become familiar and which seems to be operating satisfactorily, we should retain it. We do not want to risk giving the impression of suggesting that the concept of a psychiatric illness should in any way be diluted. Moreover, including the word recognisable may enable the courts to maintain more easily the traditional line between compensatable 12 psychiatric illness and non-compensatable mere mental distress. We therefore recommend that: (2) while the legislation should refer to a recognisable psychiatric illness - that being the familiar judicial terminology to denote more than mere mental distress - a definition of what is a recognisable psychiatric illness should not be laid down in legislation. (Draft Bill, clause 1(2), 2(2), 4(a) and 5(1)) 2. REASONABLE FORESEEABILITY OF PSYCHIATRIC ILLNESS AND THE TEST OF REASONABLE FORTITUDE
5.6
5.7
Legislative provisions enacted in three Australian jurisdictions provide that certain close relatives of a person killed, injured or put in peril may recover damages in respect of psychiatric illness sustained as a result, and that other family members may recover if the accident occurs within their sight or hearing.13 There is no additional requirement that the plaintiff should establish that the defendant ought reasonably to have foreseen the possibility that he or she might suffer psychiatric illness.14 We do not think that the same approach should be adopted in our jurisdiction. We consider that, at least where the plaintiff is outside the area of reasonably foreseeable physical injury,15 the plaintiff should, as at common law, be under an obligation to show that his or her psychiatric illness was a reasonably foreseeable consequence of the defendants conduct. We consider that reasonable
5.8
the plaintiffs success or failure is to be measured . Similarly, in the arbitration that followed the Zeebrugge disaster, the arbitrators declined to make a finding that the Diagnostic and Statistical Manual of Mental Disorders contains a suitable guide to the diagnosis of PTSD. They said that PTSD was a very recent concept, and that further research and experience may necessitate the revision of its diagnosis. In very general terms the manual could be used as a useful guide to diagnosis, but should not be treated as a statute: A & Others v P & O European Ferries (Dover) Ltd, February 1989.
12
See para 5.3 above. Law Reform (Miscellaneous Provisions) Act 1944 (NSW), s 4; Law Reform (Miscellaneous Provisions) Act 1955 (ACT), s 24; Law Reform (Miscellaneous Provisions) Act 1956 (NT), s 25. These legislative provisions are considered in more detail in para 6.14 below. A similar approach is taken to the recovery of economic loss and bereavement damages under the Fatal Accidents Act 1976 although, in contrast to the Australian statutes, the dependants under the 1976 Act must first show that the deceased had an actionable claim. Page v Smith [1996] AC 155 laid down that where the plaintiff is within the range of reasonably foreseeable physical injury, foreseeability of physical injury, as opposed to foreseeability of psychiatric illness, is sufficient. See paras 2.10 to 2.11 above and 5.11 below.
13
14
15
60
foreseeability of harm to the plaintiff is fundamental to all negligence claims and should only be rejected if there are compelling policy reasons for doing so. There would not appear to be any here.16 5.9 In the Consultation Paper we stressed that in making our provisional proposals we had assumed that it would still be necessary to establish that a psychiatric illness to the plaintiff was foreseeable, that is that the defendant owed the plaintiff a duty of care. We asked consultees whether they supported the different approach taken in the Australian statutes of allowing plaintiffs within certain degrees of relationship to claim without having to establish that he or she is owed a duty of care.17 Seventy-four per cent of consultees who considered this issue were opposed to the Australian approach. We therefore recommend that: (3) under our proposed legislation it should be a requirement for liability for psychiatric illness that a duty of care be owed to the plaintiff by the defendant; and that in establishing that duty of care it should be a requirement that, at least where the plaintiff is outside the area of reasonably foreseeable physical injury, it was reasonably foreseeable that the plaintiff might suffer psychiatric illness. (Draft Bill, clause 1(2) and 2(2)) The above recommendation leaves open the question whether foreseeability of psychiatric illness should be required where physical injury to the plaintiff was reasonably foreseeable. This was the issue in Page v Smith.18 As explained above,19 this decision marked a radical departure from the previously understood position.20 The House of Lords held (by a 3-2 majority) that where the plaintiff was within the range of reasonably foreseeable physical injury, the test of reasonable foreseeability of psychiatric illness was not relevant. In relation to such a plaintiff, physical injury and psychiatric illness are both to be treated as forms of personal injury and reasonable foreseeability of either is sufficient to found a duty of care.21 Many have welcomed this decision. Although the decision was given after the publication of the Consultation Paper, so that we did not have an opportunity to ask consultees for their views, several did comment in any event. On the whole, the responses, especially from practitioners, were very favourable to the decision.22
5.10
5.11
5.12
16
See also para 6.25 below. Consultation Paper No 137, para 5.28. [1996] AC 155. See para 2.10. The judgment has been described as the new orthodoxy which has torn apart the fabric of psychiatric damage law and stitched it together again so differently as to be almost unrecognisable: P R Handford, A New Chapter in the Foresight Saga: Psychiatric Damage in the House of Lords (1996) 4 Tort L Rev 5, 6. Page v Smith [1996] AC 155, 190. See paras 2.13 to 2.14 above. Both solicitors and barristers commented favourably on Lord Lloyds approach. For example, G McCool of Leigh, Day & Co: Page v Smith has obviously helped the situation and has been a most welcome development; Davies Arnold Cooper: Lord Lloyds approach to this area seems admirable both in terms of its clarity and logic; A Hamilton
17
18
19
20
21
22
61
They suggested that it rendered the law simpler and more certain. They agreed with Lord Browne-Wilkinson, who pointed out the dangers of the court seeking to draw hard and fast lines between physical illness and its causes on the one hand and psychiatric illness and its causes on the other since recent developments in medical science suggest a much closer relationship between physical and mental processes than had previously been thought;23 and with Lord Lloyd who said that, [n]othing will be gained by treating [physical and psychiatric injury] as different kinds of personal injury, so as to require the application of different tests in law.24 5.13 Also welcomed was the fact that liability for psychiatric illness would no longer turn on the fortuitous absence of a physical injury. Had Mr Page suffered any form of physical injury in the car accident, Lord Lloyd said that: [N]obody would have stopped to consider the foreseeability of nervous shock.25 Provided he could prove that the consequential psychiatric illness was genuine and was caused by the accident, Mr Page would have been entitled to recover. However, support for the case has not been unanimous. Many commentators have criticised its reasoning,26 and several of the consultees (especially the academic consultees) to our Consultation Paper did not favour its approach. First, whilst applauding the implicit endorsement which the House of Lords gave to the legitimacy of claims for psychiatric illness, some have questioned whether psychiatric illness should, for legal purposes, be treated as the same kind of injury as a physical injury, so that foreseeability of one will suffice to found a duty of care for the other.27 To allow the plaintiff to recover for psychiatric illness simply because some physical injury to him or her was foreseeable was said to be unfair and arbitrary.28 Secondly, critics point out that the Page decision does not remove the need to draw a distinction between physical injury and psychiatric
5.14
QC and A Schaff: We adopt what Lord Lloyd said ... both as to what the law is and what the law should be; D Kemp QC: I think that the present position as summarised by Lord Lloyd ... is sensible and satisfactory; and the Law Societys Civil Litigation Committee: [W]e think that the decision of the House of Lords in Page v Smith is particularly apt.
23
Page v Smith [1996] AC 155, 182. Page v Smith [1996] AC 155, 188. Page v Smith [1996] AC 155, 187. But see [1996] AC 155, 172, per Lord Jauncey. See para 1.8 above. T K Feng, Nervous Shock to Primary Victims [1995] SJLS 649; C A Hopkins, A New Twist to Nervous Shock [1995] CLJ 491; N J Mullany, Psychiatric damage in the House of Lords - Fourth time unlucky: Page v Smith (1995) 3 Journal of Law and Medicine 112; A Sprince, Page v Smith - being primary colours House of Lords judgment (1995) 11 PN 124; P R Handford, A New Chapter in the Foresight Saga: Psychiatric Damage in the House of Lords (1996) 4 Tort L Rev 5; F McManus, Nervous Shock - Back to Square One? [1996] Jur Rev 159; F A Trindade, Nervous Shock and Negligent Conduct (1996) 112 LQR 22. But see B McDonald and J Swanton, Foreseeability in relation to negligent infliction of nervous shock (1995) 69 ALJ 945. A Sprince, Page v Smith - being primary colours House of Lords judgment (1995) 11 PN 124, 126 and T K Feng, Nervous Shock to Primary Victims [1995] SJLS 649, 650652. F McManus, Nervous Shock - Back to Square One? [1996] Jur Rev 159, 161.
24
25
26
27
28
62
29 illness. The judgment makes a fundamental distinction between primary and secondary victims, certain control mechanisms applying to secondary victims only. This distinction is based on whether the victim is within the range of foreseeable physical injury (as opposed to psychiatric illness). Therefore, while it may be irrelevant to liability whether a primary victim suffers physical injury or psychiatric illness, the line between foreseeability of physical injury and psychiatric illness is still afforded critical importance because it determines who is a primary victim. Thirdly, several commentators suggest that the decision may give rise to claims by a large number of plaintiffs and that this will risk opening the floodgates of litigation.30
5.15
Whilst we appreciate the concerns raised by those who criticise the Page decision, we do not consider that we should make any proposals for legislative reform in relation to a plaintiff who comes within the range of foreseeable physical injury. There appears to be no strong support among legal practitioners or judges (as opposed to academic lawyers)31 for Page v Smith to be reversed. Nor do we consider that sufficient time has passed to assess the impact that the decision will have. Certainly, it seems likely to enable a larger number of plaintiffs who have not suffered physical injury to claim for psychiatric illness, but it should be appreciated that in those cases where the plaintiff suffers psychiatric illness which is the reasonably foreseeable result of a fear for his or her own safety, recovery would in any event be available under Dulieu v White & Sons.32 The most obvious category of plaintiff to benefit from the Page decision will therefore be the susceptible plaintiff who suffers psychiatric illness as a result of a minor accident in which he or she is not physically harmed and which the defendant could not reasonably have foreseen would cause psychiatric illness. Lord Jauncey, in his dissenting speech in Page, gave the example of a hysterical woman who suffers psychiatric illness when her stationary car is bumped by the defendant while parking.33 But it
29
See, for example, M A Jones, Liability for Psychiatric Illness - More Principle, Less Subtlety? [1995] 4 Web JCLI. In Nervous Shock and Negligent Conduct (1996) 112 LQR 22, F A Trindade gives several examples where the number of potential primary victims may be large: What of the passengers sitting on a bus with which a negligent motorist collides? Is it only the passengers who are in close proximity to the part of the bus where the impact occurs, or every passenger on the bus who could be said to be within the range of foreseeable physical injury? What of a passenger train which is derailed by the negligence of the engine driver? Is it only the passengers sitting in the derailed carriage or all the passengers on the train who are within the range of foreseeable physical injury? And what of the situation of a disabled aircraft which flies over a city and then crashes into a residential building? And see, T K Feng, Nervous Shock to Primary Victims [1995] SJLS 649, 651; N J Mullany, Psychiatric damage in the House of Lords - Fourth time unlucky: Page v Smith (1995) 3 Journal of Law and Medicine 112; A Sprince, Page v Smith - being primary colours House of Lords judgment (1995) 11 PN 124, 128. See para 5.14 n 26. [1901] 2 KB 669. See para 2.4 above. In Page v Smith [1996] AC 155 both Lord Ackner and Lord Lloyd said that if the relevant test had been foreseeability of psychiatric illness, then it was reasonably foreseeable that a person would suffer psychiatric illness as a result of the collision that had occurred in that case: [1996] AC 155, 170, per Lord Ackner; 197, per Lord Lloyd. Page v Smith [1996] AC 155, 178. Lord Lloyd, commenting on this example, said that it would not be reasonably foreseeable that the woman would suffer physical injury: [1996] AC 155, 189-190. This highlights a further difficulty with the Page decision - assessing
30
31
32
33
63
should be noted that if recommendation (6) set out below,34 for a more flexible approach to the test of reasonable foreseeability of psychiatric illness, were to be adopted by the courts, then even the number of plaintiffs who fall within this category would be reduced. 5.16 For these reasons, we recommend that: (4) our proposed legislation should not overturn the rule laid down in Page v Smith that reasonable foreseeability of psychiatric illness is not required where physical injury to the plaintiff was reasonably foreseeable. In cases outside Page v Smith, where the test of reasonable foreseeability of psychiatric illness is applied, there are two unresolved issues in relation to the application of the test. Both are rather technical and can be solved by the courts without legislation (that is, our preferred solutions are within the interpretative reach of the courts). The first relates to the use of hindsight. Where the plaintiff suffers psychiatric illness as a result of the defendant causing the injury or imperilment of another, the courts have said that hindsight should be used in applying the test of reasonable foreseeability of psychiatric illness. As a result, the foreseeability of the illness suffered by the plaintiff is assessed in the light of all that has happened as a result of the defendants negligence.35 This is in contrast to cases of physical injury where hindsight is irrelevant.36 What the courts seem to have in mind is that, where the psychiatric illness is consequent on fear for anothers safety, one should assess whether the psychiatric illness is reasonably foreseeable on the assumption that the defendant knows what has happened to the immediate victim. We agree with the courts approach to the extent that a defendant should not be held liable for a psychiatric illness that is only foreseeable as a consequence of an accident to an immediate victim, if the accident to the immediate victim, albeit foreseeable, does not actually occur. So, for example, where a mother suffers psychiatric illness as a result of thinking about a potential accident which might have injured her son, but which in fact was avoided, the courts should assess the foreseeability of her illness on the basis that she is aware that the accident did not actually happen.37 On the other hand, we disagree with the courts approach to the extent that a defendant might be held liable for a psychiatric illness that was only foreseeable on
5.17
5.18
5.19
whether the plaintiff came within the area of reasonably foreseeable physical injury, when he or she has not in fact suffered any such physical harm. See R Colbey, Nervous Shock: The Law is Clarified and ME Enters the Fold (1996) 15 Lit 189, 192-193 who suggests that Lord Lloyds analysis of Lord Jaunceys example is flawed.
34
See para 5.27. See para 2.8 above. Page v Smith [1996] AC 155, 188-189, per Lord Lloyd. But see Lord Wright in Bourhill v Young [1943] AC 92, 110 and Lord Wilberforce in McLoughlin v OBrian [1983] 1 AC 410, 420 both of whom tend to suggest that foreseeability in negligence is always assessed with the benefit of hindsight. See also para 6.18 below.
35
36
37
64
the assumption that the defendant already had in mind an (otherwise unforeseeable) accident to the immediate victim. So for example, if a mother suffers psychiatric illness as a result of an injury to her son caused by the defendant in an accident that the defendant could not reasonably have foreseen, the courts should recognise that similarly the defendant could not reasonably have foreseen the consequential illness of the mother. 5.20 Accordingly, we recommend that: (5) although we do not think that legislation on the point is appropriate, we tend to the view that, where the plaintiff suffers psychiatric illness as a result of the death, injury or imperilment of another, the reasonable foreseeability of the plaintiff s psychiatric illness should not always be judged with hindsight. In particular, in assessing whether the psychiatric illness was a reasonably foreseeable consequence of the defendants conduct, the court should consider whether the harm or imperilment to the immediate victim (that is the accident) was, judged prior to the accident, reasonably foreseeable. The second and deceptively difficult issue concerns the reasonable fortitude test. 38 As explained above, in assessing whether psychiatric illness is reasonably foreseeable the defendant, unless he or she has special knowledge to the contrary, may assume that the plaintiff is a person of customary phlegm and of a normal standard of susceptibility.39 This is intended to exclude from compensation those who are abnormally susceptible to psychiatric illness. However, it is not clear whether the emphasis which the courts have placed on the plaintiff s fortitude is intended to be no more than an application in the realm of psychiatric illness of the standard approach to reasonable foreseeability in personal injury cases,40 or whether it represents a special restriction which applies to claims for psychiatric illness. The most detailed analysis of this issue was given by Lord Wright in Bourhill v Young41 where the plaintiff was considered to be of peculiar susceptibility, being eight months pregnant. Lord Wright seemed to regard the reasonable fortitude test as merely an application of the foreseeability test as it applies in cases of physical injury. He said: One who suffers from the terrible tendency to bleed on slight contact, which is denoted by the term a bleeder, cannot complain if he mixes with the crowd and suffers severely, perhaps fatally, from being merely
5.21
5.22
38
See para 2.7. See, for example, Bourhill v Young [1943] AC 92, 110, per Lord Wright; McLoughlin v OBrian [1983] 1 AC 410, 429, per Lord Russell of Killowen. It is clear that in cases of physical injury, the defendant is expected to foresee that the plaintiff might be particularly susceptible to certain injury. For example, in Haley v London Electricity Board [1965] AC 778 a blind plaintiff recovered damages in negligence when he tripped over a long handled hammer left by the defendants to fence a trench which they had dug in the pavement. The hammer was not a danger to sighted people. However, given the number of blind people, it was held that the defendants should have foreseen that a blind person might walk by and should therefore have taken different measures to fence the trench. [1943] AC 92.
39
40
41
65
brushed against.42 According to Lord Wright, the same principle applies in cases of nervous shock, where liability must depend on a normal standard of susceptibility and [w]hat danger of particular infirmity that would include must depend on all the circumstances.43 5.23 Yet in Page v Smith, Lord Lloyd said that the requirement that the plaintiff prove that the psychiatric illness was foreseeable in a person of normal fortitude was a special control mechanism which the courts adopted, as a matter of policy, where the plaintiff was a secondary victim in order to limit the potential number of claims.44 This tends to suggest that he regarded the requirement as a special limiting factor over and above an application of the usual foreseeability test. We sought help from consultees on this difficult question of law.45 Several consultees responded before the House of Lords had given its decision in Page,46 so that they were unable to take account of Lord Lloyds speech in their comments. However, of those who considered this issue, 24 per cent thought that the emphasis that the courts have placed on the plaintiff s reasonable fortitude was a special restriction to claims for psychiatric illness, while 70 per cent regarded it as simply part of the standard approach to reasonable foreseeability adopted in personal injury cases. We also asked consultees whether, in applying the reasonable foreseeability test, the law should continue to assume that the plaintiff is a person of reasonable fortitude or, on the contrary, whether they considered it unsatisfactory to distinguish between reasonable fortitude and abnormal sensitivity.47 Fifty-five per cent of consultees who responded to this question thought that there should continue to be an assumption of reasonable fortitude, while 45 per cent thought it unsatisfactory to distinguish reasonable fortitude from abnormal sensitivity.48
5.24
5.25
42
Ibid, 109. Ibid, 110. [1996] AC 155, 197. In the Consultation Paper, we asked consultees whether they regarded the emphasis which the courts have placed on the plaintiffs reasonable fortitude as a restriction special to claims for psychiatric illness or, on the contrary, as no more than an application in the realm of psychiatric illness of the standard approach to reasonable foreseeability in personal injury cases: Consultation Paper No 137, para 5.13. [1996] AC 155. Consultation Paper No 137, para 5.13. Several consultees with experience in psychiatric and psychological matters provided helpful responses to the question whether it is satisfactory for the law to distinguish between reasonable fortitude and abnormal sensitivity. M Mandelson said that the population varies considerably in terms of susceptibility to developing psychological disorder in response to a traumatic experience. It does so on two counts: a persons predisposition to developing adverse psychological reactions by virtue of his or her personality or personal circumstances; and a persons pre-existing psychiatric illness which is exacerbated by the effects of the traumatic episode. A significant minority of the population at large (up to 20%) will be covered by one or both counts so that it is reasonable to foresee that certain members of the population will be particularly susceptible. The British Psychological Society said that the body of psychological knowledge does not suggest a reason why the approach in psychiatric illness cases should be different to that used in physical injury cases. The Society said that it is difficult in psychological terms to define or measure a concept
43
44
45
46
47
48
66
However, these percentages do not provide a clear picture of consultees views, since some of those who thought that the assumption of reasonable fortitude should be retained did not regard this as anything over and above an application of the standard approach to reasonable foreseeability in personal injury cases, whilst others thought that it was a special restriction. Several responses also appeared to confuse issues of liability and quantum. Yet it has never been doubted - and this was not intended to be raised by our question - that the normal eggshell skull rule of remoteness applies in cases of psychiatric illness, so that once a duty of care has been established the susceptible plaintiff may recover for the full extent of his 49 or her illness. 5.26 In applying the test of foreseeability of psychiatric illness, the court must clearly adopt some objective standard, without which the test would have no content.50 However, with respect to Lord Lloyd, we suggest that to use reasonable fortitude as a means of restricting the potential number of claims for psychiatric illness, would be to adopt a blunt and arbitrary control device. As Lord Ackner commented in Page, normal fortitude is an imprecise phrase.51 Any attempt to fix upon criteria of what constituted such fortitude would be very difficult. We therefore think that allowing the defendant to assume that the plaintiff is a person of customary phlegm is best interpreted as meaning nothing more than that, in deciding whether psychiatric illness was reasonably foreseeable (and analogously to reasonable foreseeability in physical injury cases), one can take into account the robustness of the population at large to psychiatric illness. We therefore recommend that: (6) although we do not think that legislation on the point is appropriate, while, in applying the test of reasonable foreseeability of psychiatric illness, it may be helpful to continue to assume that the plaintiff is a person of reasonable fortitude, that assumption should be regarded as merely an aspect of the standard approach to reasonable foreseeability that is applied in cases of physical injury.
5.27
such as general resilience to stressors or its obverse vulnerability. A person may be resilient to one type of stressor but not to another type of stressor, and the impact of any particular stressor will vary between people. Similarly, the Royal College of Psychiatrists Mental Health Law Group said that in practice it can be very difficult for psychiatrists to make an either-or distinction between reasonable fortitude and abnormal sensitivity. The issue is essentially a judgment on moral fibre rather than a clinical diagnosis. People cope with life events differently, and the same person may cope with the same life event differently on different occasions. The Group doubts whether there is any such thing as reasonable fortitude or abnormal sensitivity insofar as these are constantly exhibited personal characteristics.
