Professional Documents
Culture Documents
Cambios de Alimentación en El Alzheimer
Cambios de Alimentación en El Alzheimer
Abstract. Eating impairment is well documented in the walking away from food at the table, to spitting out food,
late stage of Alzheimers disease (AD) but when these and choking on liquids and solids [812].
eating changes actually begin in the disease process is Eating can be defined by two components, self-
not known. Eating was defined as consisting of two com- feeding and swallowing; therefore, eating dysfunction in
ponents, self-feeding and swallowing. Self-feeding and AD may involve impaired ability in transferring food
swallowing of healthy elderly were compared with a from the table to the mouth (self-feeding), impaired
group of individuals with mild AD. AD subjects received transfer of food from the mouth into the stomach (swal-
significantly more partner-initiated cues or direct assis- lowing), or both. When self-feeding becomes impaired,
tance than controls. In addition, subject-initiated cued the individual is at risk for malnutrition and its sequelae;
behaviors occurred more frequently in the AD group. AD when self-feeding ceases altogether, the individual be-
subjects demonstrated significantly prolonged swallow comes dependent on the motivation, commitment, and
durations for the oral transit duration (cookie), pharyn- skill of another person or persons for adequate nutrition
geal response duration (liquid), and total swallow dura- and hydration. Swallowing dysfunction, regardless of
tion (liquid). This pilot study suggests that self-feeding self-feeding status, places the demented individual at risk
and swallowing changes may occur early in the course for malnutrition, aspiration pneumonia, nonoral feeding,
of AD. and death.
What is known about eating impairment in AD is
Key words: Alzheimers disease Eating Swal-
primarily limited to institutional settings [1117]. As
lowing Self-feeding Deglutition Deglutition
many as 50% of nursing home residents may require
disorders.
feeding assistance ranging from organizing and setting
up the meal to total assistance, and 32% of nursing home
residents have been reported to have swallowing prob-
Cognitive impairment is the defining feature of Alzhei- lems [7]. With two exceptions [18,19], most research has
mers disease (AD) but it is the concomitant behavioral focused on problems associated with institutional care-
and functional impairments that place the individual with givers feeding AD residents. No one to date has system-
AD at increased risk for institutionalization and death atically investigated self-feeding and swallowing in AD
[1]. The inability to eat is the most life-threatening [26] to identify patterns of and relationships between these
of all the functional impairments in AD, with 50% of the two eating components. One reason for this lack of in-
patients losing the ability to feed themselves 8 years after quiry is that eating dysfunction is most manifest in late-
diagnosis [7]. Altered eating behaviors in AD range from stage dementia when disease severity prohibits many in-
dividuals from reporting symptoms or cooperating with
oropharyngeal examinations [8,11,12].
A major goal in caring for individuals with AD is
Correspondence to: Beverly Priefer, Ph. D., Geriatric Research, Edu- to maintain independent ADL function for as long as
cation, and Clinical Center, Wm. S. Middleton Memorial Veterans
Administration Hospital, 2500 Overlook Terrace, Madison, WI 53705, possible [20]. Loss of ADL function has important im-
USA plications for either family or professional caregivers
B.A. Priefer and J. Robbins: Eating in Alzheimers Disease 213
since decreased function leads to increased dependency focused in the lateral view from the lips anteriorly to the pharyngeal
and increased care needs, both of which are emotionally wall posteriorly, and from the nasopharynx superiorly to the cervical
esophagus inferiorly. The swallow study was conducted by a speech
challenging and time consuming [2125]. Understanding pathologist following a standard protocol [31] using three different
the nature and course of eating dysfunction in AD using consistencies: a 3-ml liquid; a 3-ml semisolid (pudding mixed with
more physiological measures of dysfunction could po- barium); and a 1.5-cm cookie (coated with EZ-paste). The material
tentially result in specific problem-oriented interventions was placed in the subjects mouth, the fluoroscopy was turned on, and
earlier in the disease process. the subject was asked to swallow on command. Subjects performed two
trials of each consistency.
We hypothesized that changes in eating behav- To determine if swallowing differs in AD and elderly control
iors begin before that time when eating impairment ren- subjects, durations of five components of a swallow were analyzed
ders the individual dependent on others, or a feeding [32]. The definitions of these durations are shown in Table 1. A vid-
tube, for nutritional intake. Since early-stage AD holds eocounter timer imprinted a time code (accurate to 0.02 sec) on each
the greatest promise for intervention [26], we speculated swallow tape. Frame by frame analysis allowed quantification of the
duration of each of the following 5 swallow durations. The occurrence
that identifying early changes in eating behaviors would of aspiration, defined as the entry of material below the vocal cords,
be the first step in understanding the progressive nature was also recorded.
of these changes and essential for designing interven- A two-sample Wilcoxon test [33] was used to compare control
tions to maximize eating independence for as long and as and AD groups on the swallow duration measures. Since swallowing
safely as possible. Therefore, a pilot study was under- durations were hypothesized to be more prolonged in AD subjects, all
comparisons were made using one-tailed tests with a 4 0.05. Data
taken to design and implement an eating assessment pro- analyses were performed using the first swallow duration for each
tocol to determine if differences exist in the eating be- consistency. Averaging the durations of the two swallows per consis-
haviors of individuals with mild AD compared with nor- tency was not done because of technical difficulties in the timing of
mal age-matched controls. Specifically, the questions fluoroscopy activation for the second swallow in some of the subjects.
