Professional Documents
Culture Documents
Adherence To Cardiovascular Therapy: A Meta-Analysis of Prevalence and Clinical Consequences
Adherence To Cardiovascular Therapy: A Meta-Analysis of Prevalence and Clinical Consequences
Received 5 March 2013; revised 17 June 2013; accepted 11 July 2013; online publish-ahead-of-print 1 August 2013
Aims The aim of this study was to determine the extent to which adherence to individual vascular medications, assessed
by different methods, influences the absolute and relative risks (RRs) of cardiovascular disease (CVD) and all-cause
mortality.
.....................................................................................................................................................................................
Methods We performed a systematic review and meta-analysis of prospective epidemiological studies (cohort, nested case–
and results control, or clinical trial) identified through electronic searches using MEDLINE, Web of Science, EMBASE, and Cochrane
databases, involving adult populations (≥18 years old) and reporting risk estimates of cardiovascular medication
adherence with any CVD (defined as any fatal or non-fatal coronary heart disease, stroke or sudden cardiac death)
and/or all-cause mortality (defined as mortality from any cause) outcomes. Relative risks were combined using
random-effects models.
Forty-four unique prospective studies comprising 1 978 919 non-overlapping participants, with 135 627 CVD events
and 94 126 cases of all-cause mortality. Overall, 60% (95% CI: 52– 68%) of included participants had good adherence (ad-
herence ≥80%) to cardiovascular medications. The RRs (95% CI) of development of CVD in those with good vs. poor
(,80%) adherence were 0.85 (0.81– 0.89) and 0.81 (0.76– 0.86) for statins and antihypertensive medications, respect-
ively. Corresponding RRs of all-cause mortality were 0.55 (0.46– 0.67) and 0.71 (0.64– 0.78) for good adherence to statins
and antihypertensive agents. These associations remained consistent across subgroups representing different study char-
acteristics. Estimated absolute risk differences for any CVD associated with poor medication adherence were 13 cases for
any vascular medication, 9 cases for statins and 13 cases for antihypertensive agents, per 100 000 individuals per year.
.....................................................................................................................................................................................
Conclusion A substantial proportion of people do not adhere adequately to cardiovascular medications, and the prevalence of sub-
optimal adherence is similar across all individual CVD medications. Absolute and relative risk assessments demonstrate
that a considerable proportion of all CVD events (9% in Europe) could be attributed to poor adherence to vascular
medications alone, and that the level of optimal adherence confers a significant inverse association with subsequent
adverse outcomes. Measures to enhance adherence to help maximize the potentials of effective cardiac therapies in
the clinical setting are urgently required.
-----------------------------------------------------------------------------------------------------------------------------------------------------------
Keywords Medication adherence † Cardiovascular disease
†
R.C., H.K., and E.H. contributed equally to this work.
* Corresponding author. Tel: +31107043484, Fax: +31107044657, Email: [email protected] (H.K.)/[email protected] (O.H.F.).
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2013. For permissions please email: [email protected]
Adherence to cardiovascular therapy 2941
Introduction search terms related to the exposure (e.g. medication adherence, medi-
cation compliance, medication persistence) and medication groups
Medication adherence is defined as the extent to which a patient (e.g. hydroxymethylglutaryl-CoA reductase inhibitors, antihypertensive
takes medications as prescribed by their healthcare providers.1 Sub- agents, aspirin, adrenergic beta-antagonists, hypoglycaemic agents),
optimal adherence reduces the effectiveness of essential medications which were combined with search terms related to the outcomes (e.g.
and has been highlighted as a significant obstacle in achieving better cardiovascular diseases coronary artery disease, stroke, cerebrovascular,
mortality). We identified articles eligible for further review by performing
patient outcomes.1 – 3 An earlier World Health Organization
an initial screen of identified titles or abstracts, followed by a full-text
report described poor adherence as ‘a worldwide problem of striking
review. Authors of the retrieved papers were contacted directly for add-
magnitude’.4 Poor adherence itself though is a problem that should be
itional tabular data when required. In the case of multiple publications, the
ratios were both derived from the current meta-analysis. For these ana- articles remained for further evaluation. In the full-text assessment,
lyses, it is assumed that all participants within each study who were at high 44 of these articles (Supplementary material online, Appendix S4)
risk of CVD, either without pre-existing vascular disease (primary pre- met our inclusion criteria. Of the 78 excluded articles following full
vention) or with pre-existing vascular disease (secondary prevention), text evaluation, 39 were based on unrelated exposures, 26 reported
were offered preventive therapy. Hazard ratios and odds ratios were
on outcomes other than CVD or all-cause mortality, and 13 were
assumed to approximate the same measure of relative risk (RR, principal
duplicate publications.