49
Although the quantum of damages may be reduced to take into account the fact that the plaintiff might have suffered the illness in any event at some point in the future. See, for example, Brice v Brown [1984] 1 All ER 997. But see T Weir, Tort Liability for Psychiatric Damage: The Law of Nervous Shock [1993] CLJ 520, 521 where the prudence of applying the eggshell skull rule to cases of psychiatric illness is doubted: Vulnerability to physical lesion is pretty standard throughout the population ... but the range of psychic liability is very great indeed. As one consultee commented: In effect the test would be dissolved. Page v Smith [1996] AC 155, 170.
50
51
67
3. THE SHOCK REQUIREMENT 5.28 The requirement that, to be compensatable, the plaintiff s psychiatric illness must have been induced by a shock has been widely criticised.52 One commentator describes it as arguably the most anomalous of all the current restrictions.53 We set out in paragraph 5.29 below the main arguments put forward for the abandonment of the shock requirement. We recognise, however, that a number of arguments may be made in support of the retention of the shock test, or at least its replacement by some other similar, but perhaps less stringent, requirement.54 These arguments are set out in paragraph 5.30 below. (1) Arguments for Abandoning the Shock Requirement 5.29 We regard the following arguments to be the central ones in favour of abandoning the shock requirement: (1) The shock test produces harsh decisions, and if its sole purpose is to limit the number of potential defendants, it is a very crude method of doing so.55 The consultees with medical experience emphasised the difficulties that the shock test causes from a medical point of view. The Royal College of Psychiatrists Mental Health Law Group wrote: For psychiatrists the shock-induced requirement causes serious problems. The term is vague, has no psychiatric meaning and is emotively misleading. The requirement should be abandoned. Psychiatric evidence should require demonstration, or not, of a psychiatric disorder distinct from a normal mental reaction and, if present, its relationship with the index event. This is usually possible. The requirement to fit the evidence around the concept of whether or not
(2)
52
See, for example, F A Trindade, The Principles Governing the Recovery of Damages for Negligently Caused Nervous Shock [1986] CLJ 476, 478-480; K J Nasir, Nervous Shock and Alcock: The Judicial Buck Sops Here (1992) 55 MLR 705, 709; H Teff, Liability for Psychiatric Illness after Hillsborough [1992] OJLS 440, 447-451 and The Requirement of Sudden Shock in Liability for Negligently Inflicted Psychiatric Damage (1996) 4 Tort L Rev 44, 46-47; P R Handford, Compensation for Psychiatric Injury: The Limits of Liability (1995) 2 Psychiatry, Psychology and Law 37, 44-45; C J Lewis, Nervous Shock (1995) 139 SJ 960; N J Mullany and P R Handford, Tort Liability for Psychiatric Damage (1993) ch 8 and Hillsborough Replayed (1997) 113 LQR 410, 411-412; M A Jones, Textbook on Torts (5th ed 1996) pp 138-139 and Liability for Psychiatric Illness - More Principle, Less Subtlety? [1995] 4 Web JCLI. M A Jones, Liability for Psychiatric Illness - More Principle, Less Subtlety? [1995] 4 Web JCLI. Several consultees suggested alternative tests in an attempt to create a middle ground between retaining the shock test and abandoning it altogether. These included (i) that the illness must have been a consequence of a horrific experience; (ii) that the illness was induced by shock, sudden or prolonged, suffered at the scene of the primary victims accident, in its immediate vicinity or aftermath; (iii) that the illness results from the impact of matters which themselves formed part of the accident and its aftermath but excluding harm which has resulted from contact with more remote consequences such as the subsequent effect of the accident upon an injured person; and (iv) that the illness arises from a distressing event (or series of events). See paras 2.61 to 2.65 above.
53
54
55
68
the disorder is shock-induced has no scientific or clinical merit. 56 simply playing with words. (3)
It is
The shock requirement renders some forms of psychiatric illness, such as post-traumatic stress disorder, more readily compensatable than other psychiatric illnesses, such as depression. Yet there is no suggestion that the suffering involved in these other illnesses is not as great as in shock-induced illnesses, or that they may not be causally connected with the defendants negligence. In some cases the full extent of the injury to the immediate victim will only become apparent over a period of time. Rather than suddenly becoming aware of the full horror, the plaintiff learns of the severity of the injuries over a number of days, weeks, months or even years. Although the courts seem to be tending towards relaxing the degree of suddenness required,57 the fixing of any cut-off point for suddenness will be essentially arbitrary.58 It is clear that the test is no longer required in all cases where damages may be recovered for psychiatric illness. For example, the social worker suffering from stress in Walker v Northumberland County Council59 was not required to show that his breakdown was caused by a shock. There appear to be two elements to the shock requirement: both the sudden appreciation by sight or sound of a horrifying event and that this violently agitates the mind.60 The retention of the first of these elements would therefore mean preserving the requirements of proximity in time and space and direct perception, where the plaintiff suffers psychiatric illness as a result of the injury or imperilment of a loved one. Yet, as we explain in Part VI below,61 these requirements should, in our view, be abandoned.62
(4)
(5)
(6)
56
M Mandelson, a clinical psychologist, wrote: Psychological reactions to a traumatic episode do develop over time, partly as a result of the subjects ability to reflect on past events and anticipate future events and to consider their implications (processes which themselves take place over time), and partly because the traumatic episode may be protracted before the final outcome is known; and Professor Wessely pointed out that a sudden event is not the only accepted precipitant of psychiatric disorder, nor is it the most common. See para 2.65 above. See K J Nasir, Nervous Shock and Alcock: The Judicial Buck Stops Here (1992) 55 MLR 705, 709: The fact that one claimants experience is less protracted than anothers does not mean that the latter has suffered any the less - on the contrary, on an abstract analysis he or she will usually have suffered more. [1995] 1 All ER 737. See para 2.47 above. Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 401, per Lord Ackner. According to Brennan J in Jaensch v Coffey (1984) 155 CLR 549, 567 shock means the sudden sensory perception - that is, by seeing, hearing or touching - of a person, thing or event, which is so distressing that the perception of the phenomenon affronts or insults the plaintiffs mind and causes a recognisable psychiatric illness. See paras 6.10 to 6.16.
57
58
59
60
61
69
(7)
The requirement excludes those whom society may feel most worthy of legal support, such as the person worn down by the long-term caring of a relative injured by the defendant. We asked consultees whether the requirement that the psychiatric illness 63 must have been shock-induced should be abandoned. Sixty-six per cent of consultees who responded to this question thought that it should.64
(8)
(2) Arguments for Retaining the Shock Requirement 5.30 We regard the following arguments to be the central ones for retaining the shock requirement: (1) Abandoning the shock requirement would increase the potential number of claims, thereby causing concern that the floodgates of litigation would be opened. The difficulty, for example, of drawing the line between normal grief and a pathological grief reaction might result in claims for psychiatric illness being added to every claim for damages brought under the Fatal Accidents Act 1976. The shock test facilitates proof of causation. Without such a test, the court would be required to decide whether the plaintiff s illness, possibly occurring many months or even years after the defendants allegedly negligent act or omission, was in fact caused by that act or omission or resulted from some other intervening stressor,65 such as another bereavement, occurring during that period of time. Research has shown,
(2)
62
But see Reeve v Brisbane City Council [1995] 2 Qd R 661 where the Queensland Supreme Court treated the communication to the plaintiff of her husbands unexpected death as the relevant event. Consultation Paper No 137, para 5.40. We asked this question in the context of liability to a person who suffered psychiatric illness as a result of the death, injury or imperilment of another caused by the defendant. However, we consider that it is safe to assume that, if and to the extent that the shock test remains relevant to cases where the plaintiff suffers psychiatric illness as a result of a fear for his or her own safety or as a result of his or her direct involvement in an accident caused by the defendant (for example, as a rescuer), these consultees would support the abandonment of the test. Later in the Consultation Paper, in considering liability for psychiatric illness suffered pursuant to damage to anothers property, we asked whether consultees would be in favour of retaining all three proximity elements and also the requirement that the psychiatric illness must have been shock-induced: Consultation Paper No 137, para 5.56. 88% of consultees who responded to this question said that they would be in favour of retaining all these requirements. This would appear to be contrary to the views given by those consultees who favoured abandoning the shock requirement where the defendant has caused physical injury to another. However, we think that it is likely that in giving their responses to para 5.56 the consultees were endorsing our provisional view that where damage to anothers property is concerned, criteria should be adopted which are analogous to, but no less restrictive than, those adopted where human safety or injury to another is concerned: Consultation Paper No 137, para 5.56. It therefore seems reasonable to suppose that if the shock requirement were abandoned in the cases involving human safety or injury, many of those consultees would no longer consider that it should apply in cases which involve damage or danger to anothers property. See para 3.5 above.
63
64
65
70
for example, that exposure to multiple traumatic events increases the risk of PTSD.66 (3) The requirement maintains a level of immediacy with the primary event, thereby protecting the defendant from the possibility of claims arising years after his or her negligent conduct. The law relating to liability for negligently inflicted psychiatric illness in several other jurisdictions, including Australia and Canada, contains a shock requirement.67 The New South Wales Court of Appeal recently held that the common law shock requirement remains applicable to a claim under the New South Wales statutory provisions relating to nervous shock.68 Damages are clearly not available at law for every misfortune that may occur to a person as a result of anothers negligence. Some might suggest that psychiatric illness brought on by the long-term caring of an injured relative is one of the unfortunate vicissitudes of life for which the law ought not to provide compensation.69
(4)
(5)
5.31
While we recognise the force of the arguments for retaining the shock test, we have come to the conclusion that it should be abandoned. We believe that it is possible to counter all the arguments raised in favour of retaining the shock requirement. First, abandoning the requirement would not open the floodgates in its central meaning of a proliferation of claims arising out of a single event. The risk of a flood of claims is felt most strongly in the case of plaintiffs who suffer psychiatric illness as a result of the injury or imperilment of another. But provided the general requirement of a close tie of love and affection between the plaintiff and the immediate victim is maintained, the potential number of claims would be sensibly controlled. Secondly, we do not think that the courts will be unable to cope with the difficult issues of causation that might arise. The courts currently deal with similar problems which arise in relation to physical injury cases where a variety of factors may have contributed to an injury or disease.70 It is also worth noting that
66
B L Green, Psychosocial Research in Traumatic Stress: An Update J Traumatic Stress 1994;7:341-362; S Wessely, Liability for Psychiatric Illness J Psychosom Res 1995;39:659-669; L S OBrien, The Validity of the Diagnosis of Post Traumatic Stress Disorder [1994] JPIL 257, 271: It is not uncommon for victims to develop PTSD after a second and subsequent traumatic event. When they do symptomatology often includes both events and the question of attribution of effect between the two is difficult. Consultation Paper No 137, Appendix paras 13 and 28. Chiaverini v Hockey [1992] Aust Torts Rep 81-223. For details of the legislative provisions see para 6.14 below. Tied in with this argument may be the feeling that damages for such psychiatric illness would simply be too remote from the defendants negligence: see M Lunneys review of Tort Liability for Psychiatric Damage [1993] Med L Rev 408, 411. However, there appears to be no reason to think that the general tort principles of remoteness would not be able to operate effectively here. For example, in Bonnington Castings Ltd v Wardlaw [1956] AC 613 the plaintiff contracted pneumoconiosis as a result of exposure to silica dust where he worked. Exposure to some dust was inevitable and there was no breach of duty by his employers in respect of it. However, they were in breach of their statutory duty in failing to provide effective dust
67
68
69
70
71
the shock requirement does not prevent difficult issues of causation arising in any event. For example, in Calascione v Dixon71 damages for pathological grief reaction were denied to a mother following the death of her son in a motor accident because her pathological grief reaction was found to be caused by subsequent events, such as the acquittal of the defendant on a charge of causing death by reckless driving. She did, on the other hand, recover damages for her posttraumatic stress disorder, which was found to have been caused by the shock of the accident.72 Thirdly, physical injuries may also reveal themselves some time after the negligent act.73 Provided that the claim is made within the relevant limitation period, there is no suggestion that the plaintiff should not be able to recover for such injuries. Fourthly, the requirement for a shock, as applied in some other jurisdictions, has been criticised.74 Finally, it is difficult to justify why negligently inflicted psychiatric illness which occurs over a number of years should be singled out as a vicissitude of life for which there is no compensation. 5.32 Although several consultees suggested that if the shock test is abandoned it might be replaced by some less stringent requirement,75 any alternative similar test seems as arbitrary as the shock requirement. Such a test might be based on a timing factor. However, this would require an arbitrary cut-off date which would not fit with the medical analysis of what causes psychiatric illness and would exclude
extraction plant for certain machinery. The medical evidence could prove that the pneumoconiosis was caused by the dust, but not the source from which the dust came. The House of Lords held that the employers were liable for the plaintiffs illness on the basis that, where there are two causes of the harm suffered, it was sufficient for the plaintiff to show that the defendants breach had materially contributed to the harm. In Page v Smith (No 2) [1996] 1 WLR 855 the Court of Appeal confirmed that the same test is used in establishing causation in relation to negligently inflicted psychiatric illness. Furthermore, where a plaintiff has suffered two or more stressors, only one of which was caused by the negligence of the defendant, while the plaintiff may succeed if it can be shown that the negligently caused incident was more than a trivial or insignificant cause of the psychiatric illness, the amount of damages may be reduced to take into account the fact that the plaintiff might at some point have suffered the illness in any event: Page v Smith (No 2) [1996] 1 WLR 855, 857.
71
(1993) 19 BMLR 97 (CA). Issues of causation were also considered in Vernon v Bosley (No 1) [1997] 1 All ER 577. The plaintiff suffered psychiatric illness after being called to the scene where his children had been killed in a road accident caused by the negligent driving of their nanny. The defendant admitted negligence but argued that the plaintiffs illness was caused partly as a result of the shock of attending at the accident and partly as a result of an abnormal grief reaction to the bereavement which he had suffered. The defendant sought to limit the damages to compensation for that part of the illness ascribed to the shocking experience rather than to the grief. By majority, the Court of Appeal rejected this contention and held that in accordance with general tort principles of causation, damages were recoverable for an illness caused or at least contributed to by the negligence of the defendant notwithstanding that the illness might also be regarded as a pathological consequence of the bereavement which the plaintiff, where the immediate victim was killed, must inevitably suffer. For comment on this decision see B Mahendra, Nothing but the whole tort (1996) 146 NLJ 1022. Examples include asbestosis, dermatitis or pneumoconiosis. H Teff, The Requirement of Sudden Shock in Liability for Negligently Inflicted Psychiatric Damage (1996) 4 Tort L Rev 44, 54. See para 5.28 n 54 above.
72
73
74
75
72
deserving cases. Alternatively, the test could be based on the severity of the incident, requiring, for example, a particularly horrific experience. We do not consider that such a test would be practical, since what is horrific to one person may not be to another. It also focuses unduly on the accident type of claim rather than, for example, death or injury cased by medical negligence. Finally, some consultees and commentators have suggested that the fear of opening up the door to too many claims by abandoning the shock test may be reduced by using a test 76 based on the severity of the psychiatric illness suffered. However, we consider that any attempt to draw a distinction between a slight and a severe psychiatric illness would be artificial, and that establishing acceptable criteria for the distinction would be extremely difficult.77 5.33 Accordingly, we recommend that: (7) our proposed legislation should ensure that it shall no longer be a condition of liability for a recognisable psychiatric illness that the psychiatric illness was induced by shock. (Draft Bill, clause 1(2), 2(2) and 5(2)) 4. RECOVERY WHERE THE IMMEDIATE VICTIM IS THE DEFENDANT 5.34 As explained above,78 dicta suggest that recovery may be barred where the plaintiff suffers psychiatric illness as a result of an injury to the defendant him or herself (that is, where the defendants injury is self-inflicted). So, for example, recovery may be denied where parents suffer psychiatric illness consequent upon their childs injury in a car accident caused by the childs careless driving. Likewise, recovery may be barred where the plaintiff s psychiatric illness is caused by a loved ones suicide. The most persuasive argument in favour of this bar is that the imposition of a legal duty to look after oneself, simply in order to protect others from the likely psychiatric effects on them in the event of failure, places an undesirably restrictive burden on ones self-determination. However, one might counter this with the conflicting view that persons who deliberately or negligently place themselves in danger should foresee the possibility of the consequences of their actions for others and take responsibility for them. The self-determination argument is not given weight where a person negligently or deliberately injures or endangers him or herself and thereby causes reasonably foreseeable physical injury to another. For example, in road accident cases it is fairly common for physically injured
5.35
76
Tests suggested by consultees were based either on the length of time during which the illness is suffered, so, for example ruling out all claims where the symptoms lasted less than three months, or on the extent to which the illness impaired the ability to lead a normal life. Some commentators have also suggested such a restriction: H Teff, Liability for Negligently Inflicted Nervous Shock (1983) 99 LQR 100, 105; K J Nasir, Nervous Shock and Alcock: The Judicial Buck Stops Here (1992) 55 MLR 705, 712. The Consultation Paper asked generally whether all claims for psychiatric illness should be limited by some severity threshold: Consultation Paper No 137, para 5.44. 87% of consultees who responded to this question thought that damages should continue to be recoverable irrespective of the severity of the illness. See para 2.66.
77
78
73
passengers to sue another member of their family whose negligent driving caused the accident. 5.36 In our Consultation Paper we asked consultees whether there should be a bar to the recovery of damages for psychiatric illness where the primary victim is the defendant.79 Seventy-six per cent of consultees who responded to this question thought that there should be no such bar. Those who gave reasons for their response generally did so on the basis that, since there is no such bar where the defendants self-inflicted injury results in the plaintiff s physical injury, nor should there be a bar where psychiatric illness is caused thereby. One advantage of removing this bar would be that it would resolve the difficulties otherwise produced where the immediate victims negligence has contributed to an accident with the defendant.80 In that situation, if a relative of the immediate victim were to bring a claim for damages for psychiatric illness against the defendant, it would seem unfair not to allow the defendant a right of contribution from the immediate victim. Yet, if the bar applies, such a right of contribution would not be available under the Civil Liability (Contribution) Act 1978 as the immediate victim would not be committing a tort against the plaintiff and would not therefore be liable for the same damage as the defendant. On the other hand, it may be thought harsh that the practical effect of allowing an action against the immediate victim may sometimes be that the (uninsured) immediate victims own damages will have to be used to pay for the relatives claim for psychiatric illness. For example, say that the immediate victim, a pedestrian, suffers injury partly by his own negligence and partly as a result of the negligence of a car driver, and has secured damages for the injury (reduced to reflect his or her contributory negligence). It may seem harsh that the immediate victim has to disgorge some or all of those damages if sued by a person with whom he or she has a close tie of love and affection and who has suffered psychiatric illness as a result of the accident. However, one might query the extent to which plaintiffs would bring such claims for psychiatric illness in practice, if their result would be to jeopardise the financial position of the immediate victim - with whom, after all, he or she has a close tie of love and affection. In practice, such claims are only likely to be brought if the financial position of the immediate victim and the relative, taken together, is enhanced. Moreover, by altering the facts of the example (including reversing the insurance position) one can plainly see the harshness of applying an immediate victim bar. Say that the immediate victim is a car driver who suffers injury in an accident partly caused by his negligent driving and partly caused by the negligence of a pedestrian, who is also injured. The
5.37
5.38
79
Consultation Paper No 137, para 5.51. Lord Oliver drew attention to this difficulty in Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 418, saying that: If, for instance, the primary victim is himself 75 per cent. responsible for the accident, it would be a curious and wholly unfair situation if the plaintiff were enabled to recover damages for his or her traumatic injury from the person responsible only in a minor degree whilst he in turn remained unable to recover any contribution from the person primarily responsible since the latters negligence vis--vis the plaintiff would not even have been tortious. However, Lord Oliver still suspected that an English court, if asked to consider the question, would impose a bar on recovery where the defendant was the immediate victim.