asked were (1) Do swallowing durations differ in nor-
mal elderly and individuals with early stage AD? and
(2) Do self-feeding dependency and self-feeding behav- Self-feeding Assessment
iors differ in normal elderly and individuals with early
A standard meal was developed to assess self-feeding behaviors. This
stage AD?
meal was eaten by the subject, the spouse (for controls, spouse or
friend), and one of the authors (BAP). The meal was designed to
Methods represent the everyday challenges people encounter when self-feeding
and was patterned after the type of regular diet served to patients in a
Subjects hospital setting. The food and eating challenges were also similar to
those experienced in other common eating situations such as airline
Healthy, older adults and individuals with AD were recruited as paid meals or buffet or cafeteria-style meals. Figure 1 is a sketch of a
subjects. Normal control subjects were recruited from area senior cen- hospital meal and illustrates several self-feeding tasks that the patient is
ters and volunteer organizations. All potential control subjects had to required to perform: opening a package that contains silverware and a
score 28 or higher on the Mini-Mental State Examination (MMSE) [27] folded napkin, opening a carton of milk, opening a salad dressing
and did not suffer from any neuromuscular conditions that interfered packet and pouring the dressing on the salad, opening salt, pepper, and
with their motor ability to self-feed. AD subjects were recruited from cream packets, buttering bread, and using a knife and fork to cut meat.
an area university, the Veterans Administration, community memory For this study the hospital meal was modified by removing the tray and
disorders/geriatrics clinics, and the local chapter of the Alzheimers serving the food in a more restaurant-like environment with all of the
Association. All AD subjects were initially screened for an established food served in dishes sitting on a placemat on the table. The test meal
diagnosis of probable AD according to the NINCDS-ADRDA criteria (Fig. 2) consisted of a salad, a choice of salad dressing packets, roast
[28], scored <5 on the Hachinski Ischemic Scale [29], and met the beef or turkey, gravy in two small souffle cups placed on the plate,
criteria for mild dementia as defined by the Clinical Dementia Rating baked or mashed potatoes, peas, roll, Jello, chocolate chip cookies in
Scale [30]. plastic wrapping, and coffee, tea, or milk. A bowl on the table con-
All subjects underwent additional screening procedures includ- tained packets of salt and pepper and small containers of cream.
ing dental and neurological examinations performed by a university- All participants dined together at a table in a dining room in the
affiliated dentist and neurologist. The purpose of the dental examina- rehabilitation area of a hospital. Two cameras were used: one focused
tion was to rule out any oral pathology that might interfere with the on the subject and one on the subject and his/her partner.
subjects ability to chew and/or manipulate food in the mouth. The Videotapes were analyzed on a playback recorder that allowed
purpose of the neurological exam was to rule out any upper extremity frame-by-frame viewing and was equipped with a second by second
impairment that might interfere with self-feeding and, in the case of the video counter. Tapes were initially viewed by BAP to record the se-
AD subjects, to confirm the diagnosis of AD by ruling out any focal quence of events and to refine data collection techniques. Specific
findings suggestive of vascular dementia. All subjects signed consent self-feeding tasks associated with each of the food or beverage items
forms approved by the university internal review board. available during the meal were printed on a data collection form (Table
2). BAP and a second rater, a graduate research assistant, recorded the
Swallowing Assessment time the behavior occurred, whether the behavior was cued by the
spouse/friend, whether it seemed appropriate, and any additional com-
Subjects were seated in an upright position in the radiology suite for a ments. For condiment-type foods, such as butter and gravy, notation
videofluoroscopic recording of their swallowing. The fluoroscopy was was made as to how they were used (i.e., first cup of gravy on meat
214 B.A. Priefer and J. Robbins: Eating in Alzheimers Disease
Oral transit duration (OTD) Time from initiation of the posterior bolus movement in the oral cavity to arrival of the bolus at the ramus
of the manidible. Posterior bolus movement is a voluntary action under cortical control.
Stage transition duration (STD) Time from arrival of the bolus at the head of the ramus of the mandible to the beginning of maximum hyoid
bone excursion. Hyoid excursion is thought to contribute to the upward and forward movement of the
larynx which contributes to the closing off of the airway during passage of the bolus through the pharynx.
STD ranges from a negative value, indicating the hyolaryngeal excursion begins before the bolus reaches
the ramus of the mandible, to a positive value indicating initiation of excursion after the bolus has reached
the ramus.
Pharyngeal response duration (PRD) Time from the beginning of maximum hyoid excursion to the hyoid returning to rest. This duration
represents the more automatic pharyngeal phase of swallowing. If the pharnygeal response begins after the
bolus reaches the ramus of the mandible (a positive STD value), the bolus may approach an unprotected
airway (larynx and vocal cords open).
Pharnygeal transit duration (PTD) Time from the arrival of the head of the bolus at the ramus of the mandible to the tail of the bolus through
the upper esophageal sphincter (UES).
Total swallow duration (TSD) Time from the beginning of the posterior movement of the bolus in the oral cavity to the tail of bolus
passing through the UES. Once the bolus is through the UES, the esophageal phase of swallowing begins.
Napkin Gravy
Unwraps First cup
Initial placement Second cup
Cutlery Disposition of cups
Initial placement of knife, Butter
fork, spoon First pat
Salad Second pat
Opens lid Third pat
Chooses dressing Disposition of wrapper
Pours dressing Beverages
Disposition of used packet Cream in coffee
First bite Sugar in coffee
Fig. 1. Hospital meal illustrating self-feeding tasks: opening a sil- Roll First sip coffee
veware and napkin package, opening a carton of milk, opening a salad Breaks/cuts/opens Opens milk
dressing packet and pouring dressing on the salad, opening salt, pepper, First bite Opens straw
and cream packets, buttering bread, and using a knife and fork. Potato (baked/mashed) First sip milk
Cuts-if applicable Jello
First bite Leaves in place
Meat Moves to other area of table
Cuts First bite
First bite Cookie
Peas Unwraps
First bite First bite
Disposition of wrapper
perform the directed behavior within 15 sec of the cue with no inter- the unfamiliar taste or texture of the substance resulted in
vening behavior between the administration of the cue and the cued the increased OTD duration.
behavior performed by the subject. For cue [3 to be considered a cue,
the subject had to finish the initiated behavior.
Table 3 illustrates the means, SDs, and signifi-
Assistance was coded as either partial or total assistance as cance of control and AD subject swallow durations for
follows: (1) partial assistanceperforming part of a behavior; (2) total each of the three consistencies. Swallow durations for the
assistanceperforming the entire task for the individual. AD subjects were significantly longer than the control
The two raters independently viewed the videotapes and iden- subjects for cookie OTD, liquid pharyngeal response du-
tified episodes of cueing and assisted behaviors. When the two raters
differed in identification of a cue or assistance, a third person viewed
ration (PRD), and liquid stage transition duration (TSD).
the videotape and rendered a decision. A one-tailed Fisher-Exact test No subjects aspirated any materials. Two control sub-
for proportions was used to test for group differences in frequency of jects and 3 AD subjects demonstrated laryngeal penetra-
cueing and/or assisted eating behaviors. tion defined as material in the laryngeal vestibule above
In addition to assessing for subject eating behaviors that were the vocal cords.
cued and/or assisted, appropriateness of eating behaviors was also as-
sessed. Defining appropriate/inappropriate and normal/abnormal eating
A secondary analysis of the subjects >70 years
behaviors is challenging since raters make decisions about the appro- old was performed since age-related changes in STD,
priateness of eating behaviors based on differing past experiences, PTD, and TSD durations have been reported in individu-
cultural habits, and personal expectations. A methodology is currently als >70 years compared with middle-aged subjects [32].
being tested to define normal/abnormal, acceptable/unacceptable eating This analysis included 8 control (mean age 76 4.3
behaviors.
years) and 5 AD subjects (mean age 76 3.7 years).
Table 4 illustrates that for subjects over 70 years, all
solid (cookie) durations except PRD and two liquid du-
Results rations, OTD and TSD, were significantly more pro-
longed in AD subjects.
Subjects
Fifteen healthy, elderly control and 10 individuals with
mild stage AD participated in the study. The mean age of Self-feeding
the 10 AD subjects was 68 years (SD 4 10) and of the Twenty-four subjects (15 control 9 AD) participated in
15 control subjects, 73 years (SD 4 5). All subjects were the videotaped meal. One AD subject was unable to com-
ambulatory and lived in the community. AD subjects plete the meal portion of the study. This subject, un-
lived with a spouse who participated in answering ques- known to the researchers, developed flu symptoms ear-
tionnaires and in eating the videotaped meal. Control lier in the day and, after starting to eat lunch, felt too ill
subjects were accompanied by a spouse or friend who to continue.
participated in the videotaped meal. Table 5 illustrates the number and type of cues
The majority of subjects retained most of their initiated by subject partners. One cue was directed at a
natural dentition. Control subjects had an average of 27 single control subject whereas a total of 16 cues were
total teeth with an average of 24 natural teeth; AD sub- directed at four different AD subjects with each of these
jects had an average of 26 total teeth with an average of four AD subjects receiving at least two self-feeding cues
20 natural teeth. No subject wore a complete set of den- from their spouses. All of the cues, except the control
tures. With the exception of one control subject judged to subjects cue (eat cookies) and two of the AD cues (drink
have slightly diminished saliva, all subjects were judged milk), focused on behaviors involving the manipulation
by the examining dentist to have normal saliva. One or use of condiments such as the salad dressing or gravy.