80
74
pedestrian receives compensation from the car driver, reduced to reflect his or her contributory negligence. If the loved one of the driver suffers psychiatric illness as a result of the drivers injury and claims from the uninsured pedestrian, it would seem unfair not to allow the pedestrian a contribution from the car driver to the extent that the accident was caused by the drivers negligence. 5.39 An alternative approach to solving the problem created where the immediate victim has contributed to an accident with the defendant, would be to reduce the plaintiff s damages in line with the contributory negligence of the immediate victim. But this is not attractive. It would be contrary to the underlying principle 81 that the defendant owes a separate duty of care directly to the plaintiff, and would mean that the plaintiff was unable to obtain full compensation for his or her psychiatric illness. We do not therefore regard this as a satisfactory alternative to the removal of the bar to recovery where the defendant is the immediate victim. However, the problem is more difficult to resolve if one moves away from standard examples of negligently caused self-injury (as in traffic or work accidents). Away from those standard examples, the self-determination argument can be seen to have a stronger role to play in cases where the defendants self-inflicted injury causes anothers psychiatric illness than it does where it results in anothers physical harm. We explain in Part VI that, where a plaintiff suffers psychiatric illness as a result of the injury or imperilment of another, the requirement that the plaintiff should be present at the scene of the accident or aftermath and have direct perception of it, should be abandoned.82 But if, in addition, one went on to remove the bar to recovery where the defendant is the immediate victim one might severely curtail a persons autonomy to carry out activities that are very dangerous to him or herself. Where a person is intent on undertaking some activity which carries a high risk of self-injury (for example, very dangerous sports), he or she can generally arrange to do so in such a way that would not put another person in physical danger. The duty of care not to cause physical injury to another does not therefore restrict a persons autonomy to carry out the very dangerous activity, but rather restricts the manner or place in which he or she may do it. However, where it is foreseeable that if self-harm occurs a person who has a close tie of love and affection may suffer psychiatric illness as a result, there would be no way of carrying out the dangerous activity at all, without the potential tortious liability. While it might be argued that the person carrying out the activity is simply not negligent (because a reasonable man might undertake such activities) it would seem preferable to clarify at the outset that no duty of care should be owed. That is, no duty of care should be imposed where to do so would unduly hinder a persons desire to carry out very dangerous activities which present no risk of physical injury to anyone other than him or herself, and would therefore unacceptably restrict a persons self-determination. Similarly, where the defendant has deliberately chosen to bring about his or her own injury or death, rather than simply running the risk of that injury or death by undertaking a very dangerous activity, it might not be appropriate in all cases to impose a duty of care. For example, a person might choose to refuse medical
5.40
5.41
81
82
75
treatment, perhaps on religious grounds, knowing that this will result in his or her own injury or death. 5.42 We think that in all the cases referred to in the two previous paragraphs there is a difficult balance to be arrived at between respecting self-determination and requiring proper concern to be shown for the consequences for others of choosing to harm or incur the risk of harm to oneself. This should be a matter for the courts, to be dealt with sensitively on a case by case basis. We recognise that at first sight it might seem odd if the defendant could escape liability where his or her conduct was deliberate but not where it was merely negligent. However, we believe that this is an inherent aspect of respect for self-determination. We therefore consider that, while the general bar to recovery where the defendant is the immediate victim should be removed - thereby solving the contribution problem in standard examples of negligently-caused self injury - the courts should have scope to decide not to impose a duty of care if satisfied that it would not be just and reasonable to do so because the defendant has chosen to cause his or her death, injury or imperilment. Accordingly, we recommend that: (8) our proposed legislation should ensure that it shall not be a bar to liability for a recognisable psychiatric illness that the illness results from the death, injury or imperilment of the defendant, but that the courts should have scope to decide not to impose a duty of care where satisfied that its imposition would not be just and reasonable because the defendant chose to cause his or her death, injury or imperilment. (Draft Bill, clause 2(2), 2(4)(a) and 5(3)) Although in some respects leaping ahead of ourselves, this seems the most convenient point to explain how our draft Bill - and, more specifically, clause 2(4) and 5(3) - enacts recommendation (8). Where the plaintiff suffers psychiatric illness as a result of the death, injury or imperilment of the defendant with whom he or she has a close tie of love and affection, our proposed legislation creates a new statutory duty of care for the purposes of the tort of negligence. The draft legislation positively spells out all the elements that make up that duty of care. Our reasons for recommending legislation in this form are fully explained in paragraphs 6.19 to 6.23 below. In this situation, therefore, the draft Bill (at clause 2(4)(a)) sets out expressly that the courts may decide not to impose the duty of care in accordance with recommendation (8). Where, on the other hand, the plaintiff suffers psychiatric illness as a result of the death, injury or imperilment of the immediate victim with whom he or she has no close tie of love and affection, our proposed legislation does not establish any new duty of care. The draft Bill (at clause 5(3)) simply provides that where a claim is brought under the common law duty of care, the court may allow the claim even if the psychiatric illness results from the defendant causing his or her own death, injury or imperilment. We do not consider it necessary to lay down in the legislation that the courts shall have scope to decide not to impose the duty of care in certain circumstances. This is because, in relation to the common law duty of care, there is inherent judicial
5.43
5.44
76
scope to decide that no duty of care is owed embodied in the now standard test 83 that no duty of care is owed unless that is just and reasonable. 5. THE CLASSIFICATION OF PRIMARY AND SECONDARY VICTIMS 5.45 We have seen that in recent cases the courts have thought it important to distinguish between primary and secondary victims; and that there is confusing inconsistency as to where and why that distinction is being drawn.84 In this section we consider how one might cut through the confusion, whether one approach to the distinction is better than another, and whether statutory reform on this distinction should be recommended. Broadly, two different classifications of primary victim (which we can loosely label as the narrow and broad classifications) have been used by the courts. The narrow classification, based on that adopted by Lord Lloyd in Page v Smith,85 would include as primary victims only those who come within the area of reasonably foreseeable physical injury. Why should these plaintiffs be singled out and treated differently from other plaintiffs who suffer psychiatric illness? First, one could argue that such plaintiffs should be singled out and treated in the same way as plaintiffs who suffer psychiatric illness consequent upon a physical injury. In Page, Lord Lloyd asked rhetorically whether it can be the law that the fortuitous absence of actual physical injury means that a different test has to be applied.86 Secondly, one could argue that there is a stronger case for recovery for psychiatric illness where the plaintiff was within physical danger. Thirdly, the fear of a proliferation of claims stemming from a single event is reduced if claimants are limited to those who were within physical danger. On the other hand, each of these arguments may be countered. First, although one can accept that it is arbitrary that a physical injury, however trivial, should result in a different test for recovery being applied, it could also be said that, where the harm suffered is psychiatric illness, an equally arbitrary line is drawn by treating differently the plaintiff who happens to be within, rather than outside, the area of physical danger.87 A modification of this narrow classification, to include
5.46
5.47
83
See further para 6.36 below. See paras 2.52 to 2.60. [1996] AC 155. [1996] AC 155, 187. But see Lord Jauncey [1996] AC 155, 172 who said that: While it is not uncommon for a severe physical injury to give rise to some degree of psychiatric illness it is not the law that such illness is presumed to be a foreseeable consequence of every physical injury, rather does each case depend on its own circumstances. See also B McDonald and J Swanton, Foreseeability in relation to negligent infliction of nervous shock (1995) 69 ALJ 945, 947. F A Trindade, Nervous Shock and Negligent Conduct (1996) 112 LQR 22, 24-25 asks: What is so magical about being within the range of foreseeable physical injury, except perhaps the mistaken view that the number of potential claimants will be limited by the nature of the case? The arbitrariness of this dividing line is shown by the decision in Young v Charles Church (Southern) Ltd, The Times 1 May 1997 (for the facts of the case see para 2.15 above). The plaintiff did not suffer psychiatric illness because of a fear for his own safety but rather because of the impact upon him of the death of his colleague. However, because he happened to come within the area of physical injury, he was able to recover under Page: Unlike Page v Smith, the plaintiffs injury was caused by the impact upon him
84
85
86
87
77
only those who reasonably feared for their own safety, might be more easily justified, since fear for ones own safety is recognised as a stressor that may cause 88 psychiatric illness. Yet, the suggestion that more favourable treatment should be given to plaintiffs who feared for their own safety than is given to plaintiffs who feared for the safety of another was historically rejected on the basis that the law could not support such a distinction.89 Secondly, the distinction tends to suggest that physical injury is more worthy of legal support than psychiatric illness, something which the medical consultees to our Consultation Paper were anxious to emphasise should be rejected.90 Thirdly, it tends to ignore the fact that psychiatric illness may be suffered in many situations (such as through stress at work), where no person has been put at risk of physical injury. 5.48 In the light of these counter-arguments, we are far from convinced that there is a valid reason for isolating those plaintiffs who came within the area of physical danger and treating them differently from those who suffer psychiatric illness but who were not at any stage in physical danger. What about a broad classification of primary victim, such as that used by Lord Oliver in Alcock v Chief Constable of South Yorkshire Police,91 based on those who
5.49
of the dreadful injuries and death of [his fellow employee], but in my judgment no valid distinction exists. The fact is that the plaintiff ... was at risk of physical injury, which he was fortunate to avoid, from an accident which could be foreseen, and his mental injury was caused by the accident which occurred. Transcript No QBENF 96/0920/C, see the judgment of Evans LJ at pp 12-13 of the transcript.
88
See para 3.6 above. Hambrook v Stokes Bros [1925] 1 KB 141, 151. On the difficulty of squaring the decision in Page v Smith with earlier case law, see N J Mullany, Psychiatric damage in the House of Lords - Fourth time unlucky: Page v Smith (1995) 3 Journal of Law and Medicine 112; P R Handford, A New Chapter in the Foresight Saga: Psychiatric Damage in the House of Lords (1996) 4 Tort L Rev 5, 7; A Sprince, Page v Smith - being primary colours House of Lords judgment (1995) 11 PN 124, 126-127. But see B McDonald and J Swanton, Foreseeability in relation to negligent infliction of nervous shock (1995) 69 ALJ 945, 946 who say that it is well-established that the law of tort draws a strict distinction between liability to primary victims and liability to secondary victims so that it is strongly arguable that dicta in the secondary victim cases are not directly applicable to primary victims. That suffering from a psychiatric illness may be as real and frequently no less painful and disabling than suffering from a physical injury was recognised by Lord Bridge in McLoughlin v OBrian [1983] 1 AC 410, 433. See also M A Jones, Liability for Psychiatric Illness - More Principle, Less Subtlety? [1995] 4 Web JCLI which is critical that as a result of Page, [t]he presence or even the foreseeability of physical injury somehow legitimates the plaintiffs claim for psychiatric injury. (Although one can perhaps argue that a distinction is given legal recognition in that the level of damages recoverable for psychiatric illness is low compared to that available for physical injury: see the Judicial Studies Boards Guidelines for the Assessment of General Damages in Personal Injury Cases (3rd ed 1996). In the Consultation Paper we asked consultees whether they thought that the problems of assessing damages for psychiatric illness are so much greater than for other types of personal injury, that a different method or regime should be adopted for the assessment of damages for psychiatric illness than is adopted for assessing damages for other types of personal injury: Consultation Paper No 137, para 2.55. 79% of consultees who considered this question were not in favour of the adoption of any different method. Generally consultees thought that although the area is more difficult than that of other types of personal injury, the principles are the same.) [1992] 1 AC 310. See para 2.53 above.
89
90
91
78
participate in the negligent event? This has the advantage of providing an explanation as to why some plaintiffs are able to recover while others, like Janet 92 Smith in Frost v Chief Constable of South Yorkshire Police, cannot. Those police officers in Frost who recovered were directly involved in the consequences flowing from the defendants negligence, unlike Janet Smith, who arrived at the ground several hours after the event. Under a broad classification of primary victim, recovery for psychiatric illness is available under the present law to plaintiffs who show proximity either in terms of participation in the event which causes the illness as primary victims, or in terms of their relationship to the immediate victim and their perception of the accident or its immediate aftermath as secondary victims. The main objection to the narrow classification is overcome: the broad classification supports the idea that recovery for psychiatric illness exists as a distinct harm, as worthy of support as physical injury. It recognises that the law relating to recovery for psychiatric illness has developed to an extent where it has a validity of its own independent of any physical injury or risk of physical injury.93 5.50 The difficulty with this broad approach is that it provides little guidance as to where the line between primary and secondary victims should be drawn. What amounts to direct involvement and what is the event with which a plaintiff need be directly involved are not clear. Even within the class of primary victim construed broadly there are different degrees of participation that are relevant. As we have seen,94 some officers in Frost were sufficiently involved to qualify as employees but not as rescuers, whilst another was sufficiently involved to be owed a duty as a rescuer, but not as an employee. Furthermore, while the broader classification may be helpful in recognising the existence of liability in nonaccident type situations, it does not by itself resolve all issues in relation to liability. For example, it would be difficult to regard a plaintiff who suffers psychiatric illness pursuant to property damage as other than a primary victim, yet this is a prime example of a situation in which it may be desirable for the law to impose special restrictions on liability.95 Given the different meanings that have been ascribed to the distinction between primary and secondary victims, and the difficulties in justifying the distinction whatever meaning is adopted, we consider that the distinction may be more of a hindrance than a help.96 In the Consultation Paper we thought it helpful as a
5.51
92
[1997] 3 WLR 1194. See para 2.37 above. P R Handford, A New Chapter in the Foresight Saga: Psychiatric Damage in the House of Lords (1996) 4 Tort L Rev 5, 7. See para 2.43 above. Another example might be liability for psychiatric illness brought on by the negligent communication of distressing news. See para 2.60 n 163 above. Based on Hegarty v EE Caledonia Ltd [1997] 2 Lloyds Rep 259, for example, one might think that to succeed as a rescuer the plaintiff has to show that he feared for his personal safety, yet this was clearly not so in the case of the successful officers in Frost v Chief Constable of South Yorkshire Police [1997] 3 WLR 1194. If a rescuer who fears for his safety is a primary victim, whereas a rescuer who does not is a secondary victim, what difference does this make in relation to the duty of care that is owed to them? For criticism of the distinction see M A Jones, Liability for Psychiatric Illness - More Principle, Less Subtlety [1995] 4 Web JCLI; N J Mullany, Psychiatric Damage in the House of Lords - Fourth time unlucky: Page v Smith (1995) 3 Journal of Law and Medicine 112; A Sprince, Page v
93
94
95
96
79
matter of exposition to refer to those who suffer psychiatric illness pursuant to physical injury or imperilment of another as secondary victims (all other plaintiffs 97 who suffer psychiatric illness being primary victims). But we did not see this distinction as obviating the need for the secondary victim to establish that a duty of care was owed by the defendant directly to him or her.98 Nor did we regard the distinction as being significant in policy terms (albeit that there is bound to be a fear of opening the floodgates of litigation where secondary victims are involved, which will not be present in many primary victim situations). 5.52 While we continue to recognise that using the primary/secondary victim terminology may be helpful for descriptive purposes, it seems that the distinction is being given a policy importance which it does not merit. The consequence has been that courts are being forced to draw the distinction with a precision that does not withstand close scrutiny. This no doubt accounts for the differing interpretations of where the line should be drawn.99 We have considered whether we should now recommend legislation to cut through the confusion being caused by the distinction between primary and secondary victims. It seems to us, however, that this would not be appropriate. Indeed, since we have no objections to the results of cases, we are not convinced that legislation would solve the problem. A provision saying something like, Courts shall not use the distinction between primary and secondary victims would be both odd and probably unworkable. Nevertheless we hope and expect that through this Report the courts will be encouraged to consider abandoning attaching practical consequences to whether the plaintiff may be described as a primary or secondary victim. We therefore recommend that: (9) although a legislative provision on this would not be appropriate, we tend to the view that the courts should abandon attaching practical significance, in psychiatric illness cases, to whether the plaintiff may be described as a primary or a secondary victim.
5.53
5.54
Smith - being primary colours House of Lords judgment (1995) 11 PN 124; F A Trindade, Nervous Shock and Negligent Conduct (1996) 112 LQR 22; N J Mullany and P R Handford, Hillsborough Replayed (1997) 113 LQR 410; H Teff, Psychiatric Injury in the Course of Policing: A Special Case? (1997) 5 Tort L Rev 184.
97
Consultation Paper No 137, para 2.3. Consultation Paper No 137, paras 4.12 to 4.13. See paras 2.57 to 2.59 above.
98
99
80
6.2
6.3
[1925] 1 KB 141. [1992] 1 AC 310. H Teff, Liability for Negligently Inflicted Nervous Shock (1983) 99 LQR 100; F A Trindade, The Principles Governing the Recovery of Damages for Negligently Caused Nervous Shock [1986] CLJ 476; A Unger, Undue Caution in the Lords (1991) 141 NLJ 1729; M Davie, Negligently Inflicted Psychiatric Illness: The Hillsborough case in the House of Lords (1992) 43 NILQ 237; S Hedley, Hillsborough - Morbid Musings of a Reasonable Chief Constable [1992] CLJ 16; K J Nasir, Nervous Shock and Alcock: The Judicial Buck Stops Here (1992) 55 MLR 705; H Teff, Liability for Psychiatric Illness after Hillsborough [1992] OJLS 440; J Steele, Two Cheers for Caparo: Ravenscroft v Rederiaktieblaget Transatlantic (1993) 56 MLR 244; K Wheat, Nervous Shock: Proposals for Reform [1994] JPIL 207; P R Handford, Compensation for Psychiatric Injury: The Limits of Liability (1995) 2 Psychiatry, Psychology and Law 37, 41-44; M A Jones, Liability for Psychiatric Illness - More Principle, Less Subtlety? [1995] 4 Web JCLI. See also Lord Oliver in Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 418.
81
Stokes Bros4 position and reintroduce the requirement of actual physical injury or reasonable fear of physical injury to the plaintiff. We do not think that such an approach is acceptable. The medical literature has shown that people do suffer psychiatric illness as a result of the injury or fear of injury to another.5 To exclude such people from recovery might be to deny recovery to those whom one may regard as most deserving (such as the mother in Hambrook v Stokes Bros) who fear for their familys safety over and above their own. In the Consultation Paper we asked whether consultees agreed with our provisional view that it should not be a requirement of liability that the psychiatric illness be sustained by reason of actual or apprehended physical injury to the plaintiff.6 Ninety-five per cent of those who commented on this point agreed with our provisional view. Many consultees agreed because they considered that the current wider rules were already unduly restrictive. One remarked that the reintroduction of such a requirement would be to substitute tidiness for justice. Generally, consultees regarded the old requirement of actual or apprehended physical injury to the plaintiff as an historical anachronism, out of keeping with medical understanding and incapable of producing fair results. 6.4 We have already recommended that the shock requirement, insisted on in Alcock v Chief Constable of South Yorkshire Police, should be abandoned.7 But this still leaves the central policy question of what should be done about the three proximity requirements laid down in Alcock. 2. POLICY LIMITATIONS 6.5 Most recent judicial statements have stressed that there may be no hard and fast dividing line between physical injury and psychiatric illness.8 In looking at the reform of liability for psychiatric illness - and at whether the Alcock proximity requirements are justified - the correct starting point must therefore be to ask why the rules which apply to liability for psychiatric illness should be any different from those that apply to liability for physical injury. The Consultation Paper set out five possible policy-based arguments for justifying the adoption of limiting factors in founding a duty of care for psychiatric illness: (i) the fear of a flood of claims;9 (ii) the potential for fraudulent or exaggerated
6.6
[1925] 1 KB 141. See paras 3.10 to 3.12 above. Consultation Paper No 137, para 5.6. See para 5.33 above. Young v Charles Church (Southern) Ltd, The Times 1 May 1997; Transcript No QBENF 96/0920/C at p 6, per Evans LJ; Page v Smith [1996] AC 155, 182-183, per Lord BrowneWilkinson; 187-188, per Lord Lloyd; see also Dulieu v White & Sons [1901] 2 KB 669, 677, per Kennedy J; Bourhill v Young [1943] AC 92, 103, per Lord Macmillan. See also M McCulloch et al, Post Traumatic Stress Disorder: turning the tide without opening the floodgates (1995) 35 Med Sci Law 287, 289: More recent work emphasizes the increasingly biological view that is now taken of PTSD. The floodgates argument may itself be subdivided into two distinct concerns: (i) the fear of a proliferation of claims from a single event (probably the arguments central force) and (ii) the possibility of a mass of claims from a mass of separate events. Both possibilities give rise to the concern that such a proliferation of claims would clog the court system and
82
10 claims; (iii) the problem of conflicting medical opinions; (iv) the feeling that psychiatric illness is less serious and therefore less worthy of legal support than physical injury; and (v) the reluctance to provide compensation since the plaintiff in these cases is only a secondary victim. Several responses to the Consultation Paper raised a sixth concern: (vi) that litigation may adversely affect the prognosis of those suffering a psychiatric illness.11
6.7
Counter arguments can be raised against each of these possible justifications. The Consultation Paper itself firmly rejected (iv) and (v),12 and cast doubt on (ii) and (iii). The main argument against (ii) and (iii) is that the same issues can arise in physical injury cases, yet it is not suggested that liability should be restricted where the plaintiff is physically injured. For example, while an injury to the back may be physically detected, the level of pain, suffering and disability that it causes cannot be. Neither are the plaintiff s and defendants medical experts likely to agree on the correct prognosis. Similarly, as regards (vi) above, the possibility of litigation has been shown in some instances to impact upon the number and severity of physical injuries.13
divert too many of societys resources into compensating the victims of psychiatric illness at the expense of other equally or more deserving plaintiffs. If the system fails to cope, the law will fall into disrepute and this would be a disservice to those few who most deserve legal support. In addition, the first possibility raises objections that to allow a mass of claims from a single event would place an undue burden on the defendant disproportionate to the negligent conduct.
10
The difficulty of prognosis in psychiatric illness cases was highlighted in the recent case, Vernon v Bosley (No 2) [1997] 3 WLR 683. The plaintiff had been awarded substantial damages for psychiatric illness in the High Court on the basis of evidence of a poor prognosis given by his two medical experts. The defendant appealed to the Court of Appeal, which affirmed the High Courts decision, although it reduced the damages award (Vernon v Bosley (No 1) [1997] 1 All ER 577 (see para 5.31 n 72 above). However, before the final order was drawn up, the defendants counsel became aware, having been anonymously sent copies of a judgment given in family proceedings between the plaintiff and his wife, that the plaintiffs psychiatric health had dramatically improved and that he was substantially, if not completely, recovered. In the light of this new evidence, the Court of Appeal reduced the quantum of damages in its final order accordingly. This point was most forcefully made by the medical consultees, including Professor Wessely, Dr Mayou, Dr Parkes and Dr Weller. Professor Wessely said that the Commissions understandable wish to avoid opening the floodgates arises out of a concern for the public or private purse, and a desire to avoid bringing the law into disrepute. However, what our Consultation Paper lacked was the public health argument for restricting the spread of litigation. See also, R Mayou, Psychological, Quality of Life and Legal Consequences of Road Traffic Accident Injury [1995] JPIL 277, 289. This concern was also raised in McCarthy v Chief Constable of South Yorkshire Police (unreported, 11th December 1996) where the court accepted expert evidence that the plaintiffs recovery had been impaired by the litigation, and that the end of the litigation would aid recovery only if the plaintiff were successful. Medical and legal experts working in the field impressed upon us the seriousness of psychiatric illness and how it can be as debilitating as physical injury, if not more so. The suggestion that the plaintiffs claim should fail because he is only a secondary victim does not withstand scrutiny. It is clear that the plaintiff must prove that the defendant owed a duty directly to him: see Lord Oliver in Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 411 and paras 2.22 to 2.24 above. Professor Wessely referred us to studies which suggest that the characteristics of systems of compensation may alter the epidemiology of such syndromes as whiplash or repetitive
11
12
13
83
6.8
After much deliberation, we do, however, remain persuaded that at this point in time (i), the floodgates argument, requires special policy limitations to be imposed over and above the test of reasonable foreseeability. In particular, we are concerned that the dividing line between what level of mental disturbance does and does not amount to a psychiatric illness is a matter of degree not kind14 and that the concept of psychiatric illness has widened significantly over the past few years.15 Our review of the relevant medical literature has led us to believe that the adoption of a simple foreseeability test would or could result in a significant increase in the number of claims which, at least at this point in time, would be unacceptable. This in turn might lead the courts to make use of policy considerations, concealed beneath the foreseeability test, in an attempt to restrict the number of successful claims. Such confusion could only result in an increased volume of litigation. While we accept that it is difficult to be sure that a move to a pure reasonable foreseeability test would open the floodgates of litigation, we believe that there is at least a significant risk of that consequence. It would be imprudent to take that risk when we can leave the courts free to develop the common law in the light of the effects of our more limited reforms. Our provisional recommendation in the Consultation Paper was that special limitations over and above reasonable foreseeability should continue to be applied to claims where the plaintiff has suffered psychiatric illness as a result of the injury or imperilment of another. Sixty-four per cent of consultees who considered this issue supported that provisional view, which we now confirm as a final recommendation. Accordingly, we recommend that: (10) special limitations over and above reasonable foreseeability should continue to be applied to claims for psychiatric illness where the defendant has injured or imperilled someone other than the plaintiff, and the plaintiff, as a result, has suffered psychiatric illness.