control subject was found to have a mandibular tori; no Repeated viewing of the meal videotapes resulted
other oral lesions were identified in any of the subjects. in identification of several instances where both AD and
control subjects imitated the behavior of either their part-
ner or the principal investigator (BAP). That is, the sub-
Swallowing jects observed their partner or the principal investigator
Swallowing data were eliminated for 1 control subject (BAP) in a self-feeding action, and imitated this action
who was found to have a Zenkers diverticulum on vid- within 15 sec of the observation. No intervening action
eofluoroscopy. The duration measures for the cookie occurred between the observed action and the imitated
swallow for 1 AD subject and for the semisolid swallow action. Thus, the subject used the action of another per-
for 1 control subject are not included in the group analy- son as a cue for his or her own subsequent action. For
ses since, for both of these subjects, the oral transit du- example, one woman glanced at her friend who was
ration (OTD) duration was >9 SDs from their respective opening a packet of salad dressing. The subject, who up
group means. Neither of these subjects had extreme du- to that point had already eaten some of her meat, pota-
rations on any of their other swallows. It is possible that toes, and salad without dressing, reached for the bowl
216 B.A. Priefer and J. Robbins: Eating in Alzheimers Disease
Table 3. Control vs. AD subjects by consistency and swallowing du- who had just finished opening her salad dressing packet,
ration put her hand in a position to pour the dressing when she
stopped, appearing confused as to just where she should
Control AD subjects
Swallow subjects (SD) (SD) pour it. She glanced at her husband who was reaching for
duration Consistency (n 4 14) (n 4 10) Significancea butter to butter his roll. She then looked down at her
plate, touched her roll with her other hand, and moved it
OTD Cookie 0.19 (0.19) 0.41 (0.25) p 0.05
Liquid 0.32 (0.19) 0.37 (0.18) NSb
slightly while still holding the salad packet in a ready-
Semisolid 0.44 (0.41) 0.38 (0.39) NS to-pour position. She glanced back at her husband who
STD Cookie 0.24 (0.48) 0.14 (0.32) NS was still preparing to butter his roll. She again looked
Liquid 0.05 (0.17) 0.18 (0.31) NS down at her plate and was just about to pour the salad
Semisolid 0.33 (0.61) 0.53 (0.80) NS dressing on her meat when her husband looked up, saw
PTD Cookie 0.95 (0.50) 0.77 (0.29) NS
Liquid 0.67 (0.18) 0.80 (0.26) NS
what she was about to do, pointed to her salad, and said,
Semisolid 0.90 (0.56) 1.04 (0.79) NS no, on your salad. She then successfully poured the
PRD Cookie 0.90 (0.20) 1.02 (0.42) NS salad dressing on her salad. This subject appeared to
Liquid 0.93 (0.21) 1.10 (0.19) p 0.05 know that she was unsure of what she should do with the
Semisolid 1.10 (0.27) 1.01 (0.25) NS salad dressing and actively, albeit nonverbally, sought
TSD Cookie 1.33 (0.46) 1.56 (0.62) NS
Liquid 1.30 (0.27) 1.65 (0.50) p 0.05
assistance from her husband by observing his behavior
Semisolid 1.85 (0.87) 1.93 (1.02) NS for an action she could imitate or for a cue from him as
to what to do with the dressing.
a
2-sample Wilcoxon test. We have labeled the behavior of subjects who
b
Not significant.
imitated or who sought to imitate the actions of another
as subject-initiated cued behavior as compared with part-
Table 4. Control vs. AD subjects >70 years by consistency and swal- ner-initiated cued behavior, as previously described.
lowing duration Table 6 illustrates the type and frequency of subject-
initiated cued behavior by both control and AD subjects.
Control AD subjects
There is a significant difference in the number of AD
Swallow subjects (SD) (SD)
duration Consistency (n 4 8) (n 4 5) Significancea subject-initiated cued behaviors (78%) as compared with
control subjects (33%) as measured by the Fisher-Exact
OTD Cookie 0.25 (0.21) 0.51 (0.11) p 0.05 test (p # 0.05).
Liquid 0.29 (0.12) 0.46 (0.16) p 0.05
Semisolid 0.50 (0.46) 0.51 (0.63) NSb
STD Cookie 0.02 (0.34) 0.37 (0.22) p 0.05
Liquid 0.01 (0.13) 0.23 (0.36) NS Discussion
Semisolid 0.30 (0.76) 0.52 (0.68) NS
PTD Cookie 0.66 (0.29) 0.99 (0.23) p 0.05 One objective of this study was to design an eating pro-
Liquid 0.62 (0.21) 0.88 (0.31) NS
Semisolid 0.82 (0.68) 0.97 (0.63) NS
tocol to assess both self-feeding and swallowing ability
PRD Cookie 0.96 (0.18) 0.97 (0.13) NS in individuals with AD. Swallowing ability was quanti-
Liquid 1.00 (0.24) 1.12 (0.23) NS tatively assessed by measuring swallowing durations and
Semisolid 1.09 (0.34) 0.93 (0.22) NS by the presence of aspiration/penetration as recorded on
TSD Cookie 1.23 (0.45) 1.86 (0.34) p 0.05 a videofluoroscopic swallow study. Self-feeding ability
Liquid 1.28 (0.26) 1.81 (0.60) p 0.05
Semisolid 1.89 (0.96) 1.96 (0.93) NS
was assessed by a videotape of the subject self-feeding a
noon meal. Though several earlier studies describe eating
a
2-sample Wilcoxon test. problems in late-stage dementia that are suggestive of
b
Not significant. both self-feeding and swallowing difficulties, this is the
first study to concurrently examine both the self-feeding
containing the dressing packets, chose one, opened it, and swallowing components of the eating process, and to
and poured it on her salad. do so in early-stage AD when subjects were able to co-
In general, these imitated or cued actions ap- operate with the study protocol. Assessing eating ability
peared unplanned; the subjects happened to glance at using this protocol posed little difficulty for the subjects
their meal partner or the principal investigator, noticed and their partners and required about 3 h of time. Further
an action, and then imitated the same action. Neither the study with AD subjects in moderate- and severe-stage
partner or principal investigator appeared to recognize AD are needed to determine the feasibility of the proto-
when the subject was imitating their behavior. One ex- col as the disease progresses.
ception was an AD subject who was clinically the most A second objective of this study was to determine
advanced of the early stage AD subjects. The woman, whether the two components of the eating process, self-
B.A. Priefer and J. Robbins: Eating in Alzheimers Disease 217
Type Type
(number of (number of
Subject episodes) Example Subject episodes) Example
7 Verbal (1) Shall we eat our cookies now? 15 Verbal (1) Do you want salad dressing?
(subject picks up and unwraps Partial assist (1) (Hands subject dressing packet, sub-
cookies) ject puts back in dish but chooses
another and goes on to pour
dressing on salad)
20 Verbal (3) Scrape your gravy off your Jello
(subject tries to do so)
Drink your milk (subject drinks
milk)
Do you want Italian or 1000
island? (chooses 1000)
Total assist (3) Removed lid from subjects salad
Poured dressing on subjects salad
Poured gravy on subjects potatoes
22 Verbal (3) Do you want your salad? (sub-
ject opened lid on salad)
Drink your milk (subject drank
milk)
Dont you want your gravy?
(subject eats some gravy)
Nonverbal (2) Points to salad (in conjunction with
verbal cue)
Points to gravy (in conjunction with
verbal cue)
23 Partial assist (1) Wife hands subject dressing packet
(subject proceeds to open)
feeding and swallowing, differ in normal elderly as com- studies, as well as electrical stimulation studies, suggest
pared with individuals with early-stage AD. that areas in the cerebral cortex trigger the initiation of
the swallow [38,39]. Martin and Sessle [38], in summa-
rizing numerous animal and human studies, suggest that
Swallowing the cerebral cortex, in particular the motor and premotor
We hypothesized that swallowing changes begin in the cortex, may initiate and modulate the motor responses
early stage of AD, and we examined this hypothesis by comprising the swallow. Though positron emmision to-
measuring five different swallow durations for three con- mography (PET) studies of individuals with mild to se-
sistenciesliquid, semisolid, and solid. We found sig- vere dementia indicate a relative sparing of the sensori-
nificant group differences in OTD for the solid consis- motor cortex in AD [26], Haxby et al. [40] have dem-
tency and in PRD and TSD for the liquid consistency. onstrated hypometabolism in the premotor cortex by
OTD reflects posterior bolus movement in the mouth and PET scan in individuals with moderate and severe but
is under voluntary control. Prolongation of this duration not mild AD where mild AD was defined as a MMSE
may reflect interference with the initial oral preparation score of >21. Using these investigators MMSE criteria
of the bolus and initiation of the swallow. for mild AD, our mild AD subjects would fall into the
Abnormal or absent chewing movements [11,34] moderate AD category. Therefore, it is possible that our
and abnormal tongue and jaw movement [35] have been AD subjects had some AD involvement of the premotor
reported in individuals with dementia. Since the brain- area. The premotor cortex, an association area thought to
stem is relatively spared in AD [36,37] an explanation be important in the planning, organizing, initiating, and
for oral stage swallowing changes in AD may be found sustaining of motor output [41,42], also has descending
in the cerebral cortex. The cerebral cortex is hypoth- corticospinal and corticobulbar fibers, the latter of which
esized to play a role in the initiation, modulation, and could carry information to the swallowing center in the
integration of swallowing. Animal and human lesion brain stem [41]. Premotor changes due to AD might ini-
218 B.A. Priefer and J. Robbins: Eating in Alzheimers Disease
tially be manifested as slight prolongation or disorgani- of demented nursing home residents, including AD resi-
zation of bolus movement in the mouth resulting in an dents. Although swallow durations were not quantita-
increased oral transit duration. This may be more promi- tively measured in either of these two studies, they both
nent with consistencies that require more oral manipula- suggest that AD subjects do have swallowing impair-
tion such as solids or greater oral control such as liquids. ments, and that the most common appear to occur in the
Several reports exist in the literature of coughing oral phase, in the transition between the oral phase, and
or choking [11,12,34,35] and delayed swallow [11,34] in in the initiation of the patterned pharyngeal response.
late stage dementia. Prolongation of the PRD in the total Neither of these two studies included a control group, but
sample for a liquid swallow and in the STD and PTD for our study, demonstrating differences in swallow dura-
cookie in the >70 group represent pharyngeal delay and tions between control subjects and AD subjects, adds
may place an individual at increased risk for aspiration strength to the existing literature demonstrating swallow-
[43]. Robbins et al. [44] have provided evidence in ing changes in AD prior to late-stage disease.
stroke patients that cortical input plays a role in initiation
of the pharyngeal swallowing response. Our findings
support the notion that premotor cortex involvement, Self-feeding Dependency
secondary to AD, may result in delayed initiation. We hypothesized that self-feeding dependency would be
The findings in our >70-year-old sample are in- present to a greater degree in mild stage AD compared
triguing since, despite small numbers of subjects in each with older controls as determined by the number of part-
group, several significant group differences were found ner-initiated feeding cues or direct assistance provided
that were not present in the total sample group compari- by the partner. We found AD subjects significantly more
sons. This finding may be due to the inclusion of 5 AD likely to receive self-feeding cues or direct assistance
subjects <65 years in the total sample. Gradual changes from their eating partner than control subjects. Self-
in swallow durations with age have been reported to feeding dependency is well documented in late-stage de-
occur sometime after the age of 45 such that individuals mentia [12,34,35,45]. Ours is the first study to document
>70 demonstrate significantly longer swallow durations a significant difference in self-feeding dependency in
than 45-year-olds [32]. It is possible that the younger AD subjects with mild-stage dementia compared with older
subjects in our sample did have prolongation of some adult controls. The partner-initiated cues or assistance
swallow durations but that these longer durations were were directed primarily toward self-feeding tasks related
not apparent when compared with older controls whose to food preparation such as pouring dressing on the salad.