6.9
strain injury, and Dr Wade made a similar point in relation to back pain. That the strain of litigation may delay recovery was noted in the survey of personal injury claims carried out by the Law Commission: Personal Injury Compensation: How Much is Enough? (1994) Law Com No 225, para 14.14. As we have explained, see para 1.11 above, our terms of reference require us to recommend improvements to the existing tort system, and we are not therefore in a position to attempt to alleviate these problems by way of wholesale reform of the compensation system.
14
See paras 3.27 to 3.29 above. But see N J Mullany, Psychiatric damage in the House of Lords - Fourth time unlucky: Page v Smith (1995) 3 Journal of Law and Medicine 112, 119 for a forceful rejection of the floodgates argument. He says: There is nothing to suggest that the normal interlocutory mechanisms designed to excise baseless physical injury and other types of common law claims are somehow inadequate in psychiatric injury proceedings. Psychiatric injury is not a nebulous ailment: it is a broad recognisable medical category, of which there are numerous identified subcategories. For example, the earliest community studies into the prevalence of PTSD in the general population estimated lifetime prevalence to be approximately 1%, whilst two more recent studies found lifetime prevalence rates of 9% and 7.8%. It has been suggested that the increased prevalence findings might be due, at least in part, to the different diagnostic criteria used: R C Kessler et al, Posttraumatic Stress Disorder in the National Comorbidity Survey Arch Gen Psychiatry 1995;52:1048-1060. See further para 3.7 n 21 above.
15
84
3. OUR CENTRAL RECOMMENDATION 6.10 Having reached the conclusion that there should be special limitations over and above reasonable foreseeability, we need to consider whether all three Alcock proximity requirements need be maintained, and if not, whether the restrictions should focus on the relationship between the immediate victim and the plaintiff, or on the plaintiff s closeness to and means of perception of the accident.16 We believe that the imposition of all three proximity requirements is unduly restrictive, and that it is the last two limitations that have resulted in the most arbitrary decisions. How many hours after the accident the mother of an injured child manages to reach the hospital should not be the decisive factor in deciding whether the defendant may be liable for the mothers consequential psychiatric illness.17 We consider that so long as special control mechanisms over and above foreseeability are required in order to limit the potential number of claimants, the most acceptable method of achieving this is to restrict the claimants by reference to their connection with the immediate victim. Provided that the requirement for a close tie of love and affection between the plaintiff and the immediate victim is retained, the main floodgates objection of the possibility of many claims arising from a single event is limited. Furthermore, the advice we received from medical consultees supports the view that where there is a close tie of love and affection between the plaintiff and the immediate victim, the plaintiff s proximity to the accident or its aftermath is not always a relevant factor in determining his or her reaction to it.18 In the Consultation Paper we drew attention to the support for the abandonment of the requirements of closeness in time and space and direct perception from the judiciary and academics.19 In Jaensch v Coffey Deane J referred to expert opinion that: [T]he most important explanation of nervous shock resulting from injury to another is the existence of a close, constructive and loving relationship with that person (a close relative) and ... it is largely immaterial whether the close relative is at the scene of the accident or how he or she learns of it.20 Professor Teff points to
6.11
16
The overlap between the plaintiffs proximity to and means of perception of the accident has been noted by several commentators: M Davie, Negligently Inflicted Psychiatric Illness: the Hillsborough Case in the House of Lords (1992) 43 NILQ 237, 240; K J Nasir, Nervous Shock and Alcock: The Judicial Buck Stops Here (1992) 55 MLR 705, 708-709. F A Trindade, The Principles Governing the Recovery of Damages for Negligently Caused Nervous Shock [1986] CLJ 476, 492 poses the position of a bedridden plaintiff who is unable to visit his wife after she is injured in a gruesome road accident, but who suffers psychiatric illness on being told of the accident in vivid detail and seeing photographs of her injuries. Professor Yule said that in psychological terms, proximity is not always the determining factor regarding the production of distress. R Telford, A Rowlands and J Wright referred to the growing evidence that the impact of traumatic events can have serious and lasting effects on people who were not directly exposed to those events. The British Psychological Society said: Physical or temporal distance from an incident may not be a critical variable in causation. Consultation Paper No 137, paras 5.24 to 5.26. (1984) 155 CLR 549, 600. See also Ward J in Ravenscroft v Rederiaktieblaget Transatlantic [1991] 3 All ER 73, 79 who found, based on the medical evidence in relation to a plaintiff suffering from a prolonged depressive reaction, that (i) there is no diagnostic or medical difference in the anxiety status caused through witnessing an accident or through being
17
18
19
20
85
the distinction between primary and secondary responses to traumatic stimuli. He says that the primary response is exemplified by emotional responses such as fear, anger, grief and shock. These are immediate, automatic responses which are generally short lived and do not result in damages. The secondary responses are comparatively rare and longer-lasting reactions, including compensatable psychiatric illnesses. He suggests that while witnessing an accident is liable to increase the severity of a mothers primary reactions, it would normally have little if any effect on her secondary ones. He points to medical support for the view that hearing of the loss of a loved one in an accident could prompt an even stronger reaction than seeing it, given the human minds propensity for constructing an image of an event even more gruesome than the reality.21 6.12 Two cases, in particular, illustrate the arbitrary and, we believe, unduly restrictive decisions that are reached as a result of the Alcock requirements of closeness to the accident or its immediate aftermath and direct perception of it. In Taylor v Somerset Health Authority22 the plaintiff s husband had suffered a heart attack at work, the result of a failure by the defendants many months earlier to diagnose his serious heart disease. The plaintiff was informed of her husbands attack and within 20 minutes had arrived at the hospital to which he had been taken. After a wait of some further 20 minutes she was informed that her husband had died. Refusing to believe it, she went to see her husbands body in the hospital mortuary. Her claim for damages for the psychiatric illness which she consequently suffered was denied on the basis that she had not directly perceived the immediate aftermath of the event. Being told of her husbands death was not sufficient because the law would not compensate for shock brought about by communication by a third party, and her visit to the mortuary to confirm his death went to the fact of death rather than the circumstances in which death was brought about. The court also emphasised that, whereas in McLoughlin v OBrian,23 the plaintiff had seen her family in very much the state they would have been in at the scene of the accident, in this case the husbands body bore no marks or signs of the sort that would have conjured up for the plaintiff the circumstances
present at the aftermath or through simply learning about it from another; and (ii) the closer the relationship between the patient and the deceased and the more disturbing the circumstances and nature of the death, the more likely the depressive reaction. In that case the plaintiffs son was crushed by machinery while working on the cargo deck of a vessel owned by the defendants. He was immediately taken to hospital but died two hours later. The plaintiff, his mother, arrived at the hospital some 20 minutes later and broke down on being told of his death. She was diagnosed as suffering from a prolonged depressive reaction which was likely to remain debilitating for at least two more years. Although successful at first instance, the decision was overturned by the Court of Appeal [1992] 2 All ER 470 (Note) as being inconsistent with Alcock.
21
H Teff, Liability for Negligently Inflicted Nervous Shock (1983) 99 LQR 100, 106-107. See also H Teff, The Hillsborough Football Disaster and Claims for Nervous Shock (1992) 32 Med Sci Law 251, 252; F A Trindade, The Principles Governing the Recovery of Damages for Negligently Caused Nervous Shock [1986] CLJ 476, 487-489; K J Nasir, Nervous Shock and Alcock: The Judicial Buck Stops Here (1992) 55 MLR 705, 708; M A Jones, Liability for Psychiatric Illness - More Principle, Less Subtlety? [1995] 4 Web JCLI. [1993] 4 Med LR 34. [1983] 1 AC 410.
22
23
86
24 of his fatal attack. In Taylorson v Shieldness Produce Ltd the plaintiffs son died three days after a serious accident. The plaintiffs were unable to see their son immediately after the accident while he was undergoing examination and treatment. His father saw him in intensive care some ten hours later, although his mother was dissuaded from doing so until the next day. The Court of Appeal rejected their claims in respect of their consequential psychiatric illness, on the basis, inter alia,25 that the immediate aftermath did not extend beyond the immediate post accident treatment to their sons deterioration over the subsequent few days.
6.13
The restrictive nature of these two proximity requirements is further highlighted by the different outcome in Alcock v Chief Constable of South Yorkshire Police26 and Frost v Chief Constable of South Yorkshire Police.27 No matter how close their relationship with the immediate victim, the plaintiffs in Alcock were unable to recover damages for their psychiatric illness. Many might consider such a decision unjust, when the officers who helped at the scene of the tragedy were entitled to recover.28 Some guidance may be obtained from the experience of the Australian jurisdictions which have enacted legislation relating to liability for negligently inflicted psychiatric illness. In New South Wales, the Australian Capital Territory and the Northern Territory legislative provisions permit the spouse (defined in New South Wales to include a de facto spouse) or parent (defined to include stepparent, grandparent and persons in loco parentis) of a person killed, injured or put in peril by the defendants wrongful act to recover damages for mental or nervous shock suffered as a result, regardless of whether they saw or heard the accident. Any other member of the family (defined as child, grandchild, stepchild, any person to whom another stands in loco parentis, brother, sister, half-brother and half-sister) may claim where the immediate victim was killed, injured or put in peril within their sight or hearing.29 Rather than finding that this legislation has resulted in a flood of claims,30 it has come to be regarded as unduly restrictive. In
6.14
24
[1994] PIQR P329. The Court of Appeal also held that the plaintiffs psychiatric illness was brought about by grief rather than by shock. See para 2.64 above. [1992] 1 AC 310. [1997] 3 WLR 1194. See para 1.1 above. Law Reform (Miscellaneous Provisions) Act 1944 (NSW), s 4 as amended by Law Reform (Miscellaneous Provisions) (De Facto Relationships) Amendment Act 1984, s 3; Law Reform (Miscellaneous Provisions) Act 1955 (ACT), s 24; Law Reform (Miscellaneous Provisions) Act 1956 (NT), s 25. Although note that in New South Wales there is legislation, applicable to motor vehicle accidents and accidents suffered in the course of employment, which restricts recovery for psychological or psychiatric injury to certain categories of claimant: Motor Accidents Act 1988, s77 (NSW) and Workers Compensation Act 1987, s151P (NSW) as amended by Workers Compensation (Benefits) Amendment Act 1989. See Consultation Paper No 137, Appendix para 21. In a letter to us dated 17 January 1995 the New South Wales Law Reform Commission wrote: It is safe to assert that the 1944 Act has not led to an opening of the floodgates in NSW as far as claims for psychiatric illness are concerned. We would like to thank the Commission for its help on this matter.
25
26
27
28
29
30
87
31 Coates v Government Insurance Office of New South Wales the Court of Appeal expressed the view32 that the New South Wales legislation did not deny a person the right to rely on the common law to recover damages for psychiatric illness. Although the plaintiffs claim failed,33 Kirby P said that under the common law it was not necessary for the plaintiffs (the children of the deceased) to have witnessed or heard the accident or aftermath in order to recover. Kirby Ps obiter comments were followed in Quayle v State of New South Wales,34 where one of the plaintiffs successfully claimed damages for psychiatric illness suffered after being told of his brothers death without himself having seen the accident or its aftermath.
6.15
In the Consultation Paper we provisionally recommended that the requirements of closeness in time and space, and perception through ones own unaided senses, should be abandoned where there is a close tie of love and affection between the plaintiff and the immediate victim.35 This indeed was our central provisional recommendation. Eighty-one per cent of consultees who considered this question agreed with this provisional recommendation.36 Consultees emphasised that these two Alcock restrictions were at variance with the current psychiatric understanding of the development of mental disorders following trauma.37 Most commented on the artificial and arbitrary38 nature of the requirements, finding the distinction between the immediate aftermath and the aftermath inexplicable.39 We therefore recommend (and this is our central recommendation in this Report) that:
6.16
31
(1995) 36 NSWLR 1. Per Kirby P, Clarke JA tentatively agreeing, Gleeson CJ not deciding. The plaintiffs were unable to establish that they were suffering a recognisable psychiatric illness. Neither Gleeson CJ nor Clarke JA ruled on the issue of proximity. [1995] Aust Torts Rep 81-367. Noted in N J Mullany, Recovery for Psychiatric Injury by Report: Another Small Step Forward (1996) 4 Tort L Rev 96. See also Pham v Lawson (1997) 68 SASR 124. Consultation Paper No 137, para 5.27. Three of these consultees explicitly or implicitly noted that they agreed with our central provisional recommendation on the basis that the shock requirement was maintained (contrary to what is now our recommendation in para 5.33 above). But at least 62% of the consultees who supported our central provisional recommendation also favoured removing the requirement that the illness be shock-induced. We say at least because the figure could have been as high as 82%; unfortunately 20% of those who supported the central provisional recommendation did not give any response to the shock question. The views of many consultees may be summed up by the comments made by U Cheer: Such requirements [of closeness in time and space and perception through ones unaided senses] are artificial, bear little relation to the existence of psychiatric illness in fact, create uncertainty in the law, bring the law into disrepute ... and hence do not operate as logical, fair and workable limitations on the duty of defendants. One consultee commented that the Alcock requirements evolved in the context of traumatic events or accidents and therefore operated randomly and unfairly in claims brought by relatives of medical negligence victims. The response from the Police Federation of England and Wales was typical on this point: It further seems quite extraordinary that a plaintiff has more prospects of succeeding should he/she see the victim in a bloodied state.
32
33
34
35
36
37
38
39
88
(11)
there should be legislation laying down that a plaintiff, who suffers a reasonably foreseeable recognisable psychiatric illness as a result of the death, injury or imperilment of a person with whom he or she has a close tie of love and affection, should be entitled to recover damages from the negligent defendant in respect of that illness, regardless of the plaintiff s closeness (in time and space) to the accident or its aftermath or the means by which the plaintiff learns of it. (Draft Bill, clause 1(2), 1(3), 2(2) and 2(3))
6.17
We also explained in the Consultation Paper that, if this central recommendation were accepted, the question would arise as to the circumstances in which the communication or broadcasting of true news breaks the chain of causation between the negligence of the defendant causing the injury to the immediate victim and the psychiatric illness of the plaintiff. We were of the provisional view that the application of the normal principles of causation would satisfactorily 40 Ninety-two per cent of consultees who considered this resolve this problem. issue agreed with our provisional view. While it is not doubted that a plaintiff who satisfies the Alcock criteria should recover damages for psychiatric illness suffered as a result of the imperilment of the immediate victim, whether or not the immediate victim is actually injured,41 it has been suggested that a plaintiff should also be able to recover damages for psychiatric illness suffered where the immediate victim was not in fact imperilled, but the illness results from the plaintiff s reasonable fear that he or she had been.42 We recognise the force of this argument, since the cause of the psychiatric illness will be the same: the plaintiff s reasonable fear that a loved one may be harmed. We also accept that there will be cases where the distinction between a person who has, and a person who has not, been imperilled will be a difficult one to draw. However, as we explained above, we consider that special restrictions over and above reasonable foreseeability are needed in relation to negligently inflicted psychiatric illness in order to limit the potential number of claims.43 That limitation is achieved by restricting recovery to those who can show a sufficient closeness of relationship to another who has been killed, injured or imperilled. If recovery were extended to include those who reasonably believed that their loved
6.18
40
Consultation Paper No 137, para 5.52. See para 2.23 above. For a hypothetical example of this situation, see para 6.69 below. P G Heffey, The Negligent Infliction of Nervous Shock in Road and Industrial Accidents (1974) 48 ALJ 196, 204 and 210. See also the dissenting speech of Evatt J in Chester v Waverley Corporation (1939) 62 CLR 1, 32: In principle it is difficult to see why ... a defendant whose carelessness had undoubtedly injured the plaintiff by causing alarm and nervous shock should obtain immunity because the plaintiffs reasonable fears turned out to be mistaken. And see the dissenting speech of Sargant LJ in Hambrook v Stokes Bros. He was clearly concerned that, if Mrs Hambrooks claim were to be allowed, there would be no logical stopping point for liability. He asked: Where, as in this case, the apprehended danger is out of the sight of the plaintiff, ought the plaintiff to be entitled to recover for the illness by shock, if the facts were that the person whose safety was in question had turned off the dangerous highway, or had for some other reason never been in imminent danger at all?: [1925] 1 KB 141, 164. See para 6.8.
41
42
43
89
one had been imperilled, the potential number of plaintiffs would become limited only by the concept of reasonable foreseeability, and we fear that the policy against opening the floodgates of litigation would be undermined. We consider, therefore, that at this stage legislation should draw the line at where the loved one has in fact been killed, injured or imperilled by the defendant. 4. THE METHOD OF LEGISLATING ON OUR CENTRAL RECOMMENDATION 6.19 There are two alternative methods by which recommendation (11) could be given legislative effect. The first method would be a legislative provision which stated that a claim for psychiatric illness could succeed at common law even where the plaintiff was neither close to the accident or aftermath nor had direct perception of it. We were initially attracted by the simplicity of this approach, but we eventually rejected it for two reasons. First, we were uncertain whether, having removed these two bars, it would be clear that there remained at common law a right to recover damages for psychiatric illness suffered pursuant to anothers injury. That is, we would have legislatively repealed the ratio in Alcock v Chief Constable of South Yorkshire Police,44 without affirming that any part of it should remain intact. Secondly, we were concerned that even if it were clear that a right to recover damages for psychiatric illness based on the plaintiff s relationship with the immediate victim remained, the courts would be in a position to impose new restrictions on liability, in place of the two bars that had been removed. We wish to remove, or minimise so far as possible, the scope for this. The second method, which we have adopted, is to impose a new statutory duty of care in relation to psychiatric illness - with its elements positively spelt out in the statute - that is not restricted by reference to the plaintiff s closeness to the accident and direct perception of it. This positive approach removes any doubt that the plaintiff does have a right of recovery based on reasonable foreseeability and his or her relationship with the immediate victim, and prevents any further bars to recovery from being imposed other than those provided for in our legislation. We should emphasise the novelty of the method of legislative reform which we have adopted. Rather than laying down all the requirements of liability, we have provided for one, albeit central, component of liability: the existence of a duty of care. We intend that all other aspects of the tort of negligence, for example the rules relating to the standard of care, causation, remoteness and contributory negligence, are to apply in the normal way. Although one might draw comparisons with the Occupiers Liability Acts 1957 and 1984, which have been described as simply applied negligence,45 our proposals are not directly analogous. We do not propose to set up a new statutory tort relating to liability for psychiatric illness, but rather to lay down one segment of a finding of liability under the tort of negligence, the duty of care, but otherwise to leave the common law rules in play. In spelling out the new duty of care, we have borne in mind recommendations (7) and (8) above. It would plainly be unacceptable to remove those bars from the
6.20
6.21
6.22
44
45
90
common law and yet to re-erect them as ingredients of our proposed new statutory duty of care. Recommendation (7) - the removal of the shock requirement - is easy to accommodate, by our simply not recommending that shock should be an element of our proposed new duty of care. But in order to accommodate recommendation (8) - relating to the removal of the bar to recovery where the defendant is the immediate victim - we think that it is easier and clearer to set out a duty of care for where the defendant is the immediate victim that is separate from the duty of care imposed in the usual situation where the defendant is not the immediate victim. This is because the policy concerns are not identical and, where the defendant is the immediate victim, one cannot draw on any policies that negate the defendant owing a duty of care to the immediate victim (because the defendant cannot owe a duty of care to him or herself). 6.23 Accordingly, we recommend that: (12) to implement recommendation (11):(a) our proposed legislation should adopt the method of imposing a statutory duty of care to avoid psychiatric illness (with its elements positively spelt out in the statute) for the purposes of the tort of negligence; (Draft Bill, clause 1 and 2) our proposed legislation should actually set out two new duties of care, one for the usual situation where the defendant is not the immediate victim, and the second for the rarer situation where the defendant is the immediate victim. (Draft Bill, clause 1 and 2)
(b)
5. THE ELEMENTS OF THE NEW STATUTORY DUTY OF CARE WHERE THE DEFENDANT IS NOT THE IMMEDIATE VICTIM (1) Those to whom the New Duty of Care is Owed: a Close Tie of Love and Affection 6.24 After Alcock, a close tie of love and affection is rebuttably presumed in the case of a parent, child or spouse (and possibly fianc(e)) of the immediate victim. In other cases, such as more distant relatives or friends, the plaintiff is required to prove that such a close tie of love and affection existed.46 We consider that the class of relationships in which the tie may be presumed is currently drawn too narrowly and that in certain instances the plaintiff should be deemed (without being put to proof) to have had such a tie. In coming to this conclusion we have attempted to steer a path through various conflicting factors. We want to create a greater degree of certainty in relation to liability for psychiatric illness than is present in the current law.47 We dislike a regime which requires a plaintiff, who a fortiori is suffering from a psychiatric illness as a result of the death, injury or imperilment of a relative or friend, to prove that a close tie of love and affection existed or which allows for the possibility of distressing cross-examination on the
46
See para 2.25 above. For criticism of uncertainty in the law on liability for psychiatric illness see D Robertson, Liability in Negligence for Nervous Shock (1994) 57 MLR 649.