durations were prolonged as a result of age-associated Task performance can be conceptualized as being
changes. That is, in our sample, the younger AD subject relatively automatic or relatively effortful depending on
swallows may be more similar to the older control group the amount of attentional capacity one must use to per-
swallows. form the task [46]. Based on the controlled vs. automatic
Our swallow findings support the findings of two processing theory proposed by Schifrin and Schneider
other studies of swallowing and AD. Horner et al. [19] [47], Jorm [46] has suggested that individuals with de-
performed videofluoroscopic studies on moderate and mentia are likely to retain the ability to perform tasks
severely demented individuals living in both nursing requiring little conscious thought (automatic tasks) for
homes and the community and Feinberg et al. [18] con- longer periods of time than tasks requiring a great deal of
ducted a retrospective chart review of videofluoroscopies conscious thought processing (controlled tasks). Using
B.A. Priefer and J. Robbins: Eating in Alzheimers Disease 219
this automatic vs. controlled framework, Patterson et al. meal and identifying those tasks that are difficult for AD
[48] divided ADL tasks into subtasks based on whether individuals to perform. This technique also provides an
the subtask performance required automatic or controlled opportunity for examining those tasks for which caregiv-
processing. They report that AD subjects required more ers either cue or provide direct assistance to their AD
assistance to perform subtasks categorized as controlled spouse. Little is known about the decision-making pro-
compared with those categorized as automatic ADL sub- cess a caregiver uses in determining when to intervene or
tasks. In the current study, self-feeding tasks directed what type of intervention to use when their AD spouse
toward food preparation (opening a packet of dressing exhibits self-feeding difficulty. This study was an initial
and pouring it on a salad) require more conscious attempt to determine whether self-feeding dependency
thought than the actions of transporting the food into the exists in early AD. A future study with a larger sample,
mouth from the hand or silverware. The majority of part- including both mild and moderate stage AD individuals,
ner-initiated cues/assistance were directed, more con- is needed to categorize the types of self-feeding actions
trolled processing type of self-feeding tasks. that become increasingly dependent on the assistance of
We can describe the partner-initated cues, but we others as the disease progresses. This information will be
cannot attribute motive to these cues. We do not know if valuable in developing appropriate strategies for caregiv-
the partner-initiated cues were based on the partners ers to use in assisting their AD spouses to maximize
perception that the subject needed help, a genuine need independent function and minimize excess disability.
for help by the subject, or a long-standing behavioral trait In summary, this pilot study suggests that eating
of the partner to take-over and do for another. All AD changes, both swallowing and self-feeding, begin early
subjects ate with spouse-partners and 6 of the 14 control in the AD process. The following limitations of this
subjects ate with spouse-partners (9 controls ate with
study must be acknowledged. First, because of strict In-
friend-partners) but only the spouses of subjects initiated
stitutional Review Board scrutiny of protocols including
cues. This leads us to speculate that the cueing initiated
AD subjects, we used radiographic exposure time con-
by these spouses was based on their perception of a need
servatively. We elected to limit the bolus size to 3 ml and
for help regardless of whether the need was genuine.
to have the subject perform only two swallows of each
Our finding of subjects using the actions of others
consistency [53]. Two swallows per consistency allowed
as a cue to initiate an identical action is an example of
little margin for error and when technical difficulties
imitation, a common behavior used by most people at
arose during second swallows, data were lost. In future
one time or other. Imitation is a type of observational
learning which is used extensively in children but also at studies, the swallowing protocol will be modified by de-
all ages of development [49]. Using the behavior of an- creasing the number of consistencies to only liquid and
other as a cue for personal action is a useful learning solid and increasing the number of swallows per consis-
modality and is likely to occur in unfamiliar social set- tency. Second, subjects were videotaped eating a noon
tings when uncertainty exists as to what to do next. This meal in a setting different from their eating environment.
is highlighted by the fact that 33% of the control subjects Partners may have cued AD subjects more frequently
in our study, eating in an unfamiliar social setting, imi- because they did not want their AD spouse to perform
tated the self-feeding actions of another person. All but 2 poorly while being videotaped. Though this is certainly a
of the AD subjects displayed imitating behaviors. This is possibility, it is also important to examine social or pub-
not unexpected since a consequence of AD is increased lic dining behavior, since eating in restaurants or in other
uncertainty and confusion in unfamiliar settings [50]. peoples homes is likely to be a part of the couples past
The fact that these AD subjects were able to engage in experience, and continuation of this activity may depend
imitating behaviors is positive in that the ability to imi- on the eating behaviors of the AD individual and the
tate is a necessary cognitive skill for successful partner- cueing practices of the caregiver. And finally, since sub-
initiated cueing when such cueing is necessary. jects were videotaped eating only one meal, intrasubject
Cueing and graduated direct assistance (partial to variability was not assessed. Albeit videotaping allows
total assistance) are interventions frequently recom- behaviors to be accurately captured for analysis, the
mended to maintain ADL functioning in the AD popu- analysis of these tapes is time intensive [54]. In our
lation [20,51,52]. The appropriate use of cues and direct study, meals (first to last bite) averaged 31 min. Raters
assistance in AD is a challenging issue since cues or spent 48 h per tape coding behaviors. No studies to
direct assistance can either compensate for a pure dis- date, employing videotape technology to assess eating
ability and help to maintain independence or they can behaviors, have provided intrasubject variability data al-
contribute to excess disability by providing assistance though such information is needed.