47
91
issue.48 We are, however, aware that it would not be possible to draw up a list that would include all categories of relationship where such a tie might exist without including many who in fact were not close. We therefore propose that a narrow list should be drawn of those who may be deemed to have had a close tie of love and affection with the immediate victim (hereinafter referred to as the fixed list) but would allow a plaintiff outside the list to prove that his or her relationship with the immediate victim was equally close. This proposal was not one of the various options specifically suggested in the Consultation Paper.49 However, we asked consultees for any other suggestions as to the formulation of the list and some put this idea forward as a preferable alternative. We are persuaded that it represents the best approach. 6.25 This seems the most appropriate point to clarify that there will be a close correlation between the requirement for a close tie of love and affection and the test of reasonable foreseeability.50 That is, if the plaintiff satisfies the requirement for a close tie of love and affection (and is assumed to be a person of reasonable fortitude), then he or she will always, or almost always,51 fall within the class of those whom it is reasonably foreseeable might suffer psychiatric illness as a result of the death, injury or imperilment of the immediate victim. We therefore recommend that: (13) the legislation should lay down a fixed list of relationships where a close tie of love and affection shall be deemed to exist, while allowing a plaintiff outside the list to prove that a close tie of love and affection existed between him or herself and the immediate victim; (Draft Bill, clause 3(1)-(5)) We further recommend, and set out the details of our reasoning in the following paragraphs, that:
6.26
6.27
48
On consultation two QCs referred to the distress that would be caused to plaintiffs by cross-examination on their love for the immediate victim. See also, M A Jones, Liability for Psychiatric Illness - More Principle, Less Subtlety? [1995] 4 Web JCLI: Is it really necessary in the interests of justice or even good policy, to conduct detailed enquiry into the personal emotional lives of plaintiffs, in effect questioning their love for the primary victim at a time when, if the allegations are accurate, they are extremely vulnerable emotionally? We asked consultees whether there should be: (a) a fixed list of qualifying relationships of close love and affection; or (b) a list of relationships in which there is a rebuttable presumption of a close tie of love and affection, while also allowing a plaintiff not on that list to prove a close tie of love and affection; or (c) a list of relationships in which there is a rebuttable presumption of a close tie of love and affection, while not allowing a plaintiff outside that list to prove a close tie of love and affection; or (d) no list at all, so that the plaintiff has to prove on the facts of each case a close tie of love and affection; or (e) an approach different to any of (a) to (d): Consultation Paper No 137, para 5.17. See paras 5.7 to 5.10 above. Of course, we do not seek to deny that it is reasonably foreseeable that plaintiffs other than those having a close tie of love and affection to the immediate victim (eg rescuers or bystanders) might suffer a psychiatric illness consequent on the immediate victims death, injury or imperilment. A conceivable example of where this might not be so is where a mother who abandoned her son at birth and had no subsequent contact with him, suffers psychiatric illness on reading, many years later, of her sons death in a road accident. The court might consider that her illness was not a reasonably foreseeable consequence of the defendants conduct.
49
50
51
92
(14)
the fixed list of relationships where a close tie of love and affection is deemed to exist should consist of the following relationships: (a) (b) (c) (d) (e) spouse; parent; child; brother or sister; cohabitant, defined as being a person who, although not married to the immediate victim, had lived with him or her as man and wife (or, if of the same gender, in the equivalent relationship) for a period of at least two years. (Draft Bill, clause 3(2), 3(4) and 3(5))
(a) the fixed list (i) parents, children and spouses 6.28 We propose that parents, children and the spouse of the immediate victim (in favour of whom there is currently a rebuttable presumption) should be included on the fixed list. Parents would include those who had adopted the immediate 52 victim; and children would include adopted children of the immediate victim. We considered whether stepparents and stepchildren should be included on the fixed list, since many clearly have a tie of love and affection as close as any parent and child. However, the relationship between a stepparent and a stepchild can clearly vary enormously, and so as not to include on the fixed list many who were in fact not close, we would need to put some restriction on this class. We also considered including plaintiffs who had treated the immediate victim as their child;53 and plaintiffs who had treated the immediate victim as their parent.54 However, to the extent that this test would be satisfied by a plaintiff producing factual evidence of his or her tie of love and affection with the immediate victim, this would amount to proving that a close tie of love and affection existed, and there would be no advantage to his or her inclusion on the fixed list. On the other hand, there might quite often be no close tie of love and affection if this test were satisfied merely by, for example, evidence that the plaintiff made financial provision for the immediate victim. It seemed to us that any sensible restriction on this class should involve the plaintiff in producing evidence that a close tie in fact existed, so that no benefit would be gained by its inclusion on the fixed list. However, we would confidently expect that many stepparents and stepchildren will readily be able to produce factual evidence that a close tie of love and affection
52
Section 39(6) of the Adoption Act 1976. In Hinz v Berry [1970] 2 QB 40 the plaintiffs feelings for her foster children were assumed without question to be the same as those for her natural children. These categories are included as dependants in s 1(3)(d) and (f) of the Fatal Accidents Act 1976. The Act does not, however, make provision for a residual category of dependant equivalent to our proposed category of those who can establish a close tie of love and affection to the immediate victim.
53
54
93
existed between them, and therefore successfully claim despite being outside the fixed list. (ii) brothers and sisters 6.29 We consider that there is a strong case for including siblings on the fixed list.55 We recognise that the relationship between brothers and sisters varies from family to family, and that creating an irrebuttable presumption in favour of siblings might allow recovery where the plaintiff could not otherwise produce sufficient evidence of a close tie. However, we consider that this risk is outweighed by the benefits gained from removing the distressing obligation to prove sibling love in each case. The very fact that the plaintiff is suffering from a psychiatric illness as a result of his or her brother or sisters injury must in itself go some way to suggest that there was a tie between them. On consultation our provisional view that any list (whether fixed or rebuttable) should include brothers and sisters56 was accepted by 93 per cent of consultees who responded to this question. We also considered whether provision should be made to include half-brothers and half-sisters on the fixed list. However, as with stepparents and stepchildren, we would have needed to put some limit on this class, so as not to include many who were in fact not close. Such a restriction could be by reference to a requirement that the halfsiblings were brought up in the same household. But again, we think that the most sensible restriction would be proof of a close tie of love and affection. There would therefore be no advantage in including half-siblings on the fixed list. (iii) cohabitants 6.30 In compiling the fixed list we have been looking for those people with whom the immediate victim had the closest of relationships. In the light of the number of couples that live together outside marriage we believe that recognition should be given to committed heterosexual and same sex relationships. Where the parties have chosen to enter into such a committed relationship and remain in it, the close tie may be reasonably deemed to exist. In contrast, we wish to exclude transitory relationships where it would not be appropriate to presume that the closest ties of love and affection had yet been forged. We therefore propose that an irrebuttable presumption is drawn in favour of those couples who have cohabited for a period of at least two years. We acknowledge that this two year cut off point may be thought to be arbitrary. But this was the period of cohabitation chosen for claims under the Fatal Accidents Act 1976 and
6.31
55
Hidden J, the judge at first instance in Alcock v Chief Constable of South Yorkshire Police, held that the relationship between brothers and sisters could be presumed to be sufficiently proximate to impose liability: [1992] 1 AC 310, 337-339. However his decision was overturned by the Court of Appeal and the House of Lords. Lord Ackner said [1992] 1 AC 310, 406: The quality of brotherly love is well known to differ widely. See also H Teff, Liability for Psychiatric Illness after Hillsborough [1992] OLJS 440, 445-446. In Turbyfield v Great Western Railway (1937) 54 TLR 221 an eight year old girl was awarded damages for the shock of being an unwilling witness to an accident caused by the defendant that fatally injured her twin sister. Consultation Paper No 137, para 5.19.
56
94
57 seems to have worked satisfactorily in that context. Moreover, a person who has cohabited with the immediate victim for a lesser period of time will have the possibility of proving that their tie of love and affection was equally as close, and factors other than time, such as the production of a child of the relationship, might be relevant. In the Consultation Paper we provisionally suggested that any list (fixed or rebuttable) should include stable heterosexual and homosexual relationships defined using a two year test.58 Ninety-four per cent of consultees who responded to this question agreed with the inclusion of stable heterosexual relationships and 87 per cent with the inclusion of stable homosexual relationships.
(b) outside the fixed list 6.32 There are many other persons who may have had a particularly close relationship with the immediate victim, such as a grandparent, grandchild, uncle, aunt or friend. In Alcock, for example, Lord Keith thought that the closeness of the tie could be presumed between fiance(e)s.59 We consider, however, that the further one moves away from the nuclear family, the more difficult it becomes to generalise about the degree of commitment involved in a relationship. We therefore propose that the fixed list should not be extended beyond that set out above, but that any plaintiff not included in that list may prove that his or her tie of love and affection was as close as those on it. It has been suggested that what is required is not a tie based on love and affection, but rather a tie of care.60 This would include, for example, a teacher-pupil or patient-nurse relationship. However, we consider that such an approach is too broad. We are not suggesting that plaintiffs who are not relatives of the immediate victim should be excluded, but we consider that it should be sufficient to say that anyone not on the list must prove that he or she had a close tie of love and affection with the immediate victim.61 (c) the timing of the close tie of love and affection test 6.34 Little attention has been paid to the question of when the plaintiff need show that he or she has a close tie of love and affection with the immediate victim. Presumably this is because the requirements of closeness to the accident, direct perception of it and shock have always linked the plaintiff to the point in time of the accident to the immediate victim. However, we have recommended that these requirements should be removed. We therefore need to specify in the proposed legislation the time at which the close tie of love and affection test must be
6.33
57
The wording is used in the definition of a dependant, although the category is limited to heterosexual relationships: s 1(3)(b) of the Fatal Accidents Act 1976. Consultation Paper No 137, para 5.19. [1992] 1 AC 310, 398. F A Trindade, The Principles Governing the Recovery of Damages for Negligently Caused Nervous Shock [1986] CLJ 476, 488. We consider that it would not be helpful to attempt to define the elements that make up a close tie of love and affection since the circumstances of the plaintiffs relationship with the immediate victim may be infinitely variable. Rather, as at common law, the courts should continue to give the words their plain meaning, resolving borderline factual issues if and when they arise on a case by case basis.
58
59
60
61
95
satisfied. At first sight, the most obvious time is that of the defendants act or omission which causes the death, injury or imperilment of the immediate victim. It is at this point that the defendant should have regard to those whom he or she can reasonably foresee might suffer as a result of his or her actions. However, we consider that this may not be wide enough. It might also be reasonably foreseeable that a person who later forms a close tie of love and affection with the immediate victim will suffer psychiatric illness, maybe, for example, as a result of long-term caring for the immediate victim who was initially a stranger. Since the requirement for a close tie of love and affection is purely a controlling tool, essentially adopted in addition to the reasonable foreseeability test in order to limit any possible flood of claims, we see no reason not to apply the test as generously as possible. We therefore consider that, as an alternative to there being a close tie of love and affection at the time of the defendants act or omission, the requirement may be satisfied where the plaintiff has a close tie of love and affection at the onset of his or her psychiatric illness. 6.35 We therefore recommend that: (15) the legislation should provide that the requirement for a close tie of love and affection between the plaintiff and the immediate victim may be satisfied either at the time of the defendants act or omission or at the onset of the plaintiff s psychiatric illness. (Draft Bill, clause 1(3)(b)) (2) Additional Policy Restrictions 6.36 The close tie of love and affection test can be regarded as a policy restriction on reasonable foreseeability designed to avoid the possibility of a flood of claims in respect of psychiatric illness suffered as a result of anothers death, injury or imperilment. But is there a need for further policy restrictions? A plaintiff who suffers physical harm need often, in practice, only show that some physical injury was a reasonably foreseeable consequence of the defendants conduct in order to establish that he or she was owed a duty of care. The additional two tests of proximity and whether it is just and reasonable to impose a duty of care, although relevant,62 will rarely be in issue.63 However, the courts do retain a certain flexibility to find that the defendant owes no duty of care in the circumstances of the particular case before them, for reasons that have nothing to do with the type of injury that the plaintiff has suffered. For example, special considerations apply where the plaintiff s injury or loss results from the defendants omission rather than commission64 or where the defendant is a public body.65 In addition, the particular circumstances of the case may militate against finding a duty of care. For example, in a claim brought against the police on behalf of the estate of a murder victim, the House of Lords held that as a matter of public policy the police
62
See Marc Rich & Co AG v Bishop Rock Marine Co Ltd [1996] AC 211 where, in a case relating to physical damage, the House of Lords said that all three elements of the tripartite test of negligence were necessary whatever the nature of the harm sustained by the plaintiff. Winfield and Jolowicz on Tort (14th ed 1994) p 84 and M A Jones, Textbook on Torts (5th ed 1996) p 34. See Winfield & Jolowicz on Tort (14th ed 1994) pp 102-109. See Winfield & Jolowicz on Tort (14th ed 1994) pp 109-113.
63
64
65
96
are immune from actions for negligence in respect of their activities in the 66 investigation and suppression of crime. We were concerned in defining our new duty of care in relation to psychiatric illness that the courts should retain this flexibility to deny a duty of care on policy grounds, while at the same time ensuring that liability should not be denied by the courts for what, in our view, would be unacceptable reasons based on the fact that the plaintiff has suffered a psychiatric illness as opposed to any physical injury. 6.37 Laying down in legislation all the circumstances in which it might not be just and reasonable to impose a duty of care where the plaintiff suffers psychiatric illness as a result of the death, injury or imperilment of a third person proved not to be practicable, not least because one cannot foresee the varied and miscellaneous situations in which liability might arise. Nor would it seem acceptable to go to the other extreme of providing the courts with a wide open discretion not to impose a duty of care on any policy grounds, since this would permit arguments about, for example, floodgates or the risk of fraudulent claims, to creep back into the courts reasoning. This might result in the whole purpose of our proposed legislation being defeated. The only policy considerations that we want the courts to consider (if a close tie of love and affection exists) are those that would be relevant even in deciding whether the defendant owed a duty of care not to cause physical injury: that is, the only policy considerations are those that would be relevant in deciding whether the defendant owed a duty of care not to cause physical injury to the immediate victim. Where, for example, the defendant was a mere passer-by who failed to warn the immediate victim of some impending danger, the defendant would not normally owe a duty of care to the immediate victim because there is no general duty to act for the benefit of another. On the same reasoning (that there is no general duty to act for the benefit of another) the defendant should not normally owe a duty of care to a loved one who suffered psychiatric illness consequent on the immediate victims injury. On the other hand, there may be situations where no duty of care as regards physical injury is owed to the immediate victim and yet there would be no policy inconsistency in holding that a duty of care is owed to a loved one of the immediate victim. For example, the defendant may owe no duty of care to the immediate victim because the immediate victims injury was sustained while they were pursuing a criminal activity, and the defendant can raise a successful plea of ex turpi causa. However, imposing a duty of care in respect of psychiatric illness suffered by the loved one of the immediate victim might not be inconsistent with the public policy reasoning which denies the duty of care to the immediate victim; and, as we have explained, there is no necessary reason why the claim for psychiatric illness (which rests on an independent duty of care owed to the plaintiff) should fail merely because no duty of care was owed to the immediate victim.67 Accordingly, we consider that the appropriate way to deal with policy restrictions on our new duty of care (over and above the close tie of love and affection) is as follows: the courts should be given scope to decide not to impose our proposed
6.38
6.39
66
Hill v Chief Constable of West Yorkshire Police [1989] AC 53. See paras 2.23 to 2.24 above.
67
97
new duty of care where satisfied that it would not be just and reasonable to impose the duty because of any factor by virtue of which the defendant owed no duty of care to the immediate victim. 6.40 Another example of what could be satisfactorily dealt with by the approach suggested in the last paragraph, is the need not to restrict unduly a persons selfdetermination. We have already discussed this where the defendant is the 68 But this policy can also be relevant where the defendant immediate victim. injures a third person and the plaintiff suffers psychiatric illness as a result. A persons freedom to take part in some dangerous activity which he or she recognises might result in another person causing him or her physical injury, or even a persons intent that another person should deliberately injure him or her, is not restricted by the duty of care not to cause physical injury. This is because the defendant can rely on the immediate victims consent as a defence to his or her claim for damages.69 However, were the defendant to face potential liability for psychiatric illness suffered by the loved one of the immediate victim, his or her willingness to carry out the activity would be restricted. As in cases where the defendant is the immediate victim, these issues are best dealt with by the courts on a case by case basis. The courts could deal with these issues by being given scope to decide not to impose the duty of care where satisfied that it would not be just and reasonable to do so because of any factor by virtue of which the defendant owed no duty of care to the immediate victim; or, more specifically - given that volenti is commonly regarded as not going to the existence of the duty of care70 because the immediate victim voluntarily accepted the risk of his or her death, injury or imperilment. We therefore recommend that: (16) where the plaintiff suffers psychiatric illness as a result of the defendant causing the death, injury or imperilment of another (the immediate victim), our proposed new duty of care should not be imposed if the court is satisfied that its imposition would not be just and reasonable either because of any factor by virtue of which the defendant owed no duty of care to the immediate victim, or because the immediate victim voluntarily accepted the risk that the defendants act or omission might cause his or her death, injury or imperilment. (Draft Bill, clause 1(4)(a) and (b))
6.41
68
See paras 5.34 to 5.43 above. Although in certain situations the criminal law may act as a deterrent: R v Brown [1994] 1 AC 212. Some writers argue that where the immediate victim voluntarily accepted the risk of injury this exempts the defendant from the duty of care which he or she would otherwise have owed: see A E Jaffey, Volenti Non Fit Injuria [1985] CLJ 87, 105 and Salmond & Heuston on the Law of Torts (21st ed 1996) p 472. But other writers treat a persons consent as reducing the normal standard of care or as being a defence to a breach of duty: for a discussion of these issues see Clerk and Lindsell on Torts (17th ed 1995) pp 88-90 and M A Jones, Textbook on Torts (5th ed 1996) pp 465-467 .
69
70
98
(a)
the plaintiff voluntarily accepted the risk of suffering the illness; the plaintiff excluded the duty; it would not be just and reasonable to impose the duty because the plaintiff was involved in conduct that is illegal or contrary to public policy. (Draft Bill, clause 1(4)(c) and 1(5))
(b) (c)
(4) No New Duty of Care where the Defendants Liability is governed by a Statutory Regime 6.43 In certain situations the defendants liability to the plaintiff may be governed by a statutory regime. For example, the international carriage of passengers by air, sea or rail is governed by various international conventions that are given the force of law in England by various statutes.76 These conventions set out the circumstances
71
See para 6.21 above. See para 6.40 n 70 above where we note that volenti may further be analysed as showing that there has been no breach of an admitted duty of care. Whilst it may be more common for an exclusion clause to refer to an exclusion of liability in respect of the defendants actions, the fact that a defendant might alternatively seek to exclude his or her duty is recognised by s 13(1) of the Unfair Contract Terms Act 1977. Under s 1(3) and 2(1) of the Unfair Contract Terms Act 1977 a person cannot validly exclude liability for negligently caused personal injury where the liability is business liability. Mere involvement in conduct that is illegal cannot always be a defence. The illegality, for example, may be of a relatively trivial kind (eg the fact that a plaintiff is speeding does not rule out a claim for negligent personal injury in the event of a car accident). See M A Jones, Textbook on Torts (5th ed 1996) pp 476-483 for a discussion of the different approaches that have been taken to this defence. International carriage of passengers by air is governed by the Warsaw Convention 1929, as amended by the Hague Protocol 1955, and by the Guadalajara Convention 1961 (dealing with the rights and liabilities of the actual carrier) given statutory force by the Carriage by
72
73
74
75
76
99
in which a carrier will be liable for damage sustained as a result of injury to a passenger suffered in the course of his or her journey. The rules of the 77 conventions prevent a carrier from excluding this liability, but impose a fixed limit on the carriers liability in respect of each passenger.78 The conventions relating to air and sea transport have been modified and applied to domestic air79 and sea80 transport, and the convention relating to air transport has been applied to the carriage of passengers by hovercraft.81 6.44 There is some doubt whether the conventions are intended to lay down an exclusive cause of action in respect of cases falling within their scope. In the case of the convention relating to international carriage by air, this doubt has been resolved by the House of Lords in Sidhu v British Airways Plc.82 The House of Lords held that the convention provides an exclusive regime in respect of the carriers liability to a passenger for injury sustained in the course of, or arising out of, international air carriage. A passenger injured in such circumstances does not, whether or not he or she has a remedy under the convention, have a claim under the tort of negligence. Where an international convention does provide an exclusive cause of action, how should this affect our reform proposals? If the convention establishes the liability of the carrier in terms which are construed as providing the sole cause of action, and which determine who may bring a claim, then we do not intend that our new duty of care should apply. We consider that imposing our new duty of care in these circumstances would risk undermining the purpose of the convention. The conventions represent a compromise position between facilitating the ease with
6.45
Air Act 1961 and by the Carriage by Air (Supplementary Provisions) Act 1962. The Montreal Protocol 1975, amending the Warsaw Convention, has been implemented by the Carriage by Air and Road Act 1979, but the relevant provisions of that Act have not been brought into force. International carriage of passengers by sea is governed by the Athens Convention 1974 given statutory force by the Merchant Shipping Act 1995. International carriage of passengers by rail is governed by Appendix A (CIV) of the Berne Convention concerning International Carriage by Rail 1980 (COTIF) given statutory force by the International Transport Conventions Act 1983. International carriage of passengers by road is governed by the Convention on the Contract for the International Carriage of Passengers and Luggage by Road as implemented by the Carriage of Passengers by Road Act 1974. However, the relevant provisions of this Act have not been brought into force.
77
See, eg, Article 23 of the Warsaw Convention 1929, as amended by the Hague Protocol 1955, given statutory force by the Carriage by Air Act 1961; Article 18 of the Athens Convention 1974 given statutory force by the Merchant Shipping Act 1995; and Article 32 of Appendix A (CIV) of the Berne Convention concerning International Carriage by Rail 1980 (COTIF) given statutory force by the International Transport Conventions Act 1983. See, eg, Article 22 of the Warsaw Convention 1929, as amended by the Hague Protocol 1955, given statutory force by the Carriage by Air Act 1961; Article 7 of the Athens Convention 1974 given statutory force by the Merchant Shipping Act 1995; and Article 30 of Appendix A (CIV) of the Berne Convention concerning International Carriage by Rail 1980 (COTIF) given statutory force by the International Transport Conventions Act 1983. The Carriage by Air Acts (Application of Provisions) Order 1967, SI 1967 No 480. The Carriage of Passengers and their Luggage by Sea (Domestic Carriage) Order 1987, SI 1987 No 670. The Hovercraft (Civil Liability) Order 1986, SI 1986 No 1305. [1997] AC 430.