beyond what is needed. Videotaping is a powerful tech-
nique for examining the subtasks comprising a self-fed This research was conducted at the Wm. S. Middleton Memorial Vet-
220 B.A. Priefer and J. Robbins: Eating in Alzheimers Disease
erans Administration Hospital and supported by the Alzheimers As- tional decline in Alzheimers Disease: a longitudinal study. J Am
sociation/George E. Connors Memorial Pilot Research Grant. This is Geriatr Soc 41:654661, 1993
manuscript publication 97-11 from the Madison VA GRECC. 22. Akerlund BM, Norberg A: An ethical analysis of double-bind
conflicts as experienced by care workers feeding severely de-
mented patients. Int J Nurs Stud 22(3):207216, 1985
References 23. Hu T, Huang L, Cartwright SD: Evaluation of the costs of caring
for the senile demented elderly. Gerontologist 26:158163,
1. Kukull WA, Brenner DE, Spech CE, Nochlin D, Bowen J, Mc- 1986
Cormick W, Teri L, Pfanschmidt ML, Larson EB: Causes of 24. Hotaling DL: Adapting the mealtime environment: setting the
death associated with Alzheimers Disease: impairment before stage for eating. Dysphagia 5:7783, 1990
death. J Am Geriatr Soc 42:723726, 1994 25. Aronson MK, Cox C, Guastadisegni P, Frazier C, Sherlock L,
2. Sitzmann JV, Mueller R: Enteral and parenteral feeding in the Grower R, Sternberg B, Breed J, Koren MJ: Dementia in the
dysphagic patient. Dysphagia 3:3845, 1988 nursing home: association with care needs. J Am Geriatr Soc
3. Bucht G, Sandman PO: Nutritional aspects of dementia, espe- 40:2733, 1992
cially Alzheimers disease. Age Ageing 19:s32s36, 1990 26. Herholz K: FDG PET and differential diagnosis of dementia.
4. Ciocon JO: Indications for tube feedings in elderly patients. Alzheimer Dis Assoc Disord 9:616, 1995
Dysphagia 5:15, 1990 27. Folstein M, Folstein SE, McHugh PR: Mini-mental state: a prac-
5. Litchford MD, Wakefield LM: Nutrient intakes and energy ex- tical method for grading the cognitive state of patients for the
penditures of residents with senile dementia of the Alzheimers clinician. J Psychiatr Res 12:189198, 1975
type. J Am Diet Assoc 87(2):211213, 1987
28. McKhann G, Drachman D, Folstein M, Katzman R, Price D,
6. Singh S, Mulley GP, Losowsky MS: Why are Alzheimer pa-
Stadlan EM: Clinical diagnosis of Alzheimers disease: report of
tients thin? Age Ageing 17:2128, 1988
the NINCDS-ADRDA work group. Neurology 258:34113416,
7. Volicer L, Seltzer B, Rheaume Y, Fabiszewski K, Herz L, Sha-
1983
piro R, Innis P: Progression of Alzheimer-type dementia in in-
29. Lanska DJ, Schoenberg BS: The epidemiology of dementia:
stitutionalized patients: a cross-sectional study. J Appl Gerontol
methodologic issues and approaches. In: PJ Whitehouse (ed.):
6:8394, 1987
Dementia. Philadelphia: F.A. Davis Co, 1993
8. Claggett MS: Nutritional factors relevant to Alzheimers dis-
30. Hughs CP, Berg L, Danziger WL, Coben LA, Martin RL: A new
ease. J Am Diet Assoc 89(3):392396, 1989
clinical scale for the staging of dementia. Br J Psychiatry
9. Fairburn CG, Hope RA: Changes in eating in dementia. Neuro-
140:566572, 1982
biol Aging 9:2829, 1988
31. Logemann JA, Kahrilas PJ: Relearning to swallow after stroke-
10. Morris CH, Hope RA, Fairburn CG: Eating habits in dementia:
application of maneuvers and indirect biofeedback: a case study.
a descriptive study. Br J Psychiatry 154:801806, 1989
Neurology 40:11361138, 1990
11. Siebens H, Trupe E, Siebens A, Cook F, Anshen S, Hanauer R,
32. Robbins JA, Hamilton JW, Lof GL, Kempster G: Oropharyngeal
Oster G: Correlates and consequences of eating dependency in
swallowing in normal adults of different ages. Gastroenterology
institutionalized elderly. J Am Geriatr Soc 34:192198, 1986
103:823829, 1992
12. Volicer L, Seltzer B, Rheaume Y, Karner J, Glennon M, Riley
33. Marascuilo LA, Serlin RC: Statistical Methods for the Social
ME, Crino P: Eating difficulties in patients with probable de-
and Behavioral Sciences. New York: WH Freeman & Co, 1988
mentia of the Alzheimer type. J Geriatr Psychiatr Neurol 2(4):
34. Athlin E, Norberg A, Asplund K, Jansson L: Feeding problems
188195, 1989
in severely demented patients seen from task and relationship
13. Athlin E, Norberg A, Asplund K: Caregivers perceptions and
aspects. Scand J Caring Sci 3(3):113121, 1989
interpretations of severely demented patients during feeding in a
35. Osborn CL, Marshall MJ: Self-feeding performance in nursing
task assignment system. Scand J Caring Sci 4(4):147155, 1990
home residents. J Gerontol Nurs 19(3):714, 1993
14. Athlin E, Norberg A: Interaction between the severely demented
patient and his caregiver during feeding. Scand J Caring Sci 36. Blass JP: Pathophysiology of the Alzheimers syndrome. Neu-
1(34):117123, 1987 rology 43(Suppl 4):S25S38, 1993
15. Davies AM, Snaith PA: Mealtime problems in a continuing-care 37. Henderson VW, Finch CE: The neurobiology of Alzheimers
hospital for the elderly. Age Ageing 9:10015, 1980 disease. J Neurosurg 70:335353, 1989
16. Sandman PO, Adolfsson R, Nygren C, Hallmans G, Winblad B: 38. Martin RE, Sessle BJ: The role of the cerebral cortex in swal-
Nutritional status and dietary intake in institutionalized patients lowing. Dysphagia 8:195202, 1993
with Alzheimers disease and multiinfarct dementia. J Am Geri- 39. Andre J, Car A: Inputs to the swallowing medullary neurons
atr Soc 35:3138, 1987 from the peripheral afferent fibers and the swallowing cortical
17. Kolodny V, Malek A: Improving feeding skills. J Gerontol Nurs area. Brain Res 178:567572, 1979
17(6):2024, 1991 40. Haxby JV, Grady CL, Koss E, Horwitz B, Schapiro M, Fried-
18. Feinberg MJ, Ekberg O, Segall L, Tully J: Deglutition in elderly land RP, Rapoport SI: Heterogeneous anterior-posterior meta-
patients with dementia: findings of videofluorographic evalua- bolic patterns in dementia of the Alzheimer type. Neurology
tion and impact on staging and management. Radiology 38:18531863, 1988
183:811814, 1992 41. Mesulam MM: Patterns in behavioral neuroanatomy: associa-
19. Horner J, Alberts MJ, Dawson DV, Cook GM: Swallowing in tion areas, the limbic system, and hemispheric specialization. In:
Alzheimers disease. Alzheimer Dis Assoc Disord 8:177189, MM Mesulam (ed.): Principles of Behavioral Neurology. Phila-
1994 delphia: FA Davis, 1985, pp 170
20. Gendron M, Levesque L: Evaluating the functional autonomy of 42. Walsh K: Neuropsychology. A Clinical Approach. Edinburgh:
persons with Alzheimers disease: a tool for observing four ac- Churchill Livingtone, 1987
tivities of daily living. Am J Alzheimer Care Rel Dis Res 43. Robbins J, Levine RL: Swallowing after unilateral stroke of the
8:2435, 1994 cerebral cortex: preliminary experience. Dysphagia 3:1117,
21. Green CR, Mohs RC, Schmeidler J, Aryna M, Davis KL: Func- 1988
B.A. Priefer and J. Robbins: Eating in Alzheimers Disease 221
44. Robbins JA, Levine RL, Maser A, Rosenbek JC: Swallowing 49. Flavell JH: Cognitive Development. Engelwood Cliffs: Prentice-
after unilateral cerebral stroke. Arch Phys Med Rehab Hall, Inc, 1985
74:12951300, 1993 50. Hall GR: Care of the patient with Alzheimers disease living at
45. Van Ort S, Phillips L: Feeding nursing home residents with home. Nurs Clin North Am 23:3146, 1988
Alzheimers disease. Geriatr Nurs 13(5):249253, 1992 51. Rogers JC, Snow T: An assessment of the feeding behaviors of
46. Jorm AF: Controlled and automatic information processing in the institutionalized elderly. Am J Occup Ther 36(6):375380,
senile dementia: a review. Psychol Med 16:7788, 1986 1982
47. Shiffrin R, Schneider W: Controlled and automatic human in- 52. Thralow JU, Rueter MJS: Activities of daily living and cognitive
formation processing: II. Perceptual learning, automatic attend- levels of function in dementia. Am J Alzheimers Care Rel Dis-
ing, and a general theory. Psychol Rev 84:127190, 1977 ord 7:1419, 1993
48. Patterson MB, Mack JL, Neundorfer MM, Martin RJ, Smyth 53. Logemann JA: Treatment for aspiration related to dysphagia: an
KA, Whitehouse PJ: Assessment of functional ability in Alzhei- overview. Dysphagia 1:3438, 1986
mer disease: a review and a preliminary report on the Cleveland 54. Phillips LR, Van Ort S: Measurement of mealtime interactions
Scale for Activities of Daily Living. Alzheimer Dis Assoc Disord among persons with dementing disorders. J Nurs Meas 1:4155,
6(3):145163, 1992 1993