78
79
80
81
82
100
which persons covered by their provisions may bring claims in respect of injuries suffered as a result of international transport and allowing the carrier to place limits on its potential liability. If a convention does provide an exclusive cause of action which replaces the common law rules of negligence, we do not intend that our new statutory duty of care should affect the liability of the carrier. 6.46 There is some doubt, however, about the applicability of the conventions to claims brought by non-passengers.83 Say, for example, that a husband suffered psychiatric illness when the ship on which his wife was travelling sank shortly after leaving the harbour. Could he bring a claim against the carrier under the common law tort of negligence? On the one hand, the conventions appear to be worded in terminology that suggests that they regulate the carriers liability to its passengers only, and that claims brought by non-passengers are wholly outside their scope.84 In this case, we intend that our new duty of care should apply.85 This would not undermine the purpose of the conventions, which, on this assumption, do not attempt to regulate the liability of a carrier to anyone other than its passengers. The result would be that the carriers liability would be widely extended, since non-passengers claims under our new duty of care, being brought outside the conventions, would not be subject to the conventions caps on liability.86 On the other hand, one might argue that the conventions are worded sufficiently broadly to impose liability to non-passengers.87 If so, we intend that our new duty of care would not apply. Rather the claim would be brought under the convention and the limits on liability would be applicable. There are also two possible midposition interpretations of the conventions as regards claims by non-passengers.
83
To clarify the position, we sought advice from experts in this area, and we are very grateful for their assistance. However, we received no consensus on the relevance of the conventions to claims by non-passengers. For example, Article 7(1) (Limit of liability for personal injury) of the Athens Convention 1974, given statutory force by the Merchant Shipping Act 1995, refers to: The liability of the carrier for the death of or personal injury to a passenger ... and Article 14 (Basis for claims) refers only to an action for damages for the death of or personal injury to a passenger. One consequence of this might be that a passenger on an aircraft who suffers psychiatric illness is unable to claim compensation, whereas a non-passenger who suffers psychiatric illness as a result of a passengers physical injury can. This is because the Warsaw Convention as amended by the Hague Protocol 1955, given statutory effect by the Carriage by Air Act 1961, refers to liability for bodily injury. In Sidhu v British Airways Plc [1997] AC 430 the House of Lords left open the question whether bodily injury included psychiatric illness. See para 6.43 above. A further convention (the Convention on Limitation of Liability for Maritime Claims 1976, given statutory force by the Merchant Shipping Act 1995) is relevant to maritime claims. It provides for a global limitation on the liability of a shipowner or salvor: Article 2(1)(a) provides that, whatever the basis of liability may be, claims in respect of loss of life or personal injury ... occurring on board or in direct connection with the operation of the ship or with salvage operations, and consequential loss resulting therefrom shall be subject to limitation of liability. If the sum of the individual claims exceeds this limit, each claim must be reduced pro rata. Claims brought under our proposed reforms would seem likely to be subject to the global limit: although personal injury is not defined for the purposes of the convention, it seems likely that on a purposive construction it would be treated as including claims for psychiatric illness. For example, Article 3(1) (Liability of the carrier) of the Athens Convention 1974 given statutory force by the Merchant Shipping Act 1995 provides that: The carrier shall be liable for the damage suffered as a result of the death of or personal injury to a passenger ....
84
85
86
87
101
The first is that domestic law determines the cause of action, but that those claims are subject to the limits laid down in the conventions. On this interpretation, we would intend our new duty of care to apply, with the limits then being treated like any other standard defence or feature of a claim in the tort of negligence. The second is that the convention imposes the liability but that the courts would look to the relevant domestic law to determine whether a claim could be brought. If this is the correct interpretation, we intend that, while our new duty of care would not directly apply to found a claim for negligence, it could be used to show that a claim would generally be available apart from the convention, so that a nonpassenger should be entitled to pursue a claim under the convention. On this assumption, since the claim would be brought under the convention, the individual limits on liability would be applicable.88 6.47 Liability in relation to nuclear accidents is also the subject of international conventions.89 The Nuclear Installations Act 196590 was enacted in order to give effect to these conventions. The Act sets out the circumstances in which a person will be liable for injury caused by the radioactive, toxic, explosive or other hazardous properties of nuclear matter or by the emission of ionising radiations.91 Liability for such injuries is governed exclusively by the terms of the Act,92 and an overall limit is placed on the potential liability of a person to pay compensation under the Act in respect of any one incident.93 Whether a claim for psychiatric illness suffered as a result of another persons injury or imperilment in a nuclear accident (under the present law or our proposed reforms) would be treated as within the Act must be doubtful. If not, it should be appreciated that, as would be the case under the present law, the liability suggested in our proposed reforms and applicable to the liability of a nuclear site licensee - would not be subject to any upper limit.94
88
In each of the air, sea and rail transport conventions the limit on liability is expressed as the maximum amount payable for the death or injury of a passenger (see Article 22 of the Warsaw Convention 1929, as amended by the Hague Protocol 1955, given statutory force by the Carriage by Air Act 1961; Article 7 of the Athens Convention 1974 given statutory force by the Merchant Shipping Act 1995; and Article 30 of Appendix A (CIV) of the Berne Convention concerning International Carriage by Rail 1980 (COTIF) given statutory force by the International Transport Conventions Act 1983). The claim for psychiatric illness brought by a loved one of the passenger would therefore reduce, presumably on a pro rata basis, the maximum amount recoverable for the physically injured passenger. The Paris Convention on Third Party Liability in the Field of Nuclear Energy 1960; The Brussels Convention on Third Party Liability in the Field of Nuclear Energy 1963; and The Vienna Convention on Civil Liability for Nuclear Damage 1963. As amended by the Nuclear Installations Act 1969. Sections 7-14 of the Nuclear Installations Act 1965. Section 12(1)(b) of the Nuclear Installations Act 1965. Section 16 of the Nuclear Installations Act 1965. Similarly it seems likely that claims for psychiatric illness (under the present law or our proposed reforms) suffered when another person is injured or imperilled by oil pollution or hazardous and noxious substances carried by sea would fall outside the ambit of the relevant provisions of the Merchant Shipping Act 1995 (giving statutory force to the International Convention on Civil Liability for Oil Pollution Damage 1992) and the Merchant Shipping and Maritime Security Act 1997 (giving statutory force to the International Convention on Liability and Compensation for Damage in Connection with
89
90
91
92
93
94
102
6.48
The duty which an occupier owes to persons on his or her property is set out in the Occupiers Liability Acts of 1957 and 1984. Where these Acts determine that 95 duty in place of the rules of common law, we do not intend that our new duty of care to avoid causing psychiatric illness should apply. The Occupiers Liability Act 1984 regulates the occupiers liability to persons other than visitors for psychiatric illness, injury being specifically defined in the Act to include impairment of a mental as well as a physical condition.96 However, the scope of the Occupiers Liability Act 1957 is more doubtful. It regulates the duty owed by an occupier to his visitors in respect of dangers97 on the premises and imposes a duty to see that the visitor will be reasonably safe.98 Whether this is broad enough to cover liability in respect of psychiatric illness which is not consequent on any physical injury to the plaintiff would not yet appear to have been tested in the courts. If the Act may be so broadly construed, we do not intend that our new duty of care should apply.99 If, on the other hand, the Act is construed so as to be limited to claims where the plaintiff has suffered physical injury, we would intend that our statutory duty of care should apply. In order to cater for the interaction between our new duty of care and statutes (including any statutes that may be implemented in the future), we therefore recommend that: (18) the new duty of care which we propose should not be imposed if a statutory provision regulates100 the defendants duty to the plaintiff in place of the common law rules of the tort of negligence. (Draft Bill, clause 1(6))
6.49
the Carriage of Hazardous and Noxious Substances by Sea 1996) respectively. In the former case the damage must be caused by contamination in order to come within the statutory provisions, and in the latter be caused by the hazardous or noxious nature of the hazardous and noxious substances. These claims would not therefore be subject to the upper limits set out in the Acts.
95
Although section 1 to both Acts provides that their rules shall have effect in place of the rules of the common law, there is some doubt over whether the Occupiers Liability Act 1957 replaces the common law rules in respect of danger due to activities carried out on the premises, as opposed to danger due to the state of the premises: Clerk & Lindsell on Torts (17th ed 1995) p 520. Section 1(9) of the Occupiers Liability Act 1984. Section 1(1) of the Occupiers Liability Act 1957. Section 2(2) of the Occupiers Liability Act 1957. North suggests that the Occupiers Liability Act 1957 does apply to claims for psychiatric illness: P M North, Occupiers Liability (1971) pp 92-94. This wording has been chosen because it would not be sufficient in all cases simply to refer to a statutory duty that replaces a common law duty not to cause psychiatric illness. In certain situations, because of the strict proximity requirements, there would be no duty of care imposed at common law for the statutory provision to replace; in others one might argue that the statute imposes no duty not to cause psychiatric illness, but rather imposes a liability to pay damages if psychiatric illness is suffered.
96
97
98
99
100
103
6. THE ELEMENTS OF THE NEW STATUTORY DUTY OF CARE WHERE THE DEFENDANT IS THE IMMEDIATE VICTIM 6.50 We recommend above101 that our proposed legislation should set out a new duty of care for where the defendant is the immediate victim that is separate from the new duty of care imposed in the usual situation where the defendant is not the immediate victim. We explained that this was easier and clearer than having one new duty of care because the policy concerns are not identical as between the two situations. And, where the defendant is the immediate victim, one cannot draw on any policies that negate the defendant owing a duty of care to the immediate victim - as in recommendation (16) above - because the defendant cannot owe a duty of care to him or herself. The important policy restriction in issue where the defendant is the immediate victim (and the plaintiff has a close tie of love and affection with him or her) is the need to respect the defendants self-determination. We have fully discussed this in Part V.102 We do not think that any other policy considerations will be relevant. In contrast to the usual situation where the immediate victim and the defendant are not one and the same, the distinction, for example, between acts and omissions, loses its relevance and the special rules relating to where the defendant is a public body cannot apply. If one puts to one side the policy restrictions that are different, the new duty of care can otherwise precisely mirror the elements of the new duty of care applicable where the immediate victim is not the defendant. That is, identical provisions can apply regarding (i) a close tie of love and affection (provided that one makes clear that the immediate victim and the defendant are one and the same);103 (ii) defences;104 and (iii) another exclusive statutory regime.105 We therefore recommend that: (19) (a) our proposed new duty of care to avoid causing psychiatric illness where the defendant causes his or her own death, injury or imperilment (that is, where the defendant is the immediate victim) should not be imposed where (in line with recommendation (8) above) the court is satisfied that its imposition would not be just and reasonable because the defendant chose to cause his or her death, injury or imperilment; (Draft Bill, clause 2(4)(a)) (b) the elements of that proposed new duty of care should otherwise precisely mirror those in recommendations (13)(15) regarding a close tie of love and affection, and those in recommendations (17)-(18) regarding defences and
6.51
6.52
6.53
101
See recommendation (12). See paras 5.34 to 5.43 above, including recommendation (8). See paras 6.24 to 6.35 above. See para 6.42 above. See paras 6.43 to 6.49 above.
102
103
104
105
104
another exclusive statutory regime. (Draft Bill, clause 2(3)(b), 2(4)(b), 2(5), 2(6) and 3(6)) 7. THE EFFECT OF OUR PROPOSED NEW STATUTORY DUTIES OF CARE 6.54 We believe that the imposition of our two proposed new duties of care will render the law more just. They will draw a more acceptable line between those who can and those who cannot recover damages for negligently caused psychiatric illness consequent on the death, injury or imperilment of another. We also believe that the legislative recommendations we put forward will be clear and easy to apply and, to that extent, will render the law simpler than it presently is. It may now be helpful if we put further flesh on the bones of our recommendations by providing examples of the effect that the imposition of our proposed new duties of care would have, first, in various hypothetical situations; and secondly, in respect of some past cases, where plaintiffs claims have failed. (1) How would our Proposed New Statutory Duties of Care apply to Various Hypothetical Examples? 6.56 In the hypothetical examples that follow,106 paragraphs 6.57 to 6.64 are situations where the defendant is not the immediate victim and where our recommendations would mean that there would be liability for psychiatric illness whereas under the present law there would not be liability; paragraphs 6.65 to 6.68 are situations where the defendant is the immediate victim and where our recommendations would mean that there would be liability (as in paragraph 6.65) or could be liability (as in paragraphs 6.66 to 6.68) whereas it would appear that under the present law there would be no liability; and paragraphs 6.69 to 6.71 are situations where, as under the present law, our recommendations would reach the same result of there being no liability for the psychiatric illness, albeit that there is a close tie of love and affection between the plaintiff and the immediate victim. In the Consultation Paper we asked whether one should distinguish between (a) the mother who suffers psychiatric illness as a result of seeing, or hearing about, her sons sudden death and (b) the mother who suffers psychiatric illness as a result of watching her son slowly die in hospital and (c) the mother who suffers psychiatric illness from looking after her injured son?107 No such distinction would be made if our reform proposals were adopted. All three mothers would be able to recover. A pupil from London is injured in a coach crash while on a school holiday in Newcastle. The crash is caused by the negligence of the driver. It takes some time for the schools headmaster to locate the whereabouts of the childs father, but on being told of the accident the childs father catches the first available flight and rushes to the hospital to which his son has been taken. He arrives about ten hours after the accident, by which time his son has undergone surgery, is out of danger
6.55
6.57
6.58
106
In each example we have assumed that the plaintiff can show that the defendant has breached any duty of care and can satisfy the usual tests of causation, remoteness etc. Consultation Paper No 137, para 5.38.
107
105
and is resting comfortably in a hospital ward. He would be able to recover for any psychiatric illness he suffers as a result of the injury to his son. 6.59 The day after his sister set off on holiday to California, a brother reads in the morning newspaper that the aircraft on which his sister was flying had a major mechanical defect causing it to crash into the Atlantic Ocean. He subsequently learns that his sister was killed, but her body is never found. The brother suffers pathological grief reaction. He would be able to claim against the negligent aircraft manufacturers. A wife sees a news item on television reporting on a crash between two commuter trains caused by the negligence of one of the drivers. Many passengers were killed. She knows that her husband is on board one of these trains, and, fearing for his safety, she suffers PTSD. In fact, he escaped unharmed from the accident. She would be able to recover damages against the negligent driver for any recognisable psychiatric illness which she suffered provided that she was able to satisfy the court that her husband had been put in peril by the crash. During a routine operation a doctor negligently uses unsterile equipment on his patient. As a result septicaemia sets in. The patient fails to respond to antibiotics, her condition slowly deteriorates and she dies a few months later. After her death, her widower suffers from an adjustment disorder. He would be able to claim damages from the doctor in respect of his illness. A wife is distressed by the sudden death of her husband from a heart attack. He had been the major breadwinner and she is forced to move to a smaller house. She takes up a new and stressful job in order to earn more money. Her son is then tragically killed in a fire at his office, caused by the negligence of the firms electrical contractors. She suffers major depression. Provided that she could satisfy the court that the death of her son was more than a trivial or insignificant cause of her depression, she would be able to recover damages from the contractors. However, the court should reduce the quantum of damages if satisfied that she was likely, in any event, to suffer some psychiatric illness at a later stage. The parents of a ten year old boy are killed outright in a road accident caused by the defendant. The child suffers brain damage. On his release from hospital he is looked after by his grandmother. She is distressed by the accident, but does not become ill. However, after several years of coping with her grandsons wayward behaviour she suffers major depression. Provided that she can satisfy the court that she has a close tie of love and affection with her grandson, she would be able to claim damages for her psychiatric illness. A woman had been receiving treatment for depression over a number of years. When her husband is tragically killed in a car accident caused by the negligence of another driver, she suffers pathological grief reaction. She would be able to recover for the additional illness which she suffers as a result of his death. Two cars collide at a roundabout. The driver of the first, who is 80 per cent responsible for the accident, is seriously injured. His wife, who was not present at the scene of the accident or its immediate aftermath, suffers psychiatric illness as a result. She would be able to recover full damages from the driver of the second 106
6.60
6.61
6.62
6.63
6.64
6.65
car in respect of her illness, but he could seek a contribution for 80 per cent of his liability from the injured husband. 6.66 A man is told by doctors that he requires a life-saving blood transfusion. However, this would be against his religious beliefs, and he refuses. After his death, his wife suffers psychiatric illness. If she were to claim against his estate in respect of her illness, she would not be able to recover if the court were satisfied that it would not be just and reasonable to impose a duty of care in this situation because the defendant chose to cause his own death. Two brothers are on a mountain-walking holiday. Visitors are advised not to walk on a particular path because there has been a recent rock-fall and it is anticipated that there will be another. One brother is adamant that he can look after himself and does not want to miss out walking on that path. He is killed by a rock-fall. His brother, who witnesses the aftermath of the accident, suffers PTSD. His mother, who was not present, cannot come to terms with the boys death, and suffers psychiatric illness. Both would be able to sue the estate for damages in respect of their psychiatric illness unless the court considered that it would not be just and reasonable to impose a duty of care in this situation because the boy chose 108 to cause his own death, injury or imperilment. A teenager accepts a ride in a light aircraft piloted by his friend when he knows that his friend is extremely drunk. Shortly after take-off the aircraft crashes and the teenager is seriously injured. His claim for personal injuries against the pilot is unlikely to be successful on the basis that he voluntarily accepted the risk of injury.109 If his parents were to claim damages against the pilot for a recognisable psychiatric illness suffered as a result of their sons injuries they could not recover if the court was satisfied that it would not be just and reasonable to impose a duty of care because no duty of care was owed to the son (or, more specifically, because their son voluntarily accepted the risk of injury). However, if held liable, the pilot could seek a contribution from the teenager on the basis that he was a joint tortfeasor in respect of his parents illness. The teenager would be liable to his parents unless the court was satisfied that it would not be just and reasonable to impose the duty of care because the teenager chose to cause his own injury or imperilment. As in paragraph 6.60 above, but the wifes next-door neighbours husband also usually travels home on the same train. She too watches the news report and suffers psychiatric illness. However, her husband had been delayed at work and arrived safely home that night on a later train. She remains so distressed by the thought of what might have occurred, that she does not recover for several weeks.
6.67
6.68
6.69
108
See paras 5.34 to 5.43 above. Morris v Murray and another [1991] 2 QB 6. The Carriage by Air Acts (Application of Provisions) Order 1967 would not be applicable, since the terms of the Warsaw Convention apply only to gratuitous carriage by aircraft when performed by an air transport undertaking: Carriage by Air Acts (Application of Provisions) Order 1967, SI 1967 No 480, Schedule 1, Part III A, Article 1.
109
107
She would not be able to recover damages under our proposed legislative reforms 110 because her husband was not imperilled by the accident. 6.70 A motorist is seriously injured in a road accident at a dangerous junction where land adjacent to the highway obscured visibility. His wife suffers PTSD on learning of his horrendous injuries. She claims damages from the local highway authority for failing to take action to improve visibility at the junction. The highway authority owed no duty of care to the motorist.111 We consider that the courts should not impose a duty of care to the wife on the highway authority: it would not be just and reasonable to do so because of the same factors by virtue of which the highway authority owed no duty of care to the motorist. A husband is in a minor car accident caused by the negligence of another driver. Nobody is injured, although the body work of the car is damaged. When his wife who has been receiving treatment for depression - is told about the accident, she suffers a major depressive episode. It would be unlikely that she could recover damages against the negligent driver. She could only do so if she could satisfy the court that her psychiatric illness was a reasonably foreseeable consequence of the defendants negligence, taking into account the prevalence of susceptibility to depression amongst the population at large. (2) How would our Proposed New Statutory Duties of Care apply to the Facts of Some Past Cases where Plaintiffs Claims for Psychiatric Illness have Failed? 6.72 In Alcock v Chief Constable of South Yorkshire Police,112 Brian Harrison and Stephen Jones (brothers of an immediate victim), Mr and Mrs Copoc (parents), and Brenda Hennessey, Catherine Jones and Denise Hough (sisters) would be able to recover damages for their psychiatric illness. Robert Alcock (brother-in-law), Joseph Kehoe (grandfather) and Alexandra Penk (fiance) would be able to recover provided that they could show that they had a close tie of love and affection with an immediate victim of the accident. In Sion v Hampstead Health Authority,113 assuming that negligence on the part of the hospital was proved, the father would successfully recover damages as a result of the psychiatric illness suffered following the death of his son. Similarly, in Taylorson v Shieldness Produce Ltd,114 the parents of the dying boy would be able to recover for their illness. In Ravenscroft v Rederiaktieblaget Transatlantic115 the plaintiff would be able to recover. She suffered a prolonged depressive reaction as a result of being told of
6.71
6.73
6.74
110
See para 6.18. Stovin v Wise [1996] AC 923. [1992] 1 AC 310. See para 2.20 above. [1994] 5 Med LR 170. See para 2.63 above. [1994] PIQR P329. See para 2.64 above. [1992] 2 All ER 470 (Note). See para 6.11 n 20 above.
111
112
113
114
115
108
her sons death when she arrived at the hospital to which he had been taken after 116 suffering a fatal injury. 6.75 The pregnant fish wife in Bourhill v Young117 would still be unable to recover damages. She had no close tie of love and affection with the immediate victim, so that she would fall outside our proposed new duties of care.
116
Similarly, the plaintiff who recovered damages in Hevican v Ruane [1991] 3 All ER 65, a decision doubted by the House of Lords in Alcock v Chief Constable of South Yorkshire Police [1992] 1 AC 310, 398, 401, and 418, would be entitled to recover damages for the psychiatric illness he suffered on learning of the death of his son. [1943] AC 92. See para 2.6 above.
117
109
7.2
7.3
Consultation Paper No 137, para 5.35. [1997] 3 WLR 1194. Consultation Paper No 137, para 5.35. See para 2.38 above. [1994] 2 All ER 1. [1997] 2 Lloyds Rep 259.
110
relation to rescuers. We do not have any criticism of the decision that the plaintiff in each case was not a rescuer, but we are concerned that the judgments tend to suggest that the plaintiff would only have qualified as a rescuer if he had come within the area of potential physical danger.7 Such a suggestion should be rejected. It should not be a condition for a rescuers entitlement to recover damages for psychiatric illness that he or she is in physical danger. In Frost, the police officers who recovered as rescuers were never themselves in any physical danger. Although it seems that the rescuer in Chadwick v British Railways Board8 was in physical danger from the debris of the crashed train, the case was clearly decided on the basis that the plaintiff suffered as a result of the horror of the experience rather than from a fear of personal danger.9 However, there is no suggestion in McFarlane or Hegarty that the court is attempting to narrow the category of rescuer, and it would appear that any confusion has arisen because of the difficulties in defining who qualifies as a primary victim. We would confidently expect that, in so far as there is any confusion on this issue, it will soon be dispelled by the courts and we do not think that legislation is necessary. 7.4 We therefore recommend that: (20) there is no need for legislation specifically dealing with the entitlement of a rescuer to recover damages for psychiatric illness. (2) Involuntary Participants 7.5 We expressed our provisional view in the Consultation Paper that there ought to be a special rule, as set out by Lord Oliver in Alcock v Chief Constable of South Yorkshire Police,10 applicable to involuntary participants.11 This would mean that where the defendant had put the plaintiff in the position of believing that he or she had been the involuntary cause of anothers death or injury, the plaintiff would be able to recover even though he or she had neither a tie of love and affection with such person nor was at the scene of the accident. Ninety-seven per cent of the consultees who considered this issue agreed with our provisional view. We remain of the view that an involuntary participant should be entitled to recover. As we explained above,12 we have been persuaded by floodgates type
7.6
In McFarlane v EE Caledonia Ltd [1994] 2 All ER 1, 10 Stuart-Smith LJ said: [T]he situation may arise where the plaintiff who is not originally within the area of danger comes into it later. In the ordinary way, such a person, who is a volunteer, cannot recover if he has freely and voluntarily entered the area of danger. This is not something that the tortfeasor can reasonably foresee, and the plaintiff may also be met with a defence of volenti non fit injuria. However, if he comes as a rescuer, he can recover. This is because a tortfeasor who has put A in peril by his negligence must reasonably foresee that B may come to rescue him, even if it involves risking his own safety. In Hegarty v EE Caledonia Ltd [1997] 2 Lloyds Rep 259, Brooke LJ adopts the same approach. [1967] 1 WLR 912. Waller J said that although there was an element of personal danger in what the plaintiff was doing, I think I must deal with this case on the basis that it was the horror of the whole experience which caused his reaction [1967] 1 WLR 912, 918. [1992] 1 AC 310, 408. See para 2.39 above. Consultation Paper No 137, para 5.37. See para 6.8.
10
11
12
111
arguments that some additional restrictions over and above reasonable foreseeability should continue to be applied in certain cases. However, the floodgates objection does not apply in relation to involuntary participants, so there would appear to be no reason to restrict their current rights of recovery. Furthermore, the medical literature suggests that the inclusion of such a category is justifiable.13 7.7 Since the publication of our Consultation Paper and the end of the consultation period, the decision of the Court of Appeal in Frost v Chief Constable of South Yorkshire Police14 has led to some uncertainty in this area. As we have seen, Henry LJ assumed that an essential characteristic of the cases on which Lord Oliver relied to formulate his special involuntary participant rule was that the defendant had been the employer of the plaintiff. It might therefore be suggested that the involuntary participant rule would not be applicable in other cases. We consider that a higher court might take a different view and conclude that this was not the basis on which the previous case law was decided.15 However, since the ambit of this category is hardly cemented in the common law, and consistently with our policy of recommending only minimal legislation to cure the most serious defects, we consider that this issue should be left to the common law to resolve. Accordingly, we recommend that: (21) there is no need for legislation specifically dealing with involuntary participants. (3) Employees 7.9 The Frost decision represents a development from the previously decided authorities16 by permitting recovery to employees who were directly involved in the course of their employment in an accident caused by their employers negligence. What, if anything, this direct involvement requires over and above the fact that it was reasonably foreseeable that the carrying out of their employment duties might cause psychiatric illness is not clear. The exact parameters of the duty of care have not yet been identified.17 However, it is again noteworthy that the floodgates objection, in its central meaning of exposure to unlimited liability, seems irrelevant here. This area of the common law is still evolving, and we consider that it would be sensible to leave it to develop by incremental judicial decision. We therefore recommend that:
7.8
7.10
13
For example, train drivers are sometimes involved in on-the-track accidents where a vehicle or person suddenly appears on the track and the driver has no opportunity to avoid a collision. Several studies have shown that drivers involved in such accidents may experience PTSD and other psychiatric problems afterwards: A Vatshelle and B E Moen, Serious on-the-track accidents experienced by train drivers: psychological reactions and long-term health effects J Psychosom Res 1997;42:43-52. [1997] 3 WLR 1194. See para 2.40 n 110 above. Young v Charles Church (Southern) Ltd, The Times 1 May 1997; Transcript No QBENF 96/0920/C at p 28, per Hobhouse LJ. See para 2.44 above.
14
15
16
17
112
(22)
there is no need for legislation specifically dealing with the entitlement of employees to recover damages for psychiatric illness suffered as a result of the death, injury or imperilment of another.
7.12
There is no doubt that there would be great practical difficulty in applying a more relaxed rule for recovery (by removing the need to show a close tie of love and affection to the immediate victim) in cases where the incident was particularly horrific. For example, it is difficult to understand why the tragedy unfolding at the Hillsborough stadium did not fall within this category. In other words, the definition of particularly horrific would be problematic. Another difficulty would be that even if, in exceptional circumstances, bystanders were able to recover, would those who voluntarily came to the scene of an accident out of morbid curiosity be able to recover or would they be ruled out on the grounds that their action fell within the doctrine of volenti non fit injuria or amounted to a novus actus interveniens?23 Other difficult issues might arise, such as whether the bystander would have been contributorily negligent if he or she did not attempt to get away from the scene as soon as possible, thereby voluntarily extending the degree of exposure. We invited the views of consultees as to whether mere bystanders should be able to recover for shock-induced psychiatric illness and, if so, in what circumstances.24 Fifty-six per cent of consultees who responded to this question were in favour of recovery.25 Of these, nearly one third supported the dicta of Lords Ackner, Keith and Oliver in Alcock, only allowing recovery where the accident was particularly
7.13
18
See para 2.45. [1992] 1 AC 310. [1992] 1 AC 310, 397, per Lord Keith; 403, per Lord Ackner; 416, per Lord Oliver. [1994] 2 All ER 1. [1994] 2 All ER 1, 14. See Jaensch v Coffey (1984) 155 CLR 549, 570, per Brennan J. Consultation Paper No 137, para 5.30. As we asked consultees for their views in relation to shock-induced psychiatric illness suffered by bystanders, it is possible that some consultees would have responded more cautiously had we asked for their views in relation to psychiatric illness suffered by bystanders howsoever caused. However, it seems likely that any psychiatric illness suffered by bystanders would be caused by the sudden appreciation by sight or sound of a horrifying event, so that the shock requirement would not be a significant hurdle to this class of plaintiff. We may therefore infer that abandoning the shock requirement would not have altered the responses of many consultees.
19
20
21
22
23
24
25
113
horrific. Others, who had objected to any special limitations applying to liability for negligently inflicted psychiatric illness, suggested that a simple test of reasonable foreseeability was sufficient. This would make it unnecessary to lay down the precise circumstances in which a plaintiff can claim. Several medical consultees pointed out that since it is clear that bystanders may well develop psychiatric illness as a result of exposure to shocking circumstances, there is no reason to exclude them. They pointed out that one of the stressors listed in the diagnostic criteria in DSM-IV for PTSD is witnessing an event that involves death, injury or a threat to the physical integrity of another person.26 7.14 Several of the consultees who did not favour recovery for mere bystanders thought that any liability test based on the horrific nature of the incident would be unworkable in practice. Many were concerned about the floodgates risk if bystanders could recover. One commented that: It is reasonable, in the balance of interests between plaintiffs and those who fund awards, to retain the existing rule that ... the risk of observing a catastrophe is one of the vicissitudes of life which are inherently uncompensatable.27 We do not see this area as representing a current serious defect in the law which should be reformed by legislation, but rather consider that it should be left to judicial development. The responses we received from consultees show that there is no current overwhelming demand for reform in relation to bystanders. However, since recovery was first allowed for nervous shock, the rules relating to liability for psychiatric illness have gradually been extended, and at each stage the objections raised in relation to the increased liability have been overcome. We therefore recognise that over time it may become accepted that wider recovery than that which we have proposed should be available to somebody who suffers psychiatric illness pursuant to the death, injury or imperilment of another. We would not wish our proposed legislative reforms to be construed as impeding the judicial development of liability to bystanders.28 Accordingly, we recommend that: (23) there is no need for legislation specifically dealing with bystanders. 2. WHERE THE PLAINTIFF SUFFERS PSYCHIATRIC ILLNESS OTHERWISE THAN AS A RESULT OF THE DEFENDANT CAUSING HIS OR HER OWN OR ANOTHER PERSONS PHYSICAL INJURY OR IMPERILMENT 7.17 Historically, claims for psychiatric illness have been brought by plaintiffs who have suffered illness as a result of a fear for their own or anothers safety. However, in more recent years it has been recognised that psychiatric illness may be caused by other events and some claims have been brought in cases which do not involve the physical injury or imperilment of any person. Generally, and because of its relative novelty, the law in this area is much less well developed than in the situations we have previously considered. Indeed, we consider that it has not reached a point where any legislative intervention would be helpful. Only if the common law
7.15
7.16
26
DSM-IV, para 309.81, p 424. See also ICD-10, para F43.1, p 147 and para 3.6 above. J Hodgson. See para 8.2 and recommendation (27) below.
27
28
114
comes to a settled position that is clearly unsatisfactory, as we believe is the case in relation to claims for psychiatric illness pursuant to a loved ones death, injury or imperilment, would we propose legislative reform. 7.18 The fact that we make no recommendations for legislative reform does not mean, however, that we do not recognise the importance of the recent developments that have been made. We pick out three situations below, where the case law has had to consider the relevant issues. These are: (i) liability for psychiatric illness induced through stress at work; (ii) liability for psychiatric illness pursuant to property damage; and (iii) liability for psychiatric illness pursuant to the negligent communication of distressing news. In particular, we recognise the significant impact that the recent case law relating to stress at work may have. By focusing solely on these three situations we do not intend to suggest that liability for negligently inflicted psychiatric illness should be limited to this range of claims. These examples simply provide recognition of the fact that psychiatric illness may be suffered as a result of a wide variety of different circumstances and not simply where there has been physical injury or imperilment (whether the defendants or another persons) and that in some of these circumstances the illness will have been caused by negligence. (1) Where the Plaintiff suffers Psychiatric Illness through Stress at Work 7.20 In the Consultation Paper we said that the reasoning of Colman J in Walker v 29 Northumberland County Council seemed to us to constitute a logical and just application of the law on safety at work to psychiatric illness.30 We understand that, as we had anticipated,31 the decision has led to other successful claims by employees against their employers for psychiatric illness suffered through work. Liability to employees has been described as the next growth area in relation to claims for psychiatric illness.32 In the Consultation Paper we expressed our provisional view that, subject to standard defences, there should be liability where an employer has negligently overburdened its employee with work thereby foreseeably causing him or her to suffer a psychiatric illness.33 Ninety-three per cent of consultees who considered this issue agreed with our provisional view. Although support for our provisional view was high, consultees pointed out many of the difficulties raised by the Walker decision. Several were concerned that the courts should be careful to contain liability in this area in order to avoid a flood of claims.34 Professor Freedland pointed to the ambiguity of the terms nervous
7.19
7.21
29
[1995] 1 All ER 737. Consultation Paper No 137, para 5.62. Consultation Paper No 137, para 5.62. N J Mullany, Fear for the Future: Liability for Infliction of Psychiatric Disorder in N J Mullany (ed), Torts in the Nineties (1997) ch 5 at p 107. Consultation Paper No 137, para 5.63. For example, Lord Justice Stuart-Smith said: [T]he courts are going to have to be careful to keep this new animal in bounds.
30
31
32
33
34
115
breakdown and mental breakdown used in the judgment35 and to the variety of different types of conduct that may cause psychiatric illness at work. He suggested that over-burdening with work should be seen as merely one example of a failure to provide a safe system of work. He also raised the difficulty of reconciling a duty of reasonable care for the mental health of employees with the duty of an employer to manage an enterprise efficiently. 7.22 We agree that the potential impact of the Walker decision should not be underestimated and that the decision raises a number of difficult issues. However, we remain of the view that it represents a just development in the law. We have considered whether we should suggest resolving some of the uncertainties left open by the decision by use of legislative reform. However, we do not think that at this stage any legislative intervention would be helpful. We are reluctant to suggest any legislative intervention when we believe that the common law is developing along the right lines. As Colman J recognised,36 difficult evidentiary issues of foreseeability and causation will inevitably arise. A codification of the law in this area would not resolve such questions of fact. Furthermore, we agree with Professor Freedlands suggestion, in his response to us, that the decision should be considered in the general context of an employers duty of care to his or her employees.37 There was no evidence of support among consultees generally for legislation to deal with this area. We therefore recommend that: (24) there is no need for legislation specifically dealing with liability for psychiatric illness suffered through stress at work. (2) Where the Plaintiff suffers Psychiatric Illness as a result of the Defendant Negligently causing Damage or Danger to Property 7.24 As we have explained,38 a plaintiff may recover damages for psychiatric illness which is suffered as a result of the defendant causing damage or danger to property, although the circumstances in which liability may be imposed are not clear. In the Consultation Paper, we expressed our provisional view that where the damage or danger is to the property of somebody other than the plaintiff, criteria for liability should be applied which are analogous to, but no less restrictive than, those adopted where the injury or safety of another person is involved.39 Ninety per cent of those who responded on this issue agreed with our provisional view.
7.23
7.25
35
Professor S Wessely also drew attention to the difficulties, from a psychiatrists point of view, associated with the concept of stress. [1995] 1 All ER 737, 749. He pointed out that issues concerning work-related psychiatric illness might be relevant to the implied term in an employment contract requiring the employer to avoid forfeiting the trust and confidence of the employee, to harassment in the workplace, and to protection for whistle-blowers. See paras 2.48 to 2.50 above. Consultation Paper No 137, para 5.56.
36
37
38
39
116
7.26
More difficult issues are raised where the damaged or endangered property belonged to the plaintiff. We expressed no provisional view, but we asked consultees whether they considered that psychiatric illness consequent upon damage to ones own property should be equated with psychiatric illness consequent upon physical injury to oneself, so that no special restrictions over and above reasonable foreseeability should be applied, and if not, what they considered 40 Eighty-five per cent of consultees who the special restrictions should be. responded to this question thought that no special restrictions should be applied. This would mean that recovery for psychiatric illness consequent upon damage to ones own property would be more easily available than for psychiatric illness consequent upon injury to another person (with whom one does not have a close tie of love and affection). At first sight such a scheme may seem unacceptable. First and foremost, it seems to give a higher value to property than to human life.41 Secondly, it places too great an emphasis on ownership. The attachment which a person feels for a property does not necessarily depend on his or her ownership of it. A person may be equally attached to the home in which they live, for example, whether it is they or their spouse who owns it. Several consultees expressed concern that property should not receive better protection than the person.42 Others, although not directly addressing the question raised in the Consultation Paper and therefore not reflected in the figure given above, opposed any tort liability at all for psychiatric illness consequent upon damage or danger to property. However, such a scheme may not be far removed from the present common law position on recovery of damages for mental distress. Although there would not appear to be any case law on whether damages for negligent property damage may include an element for mental distress, damages have been awarded for mental distress or anxiety in a case involving trespass to goods.43 If the law were to develop in a similar way in cases involving negligent damage to property, then it might seem odd that compensation for psychiatric illness, which is more extreme than mental distress, was not also recoverable. In addition, one could argue that the policy restrictions which apply where the plaintiff suffers psychiatric illness pursuant to the injury of another are not needed here, provided that the ownership requirement is maintained. The floodgates argument, in the sense of a fear of an unlimited number of claims arising from a single cause of action, is not relevant since the potential number of plaintiffs will be limited. And it might be that a more suitable control mechanism than simple ownership could be found. For example, those who originally created the property or who currently use it may have as close a tie to the property as those who own it.
7.27
7.28
7.29
40
Consultation Paper No 137, para 5.56. It may be worth noting that damage to property is not included in the list of stressors that may cause PTSD under the diagnostic criteria in DSM-IV (DSM-IV, para 309.81, p 424) or ICD-10 (ICD-10, para F43.1, p 147). S Hedley. Owen and Smith v Reo Motors [1934] All ER 734 (CA).
41
42
43
117
7.30
Maintaining our policy of minimal legislative intervention, we have decided not to make any recommendations for reform in this area. We do not wish to suggest that no recovery should be available where psychiatric illness is suffered pursuant to property damage. However, there has as yet been very little consideration of this area by the courts, and we feel that the law should be given a chance to develop as and when relevant cases arise. Only if the common law were to reach an unsatisfactory position, would legislative reform become expedient. We therefore recommend that: (25) there is no need for legislation specifically dealing with liability for psychiatric illness suffered as a result of damage or danger to property. (3) Where the Plaintiff suffers Psychiatric Illness as a result of the Negligent Communication of Distressing News
7.31
7.32
In the Consultation Paper we also raised the issue of psychiatric illness suffered 44 pursuant to the negligent communication of bad news. This provoked a wide range of responses from consultees. The majority thought that in certain circumstances there should be liability for the negligent communication of news, although half of these would restrict liability to cases where the news was false, and some consultees would require the defendant to have knowledge of, or be reckless as to, its falseness. Several consultees drew attention to the significance of any proposals for freedom of speech and information. Others pointed to the difficulties of causation that would arise - for example, if the news is true, how would the plaintiff establish that it was the communication of the news, and not the content, that caused the psychiatric illness. Some consultees were strongly opposed to any liability in these circumstances, whilst others thought that normal tort principles should apply without any additional restrictions. Some concern was expressed about imposing liability on officials whose duty it is to convey distressing news. In the light of these conflicting views, we do not think that it would be sensible for us to attempt to codify into legislation the circumstances, if any, in which a defendant may be liable for the negligent communication of distressing news. We believe that this question is best left to the courts to deal with, if and when such cases should arise. We therefore recommend that: (26) there is no need for legislation specifically dealing with liability for psychiatric illness suffered as a result of the negligent communication of distressing news.
7.33
7.34
44
Consultation Paper No 137, paras 5.57 to 5.60. See also para 2.33 n 88 above.
118
PART VIII REFORM V: THE RELATIONSHIP BETWEEN OUR RECOMMENDED LEGISLATION AND THE COMMON LAW
1. THE CONTINUED EXISTENCE AND DEVELOPMENT OF THE COMMON LAW DUTY OF CARE 8.1 We discuss in the next subsection whether our proposed new statutory duties of care should replace the common law duty of care in relation to psychiatric illness to the extent that they overlap. But, plainly, where there is no overlap, the common law duty of care should continue to apply. This will mean that of the various categories of plaintiff who can, or may be able to, recover under the present law, discussed in paragraphs 2.12 to 2.51 above, the common law duty of care will continue to apply to all, except those in category 1(c) described in paragraphs 2.19 to 2.33. So, for example, the common law duty of care relating to psychiatric illness will continue to apply to the plaintiff who is within the area of reasonably foreseeable physical injury, or who is not actually in danger but reasonably fears that he or she is, or who is a rescuer or an involuntary participant. The common law duty of care will also continue to apply to cases where the defendant has neither imperilled, nor caused physical injury to, any person as, for example, where the plaintiff suffers psychiatric illness through stress at work or consequent on damage to property. Two of our recommendations of general application, recommendations (7) and (8) on the removal of the shock-induced requirement and the bar to recovery where the defendant is the immediate victim,1 will have an impact on, by amending and clarifying, the common law duty of care. Otherwise we would intend that our proposed legislation should have no decisive impact on the common law duty of care owed to those who fall outside the new statutory duties. It should be entirely a matter for the courts - unconstrained by our proposals - to decide whether to expand the common law duty of care so as to embrace more widely than at present claims for negligently caused psychiatric illness brought by, for example, fellow employees or bystanders or those who suffer psychiatric illness where no physical injury or peril to another person is involved.2 The body of common law in relation to negligently caused psychiatric illness is still developing as medical knowledge in relation to psychiatric illness increases and gains credence with the courts.3 Our policy is one of recommending minimal legislative intervention and we would not wish the fact that our proposed new duties of care are limited to one particular class of plaintiff to be construed as favouring a policy that there should
8.2
See draft Bill, clause 5(2) and 5(3). See paras 2.48 to 2.51 above. See paras 7.11 to 7.15 above. In McLoughlin v OBrian, Lord Bridge referred to the earlier generation of judges who had regarded psychiatry and psychiatrists with suspicion, if not hostility. Such an attitude, he hoped, had now disappeared: [1983]1 AC 410, 433.
119
be no further judicial development of the common law duty of care in relation to psychiatric illness. 8.3 We therefore recommend that: (27) (a) where there is no overlap with our proposed new statutory duties of care, the common law duty of care in relation to psychiatric illness should continue to exist; (draft Bill, clause 5) (b) none of our legislative proposals should be construed as impeding the judicial development of the common law duty of care in relation to psychiatric illness.
2. REPLACEMENT OF THE COMMON LAW DUTY OF CARE WHERE THE NEW STATUTORY DUTIES OF CARE APPLY 8.4 Should our proposed new statutory duties of care replace the common law duty of care in relation to psychiatric illness to the extent that they overlap? In the Consultation Paper we invited consultees views as to whether any legislation that they favoured should be without prejudice to a plaintiff s rights at common law.4 The complexity of the question is shown by the split in the responses which we received: approximately half favoured retention of the common law rights and half their replacement.5 To retain the common law duty of care in relation to a plaintiff who falls within our proposed new statutory duties of care would appear to render the law unnecessarily complex and uncertain. The overlap might encourage plaintiffs needlessly to frame their action under both statute and common law. Although the Australian legislation, to which we have already referred,6 has been interpreted as leaving a plaintiff s common law rights untouched, this interpretation may have been necessitated by the fairly restricted category of persons who may claim under it. Only the spouse or parent of the immediate victim may recover without having been at the accident scene.7 As our statutory duties of care will benefit any person who can establish a close tie of love and affection, there should be no need for further common law expansion to protect this class. However, we believe that plaintiffs who fall within our proposed new duties of care should also be able to benefit from rights under the common law (as amended by recommendations (7) and (8)) and under any judicial developments of the common law where these rights do not overlap with our statutory duties of care. It is hard to see how any plaintiff on the fixed list could have a better cause of action at common law. However, a plaintiff who would only fall within our proposed new duty of care if he or she could prove a close tie of love and affection, might be
8.5
8.6
Consultation Paper No 137, para 5.70. 52% of consultees who responded to this question were in favour of legislation replacing a plaintiffs rights at common law; 45% favoured legislation being without prejudice to a plaintiffs common law rights; the remainder said that they were undecided. See para 6.14 above. See para 6.14 above.
120
better off bringing a claim under the present or a developed common law. For example, the friend of an immediate victim who provides aid at the scene of an accident might find it easier to bring a claim as a rescuer than adduce evidence of his or her relationship with the injured or deceased. Similarly if the common law were to develop to allow a claim by bystanders, a plaintiff who would have difficulty proving a close tie of love and affection to the immediate victim, but who had witnessed the accident, should be able to benefit from such a development. 8.7 We therefore recommend that: (28) the new statutory duties of care which we propose should replace the common law duty of care to the extent that they would overlap with it. (Draft Bill, clause 4) 3. COMMENCEMENT OF THE LEGISLATION 8.8 On the face of it, one might think that it is sufficient to recommend that our proposed legislation should apply to causes of action occurring after the commencement of the legislation. But because there may be a delay between the negligent conduct of the defendant and the onset of the plaintiff s psychiatric illness, we think that it must be clarified that the legislation should apply only to acts or omissions of the defendant, giving rise to the cause of action, which take place after the commencement of the legislation. This means, for example, that where the plaintiff suffers psychiatric illness after the commencement of the legislation, as a result of the injury of a loved one caused by the defendant before the commencement of the legislation, the plaintiff s claim would be governed only by common law rules. We therefore recommend that: (29) the proposed legislation should not apply to acts or omissions of the defendant8 which take place before the legislation comes into force (even if the cause of action accrues after the legislation comes into force). (Draft Bill, clause 6(1) and 6(2))
8.9
In respect of recommendation (28), one is concerned with a hypothetical defendant. For this reason, clause 6(1) does not refer to the defendants act or omission.
121
General Issues (2) While the legislation should refer to a recognisable psychiatric illness - that being the familiar judicial terminology to denote more than mere mental distress - a definition of what is a recognisable psychiatric illness should not be laid down in legislation. (Paragraph 5.6 and draft Bill, clause 1(2), 2(2), 4(a) and 5(1)) Under our proposed legislation it should be a requirement for liability for psychiatric illness that a duty of care be owed to the plaintiff by the defendant; and that in establishing that duty of care it should be a requirement that, at least where the plaintiff is outside the area of reasonably foreseeable physical injury, it was reasonably foreseeable that the plaintiff might suffer psychiatric illness. (Paragraph 5.10 and draft Bill, clause 1(2) and 2(2)) Our proposed legislation should not overturn the rule laid down in Page v Smith that reasonable foreseeability of psychiatric illness is not required where physical injury to the plaintiff was reasonably foreseeable. (Paragraph 5.16) Although we do not think that legislation on the point is appropriate, we tend to the view that, where the plaintiff suffers psychiatric illness as a result of the death, injury or imperilment of another, the reasonable foreseeability of the plaintiff s psychiatric illness should not always be judged with hindsight. In particular, in assessing whether the psychiatric illness was a reasonably foreseeable consequence of the defendants conduct, the court should consider whether the harm or imperilment to the immediate victim (that is the accident) was, judged prior to the accident, reasonably foreseeable. (Paragraph 5.20)
(3)
(4)
(5)
122
(6)
Although we do not think that legislation on the point is appropriate, while, in applying the test of reasonable foreseeability of psychiatric illness, it may be helpful to continue to assume that the plaintiff is a person of reasonable fortitude, that assumption should be regarded as merely an aspect of the standard approach to reasonable foreseeability that is applied in cases of physical injury. (Paragraph 5.27) Our proposed legislation should ensure that it shall no longer be a condition of liability for a recognisable psychiatric illness that the psychiatric illness was induced by shock. (Paragraph 5.33 and draft Bill, clause 1(2), 2(2) and 5(2)) Our proposed legislation should ensure that it shall not be a bar to liability for a recognisable psychiatric illness that the illness results from the death, injury or imperilment of the defendant, but that the courts should have scope to decide not to impose a duty of care where satisfied that its imposition would not be just and reasonable because the defendant chose to cause his or her death, injury or imperilment. (Paragraph 5.43 and draft Bill, clause 2(2), 2(4)(a) and 5(3)) Although a legislative provision on this would not be appropriate, we tend to the view that the courts should abandon attaching practical significance, in psychiatric illness cases, to whether the plaintiff may be described as a primary or a secondary victim. (Paragraph 5.54)
(7)
(8)
(9)
Core area of legislative reform (10) Special limitations over and above reasonable foreseeability should continue to be applied to claims for psychiatric illness where the defendant has injured or imperilled someone other than the plaintiff, and the plaintiff, as a result, has suffered psychiatric illness. (Paragraph 6.9) There should be legislation laying down that a plaintiff, who suffers a reasonably foreseeable recognisable psychiatric illness as a result of the death, injury or imperilment of a person with whom he or she has a close tie of love and affection, should be entitled to recover damages from the negligent defendant in respect of that illness, regardless of the plaintiff s closeness (in time and space) to the accident or its aftermath or the means by which the plaintiff learns of it. (Paragraph 6.16 and draft Bill, clause 1(2), 1(3), 2(2) and 2(3)) To implement recommendation (11):(a) our proposed legislation should adopt the method of imposing a statutory duty of care to avoid psychiatric illness (with its elements positively spelt out in the statute) for the purposes of the tort of negligence; (Paragraph 6.23 and draft Bill, clause 1 and 2)
(11)
(12)
123
(b)
our proposed legislation should actually set out two new duties of care, one for the usual situation where the defendant is not the immediate victim, and the second for the rarer situation where the defendant is the immediate victim. (Paragraph 6.23 and draft Bill, clause 1 and 2)
(13)
The legislation should lay down a fixed list of relationships where a close tie of love and affection shall be deemed to exist, while allowing a plaintiff outside the list to prove that a close tie of love and affection existed between him or herself and the immediate victim. (Paragraph 6.26 and draft Bill, clause 3(1)-(5)) The fixed list of relationships where a close tie of love and affection is deemed to exist should consist of the following relationships: (a) (b) (c) (d) (e) spouse; parent; child; brother or sister; cohabitant, defined as being a person who, although not married to the immediate victim, had lived with him or her as man and wife (or, if of the same gender, in the equivalent relationship) for a period of at least two years. (Paragraph 6.27 and draft Bill, clause 3(2), 3(4) and 3(5))
(14)
(15)
The legislation should provide that the requirement for a close tie of love and affection between the plaintiff and the immediate victim may be satisfied either at the time of the defendants act or omission or at the onset of the plaintiff s psychiatric illness. (Paragraph 6.35 and draft Bill, clause 1(3)(b)) Where the plaintiff suffers psychiatric illness as a result of the defendant causing the death, injury or imperilment of another (the immediate victim), our proposed new duty of care should not be imposed if the court is satisfied that its imposition would not be just and reasonable either because of any factor by virtue of which the defendant owed no duty of care to the immediate victim, or because the immediate victim voluntarily accepted the risk that the defendants act or omission might cause his or her death, injury or imperilment. (Paragraph 6.41 and draft Bill, clause 1(4)(a) and (b)) The legislation should provide that our proposed new duty of care shall not be imposed where: (a) the plaintiff voluntarily accepted the risk of suffering the illness; the plaintiff excluded the duty;
(16)
(17)
(b)
124
(c)
it would not be just and reasonable to impose the duty because the plaintiff was involved in conduct that is illegal or contrary to public policy. (Paragraph 6.42 and draft Bill, clause 1(4)(c) and 1(5))
(18)
The new duty of care which we propose should not be imposed if a 1 statutory provision regulates the defendants duty to the plaintiff in place of the common law rules of the tort of negligence. (Paragraph 6.49 and draft Bill, clause 1(6))
(19) (a) Our proposed new duty of care to avoid causing psychiatric illness where the defendant causes his or her own death, injury or imperilment (that is, where the defendant is the immediate victim) should not be imposed where (in line with recommendation (8) above) the court is satisfied that its imposition would not be just and reasonable because the defendant chose to cause his or her death, injury or imperilment. (Paragraph 6.53 and draft Bill, clause 2(4)(a)) The elements of that proposed new duty of care should otherwise precisely mirror those in recommendations (13)(15) regarding a close tie of love and affection, and those in recommendations (17)-(18) regarding defences and another exclusive statutory regime. (Paragraph 6.53 and draft Bill, clause 2(3)(b), 2(4)(b), 2(5), 2(6) and 3(6))
(b)
Areas where we do not recommend legislative reform (20) There is no need for legislation specifically dealing with the entitlement of a rescuer to recover damages for psychiatric illness. (Paragraph 7.4) There is no need for legislation specifically dealing with involuntary participants. (Paragraph 7.8) There is no need for legislation specifically dealing with the entitlement of employees to recover damages for psychiatric illness suffered as a result of the death, injury or imperilment of another. (Paragraph 7.10) There is no need for legislation specifically dealing with bystanders. (Paragraph 7.16) There is no need for legislation specifically dealing with liability for psychiatric illness suffered through stress at work. (Paragraph 7.23)
(21)
(22)
(23)
(24)
125
(25)
There is no need for legislation specifically dealing with liability for psychiatric illness suffered as a result of damage or danger to property. (Paragraph 7.31) There is no need for legislation specifically dealing with liability for psychiatric illness suffered as a result of the negligent communication of distressing news. (Paragraph 7.34)
(26)
Relationship between our recommended legislation and the common law (27) (a) Where there is no overlap with our proposed new statutory duties of care, the common law duty of care in relation to psychiatric illness should continue to exist. (Paragraph 8.3 and draft Bill, clause 5) None of our legislative proposals should be construed as impeding the judicial development of the common law duty of care in relation to psychiatric illness. (Paragraph 8.3)
(b)
(28)
The new statutory duties of care which we propose should replace the common law duty of care to the extent that they would overlap with it. (Paragraph 8.7 and draft Bill, clause 4) The proposed legislation should not apply to acts or omissions of the defendant2 which take place before the legislation comes into force (even if the cause of action accrues after the legislation comes into force). (Paragraph 8.9 and draft Bill, clause 6(1) and 6(2))
(29)
(Signed)
MARY ARDEN, Chairman ANDREW BURROWS DIANA FABER CHARLES HARPUM STEPHEN SILBER
126
APPENDIX A DRAFT NEGLIGENCE (PSYCHIATRIC ILLNESS) BILL [N.B. The text of the draft Bill is available as a separate downloadable PDF file] EXPLANATORY NOTES
Clause 1
Clause 1 (like clause 2) removes some (but not all) of the special restrictions that currently exist in relation to liability in the tort of negligence for psychiatric illness which is suffered as a result of the death, injury or imperilment of another person. It does so by imposing a duty of care and by spelling out all the criteria that determine the existence of that duty. Clause 1(1) provides that clause 1(2) (as qualified by clauses 1(3) - 1(6)) imposes a duty of care for the purposes of the tort of negligence and that the criteria for determining the existence of that duty of care are exclusively set out in clause 1. The Bill does not deal with any other aspect of a claim for the tort of negligence and it follows that the usual rules relating to, for example, the standard of care, causation, remoteness, and defences are applicable. Clause 1(2) lays down that a duty of care is owed to avoid causing the plaintiff to suffer a reasonably foreseeable recognisable psychiatric illness as a result of the death, injury or imperilment of another (the immediate victim). Recognisable psychiatric illness is not defined in the Bill: it is a term developed by the courts at common law and acknowledges that mere mental distress is not compensatable. Clause 1(3) qualifies the reasonable foreseeability test of clause 1(2) by laying down two restrictions. Clause 1(3)(a) provides that the defendant must have caused the death, injury or imperilment of the immediate victim. The effect is that the statutory duty of care is not imposed where, for example, the defendant causes the plaintiff to suffer psychiatric illness by the manner of reporting to him that his loved one has been hurt. Clause 1(3)(b) lays down the central restriction of there needing to be a close tie of love and affection between the plaintiff and the immediate victim. While this restriction was developed at common law, other major restrictions presently applicable at common law (in particular, that the plaintiff be close to the accident in time and space, that knowledge of the accident be experienced through the plaintiffs unaided senses, and that the illness be shock-induced: see Alcock v Chief Constable of South Yorkshire Police) are inapplicable in respect of the statutory duty of care. Clause 1(3)(b) makes clear that there must be a close tie either immediately before the defendants act or omission or immediately before the onset of the psychiatric illness (so that included, for example, is the plaintiff who suffers psychiatric illness in looking after the immediate victim who was initially a stranger). Clause 3 applies in determining whether there is a close tie of love and affection.
127
Clause 1(4) further qualifies the reasonable foreseeability test of clause 1(2) by giving the courts scope to decide that a duty of care should not be imposed, because not just and reasonable, in three situations. The first situation is where any factor by virtue of which the defendant owed no duty of care to the immediate victim also means that no duty should be owed to the plaintiff (for example, where the immediate victims harm was caused by the defendants omission not commission). The second situation is where the immediate victim has voluntarily accepted the risk of the defendant causing his death, injury or imperilment. A separate provision to (a) is included because volenti non fit injuria may be regarded not as negating the duty of care owed to the immediate victim by the defendant, but as going to the standard of care or as being a defence. The third situation is where the plaintiff was involved in conduct which is illegal or contrary to public policy. A provision is needed here because, while often treated as a defence and therefore applicable to the statutory duty of care under the usual rules of negligence, ex turpi causa non oritur actio may also be treated as a flexible factor going to the existence of the duty of care. Clause 1(5) further qualifies clause 1(2) by laying down that no duty is imposed where the plaintiff has voluntarily accepted the risk of suffering the psychiatric illness (this provision being needed because volenti non fit injuria may be regarded as going to the existence of the duty of care rather than as going to the standard of care or as being a defence); or where the plaintiff has excluded the duty (but note that, under section 2(1) of the Unfair Contract Terms Act 1977, he cannot validly do so where the liability is business liability). Clause 1(6) ensures that the statutory duty does not arise where another statutory provision governs the defendants duty to the plaintiff. We explore examples of this in paragraphs 6.43-6.48.
Clause 2
Clause 2 imposes a second statutory duty of care, in this case where the plaintiff suffers a reasonably foreseeable recognisable psychiatric illness as a result of the defendant causing his own (rather than another persons) death, injury or imperilment. In contrast, it appears that under the present law a person never owes a duty of care to avoid causing psychiatric illness as a result of harming or endangering himself. One effect of clause 2 would be that a person who is liable for breach of a duty of care under clause 1 would be able to claim contribution (under the Civil Liability (Contribution) Act 1978) from an immediate victim who was primarily responsible, through contributory negligence, for his own injuries and hence for the plaintiffs psychiatric illness. The subsections of clause 2 directly mirror the subsections of clause 1 and the explanatory notes above should therefore be referred to. But clause 2(4) requires a little more explanation. This subsection qualifies the reasonable foreseeability test of clause 2(2) by giving the courts scope to decide that a duty of care should not be imposed, because not just and reasonable, in one of two situations. There is no equivalent to clause 1(4)(a) because the defendant and the immediate victim are here one and the same, and the defendant cannot owe a duty of care to himself. The first situation is where the defendant chose to cause his own death, injury or imperilment. The idea here is that the courts should be free to respond sensitively, and to attach importance, to selfdetermination. The second situation is where the plaintiffs conduct was illegal or contrary to public policy: this precisely mirrors clause 1(4)(c).
Clause 3
Clause 3 applies in determining whether there is a close tie of love and affection for the purposes of clauses 1 and 2 (for the purposes of clause 1, it is dealing with whether the
128
plaintiff had a close tie of love and affection with the immediate victim; and for the purposes of clause 2, whether the plaintiff had a close tie of love and affection with the defendant). Clause 3(4) sets out a list of those who, by clause 3(2), shall be conclusively taken to have had a close tie of love and affection with the person killed, injured or imperilled. It follows from giving the terms in clause 3(4) their standard meaning that sons and daughters who are over 18 fall within clause 3(4)(c); while, in contrast, step-parents, stepchildren, step-brothers and step-sisters do not fall within clause 3(4). Clause 3(5)(a) refers to heterosexual cohabitants, and 3(5)(b) refers to same-sex cohabitants. Clause 3(3) provides that if the plaintiff does not fall within this list, he shall be required to show that a close tie of love and affection in fact existed.
Clause 4
This clause abolishes the common law duty of care for negligently inflicted psychiatric illness to the extent that it would otherwise overlap with the two statutory duties of care laid down in clauses 1 and 2. But where the statutory duties of care do not apply, the common law duty of care survives. This means that the common law duty of care relating to psychiatric illness (as qualified by clause 5) will continue to apply to, for example, rescuers and those who suffer psychiatric illness through stress at work.
Clause 5
Clause 5 amends and clarifies the common law duty of care for negligently inflicted psychiatric illness. Clause 5(2) removes any requirement that the illness be induced by a shock. Clause 5(3) ensures that it shall not be an absolute bar to recovery that the plaintiffs psychiatric illness results from the defendant causing his own death, injury or imperilment: but in applying the standard common law approach to determining the duty of care, the courts will have scope to decide that the imposition of a duty of care is not just and reasonable because the defendant chose to cause his death, injury or imperilment (see analogously clause 2(4)(a)).
Clause 6
Clause 6 dealing with commencement will ensure that, where there is a delay between the defendants acts or omissions and the onset of the plaintiffs psychiatric illness, the Bill will only apply where the acts or omissions occur after the commencement date. Clause 6(1) does not refer to the defendants act or omission because in respect of clause 4 (which must have effect at exactly the same time as clauses 1-3) one is concerned with a hypothetical defendant.
129
APPENDIX B
List of Persons and Organisations who Commented on Consultation Paper No 137
Consultation took place in 1995 and closed on 1 August 1995. consultees may have altered since then. The description of
GOVERNMENT BODIES Lord Chancellors Department Scottish Office, Home and Health Department JUDICIARY AND PRACTITIONERS (i) Judiciary The Hon Mr Justice Buxton His Honour Judge Cox QC The Hon Mr Justice Latham Master Macfarlane The Hon Mr Justice Phillips The Hon Mr Justice Rougier Rt Hon Lord Justice Stuart-Smith HM Council of Circuit Judges Judges of the Queens Bench Division (ii) Barristers Andrew Buchan Raymond Croxon QC Iain S Goldrein Allan Gore Adrian Hamilton QC and Alistair Schaff David Hart Benet Hytner QC David Kemp QC Brian Langstaff QC Charles J Lewis Harvey McGregor QC Colin MacKay QC Norman Marsh QC CBE Sir Michael Ogden QC A Ritchie Raymond Walker QC James Watson Adrian Whitfield QC (iii) Solicitors Simon Allen, Russell Jones & Walker Anthony Gold Lerman & Muirhead Bevan Ashford Boyes Turner & Burrows Davies Arnold Cooper 130
Roderick Dawson, Morecroft Urquhart Hextall Erskine & Co Richard Hoare Paul Hughes Geraldine McCool, Leigh, Day & Co Peter Metcalf, Hammond Suddards Michael Napier, Irwin Mitchell Maria Pittordis, Hill Taylor Dickinson Robin Thompson & Partners Nicola Solomon, Stephens Innocent Christopher Sprague, Ince & Co L J Watmore A F Whitehead, Russell Jones & Walker LEGAL ORGANISATIONS Association of Personal Injury Lawyers The Institute of Legal Executives The Law Society The Law Society of Northern Ireland Personal Injuries Bar Association Law Reform Sub-Committee Society of Public Teachers of Law Young Solicitors Group ACADEMIC LAWYERS Professor Roger Brownsword Professor Richard Buckley Peter Cane Ursula Cheer John Cooke and Anthony Harvey Professor A M Dugdale Rosalind English Associate Professor Tan Keng Feng John Fleming Dr Mark Freedland Professor G H L Fridman Professor D S Greer Professor Andrew Grubb Dr P R Handford Professor Carol Harlow Steve Hedley John Hodgson John L Horrocks Laura C H Hoyano Professor J A Jolowicz Professor Michael A Jones Professor Richard Lewis Stiewart Lindenbergh and Carel Stolker Lesley Lomax Professor R D Mackay Dr Danuta Mendelson 131
Maureen Mulholland John Murphy Ken Oliphant, Meredith Blake, Mark Lunney and Maleiha Malik Professor David Oughton and John Lowry Deborah L Parry W E Peel Professor David W Robertson Professor W V H Rogers Jane Stapleton Teresa Sutton Associate Professor Jane Swanton and Barbara McDonald Carol Tan Professor H Teff Andrew Tettenborn Rosemary Tobin Professor Stephen Todd Professor Francis Trindade Prue Vines Kay Wheat MEDICAL AND PSYCHOLOGICAL EXPERTS Professor David Alexander British Association of Psychotherapists British Medical Association British Psychological Society Professor Philip Feldman Gisli H Gudjonsson Professor Sheila Hollins International Stress Management Association (UK) Dr M J MacCulloch Miles Mandelson Dr Richard Mayou and Dr Bridget Bryant Dr David C Muss Dr Colin Murray Parkes Psychiatric Rehabilitation Association Professor R M Rosser Royal College of Nursing Royal College of Psychiatrists (Mental Health Law Group) Dr Myre Sim Rosemary Telford, Alison Rowlands and John Wright Dr James Thompson Professor Michael R Trimble Dr Gordon Turnbull Professor O L Wade Dr Malcolm Weller Dr Simon Wessely Professor William Yule OTHER ORGANISATIONS Association of British Insurers 132
Association of Chief Police Officers Association of Chief Police Officers in Scotland Association of Professional Ambulance Personnel BBC British Fire Services Association British Safety Council The Chief and Assistant Chief Fire Officers Association Combat Stress Compassionate Friends Cornhill Insurance Disabled Living & Design Justice for Victims London Fire and Defence Civil Authority National Association for Bereavement Services Order of St John Police Federation of England and Wales Road Peace Royal Insurance St Paul International Insurance Company Ltd Sun Alliance UK Direct Support after Murder & Manslaughter UK Claims Managers Association Victim Support INDIVIDUALS D C Spencer
133