summary measure). Summary RRs were calculated by pooling the
study-specific estimates for various vascular medication types (including
statins, antihypertensive, aspirin, and antidiabetic agents) using a Study characteristics
random-effects model that included between-study heterogeneity (par- The characteristics of the studies included are summarized in Table 1
allel analyses used fixed-effect models). Heterogeneity of RR estimates and Supplementary material online, Table S1. Overall, data were avail-
across studies was evaluated by the I 2 statistic.23 The possibility of publi- able on 1 978 919 unique participants with 135 627 CVD and 94 126
cation bias was evaluated using the Begg test24 and visual inspection of all-cause mortality events collected over an average follow-up
funnel plots. Heterogeneity at the level of individual studies was assessed between 1 and 10 years. The average age of the participants was
by comparing results from studies grouped according to pre-specified
63.1 years and 55% of the participants were male. Seventeen
study-level characteristics using the meta-regression technique. These
studies were based in Europe, 21 in North America, 3 in Asia-Pacific,
subsidiary analyses included study characteristics (e.g. baseline popula-
and 3 were conducted in multiple countries. Overall, 25 studies
tion, location, study design, method employed to measure adherence, ad-
herence threshold used and follow-up duration), analytical strategy (e.g. recruited patients from healthcare registers, 9 from insurance data-
adjustment for socioeconomic variables, co-morbidities and use of mul- bases, while 10 involved participants from clinical trial registers. Of
tiple medications, i.e. ‘poly pharmacy’) and outcome types. All statistical these, 26 studies assessed adherence by pharmacy refill data based
tests were two-sided and used a significance level of P , 0.05. All analyses on the MPR(14 studies) or PDC (12 studies); 16 by other indirect
were performed using Stata release 11 (StataCorp, College Station, measures including self-reports (6 studies); and 2 by direct measures
TX, USA). (e.g. electronic monitoring systems or blood tests). The majority of
the studies provided RRs for more than one medication (e.g. separate
RRs for CVD for good adherence to statins and antihypertensives).
Results Approximately half of these studies reported RRs of good adherence
to individual medications, which were adjusted for other concomi-
Literature search tant medications. Among the studies identified, 21 reported solely
A total of 2021 citations were retrieved from the electronic search on CVD outcomes, 11 reported only on all-cause mortality and 12
(Figure 1). After initial screening based on titles and abstracts, 122 reported on both outcomes.
Adherence to cardiovascular therapy 2943
Studies Participants
.......................................... ..........................................
a
n % na %
...............................................................................................................................................................................
Eligible studies
Total unique studies 44 100.0 1 978 919 100.0
Cohort 33 75.0 1 721 351 87.0
Nested case–control 8 18.2 222 160 11.2
a
Results presented are for number of studies or number of participants unless otherwise specified.
2944 R. Chowdhury et al.
Prevalence and determinants of good involving cause-specific CVD events, RRs were broadly similar for
adherence CHD and stroke outcomes for good adherence to statins and antihy-
pertensive agents (Supplementary material online, Figure S6). In 23
Data on the prevalence of good adherence were available in 34
studies with available data on all-cause mortality outcome (533 381
studies (1 230 382 participants). In these studies, the overall preva-
participants and 94 126 mortality events), the combined RR (95%
lence of good adherence to CVD medications was 60% (95% 52–
CI) for good vs. poor adherence to any CVD medication was 0.62
68) (Figure 2, Supplementary material online, Figure S1). The reported
(0.57–0.67) (Figure 4), with an I 2 estimate of 96.1% (Supplementary
proportions of good adherence ranged from 4.9 to 93.3%, across
material online, Figure S7). Corresponding RRs (95% CIs) for all-cause
studies and differed by medication type. Prevalence of good adher-
mortality were 0.55 (0.46–0.67), 0.71 (0.64–0.78), and 0.45 (0.16–
ence was 54% (95% CI: 41–67) for statins (12 studies), 59% (95%
1.29) for good adherence to statins, antihypertensive agents, and
CI 42 –77) for antihypertensives (11 studies), 70% (95% CI 49 –91)
aspirin, respectively (Figure 4 and Supplementary material online,
for aspirin (2 studies), and 69% (95% CI 59 –78) for antidiabetic med-
Figures S8 and S9). There was no association between good adher-
ications (2 studies) (Supplementary material online, Figures S1 and
ence to antidiabetic medications and all-cause mortality (two
S2). The key factors that predicted adherence rates in these studies
studies, data not shown). Findings were broadly similar when using
were age, gender, comorbidity, and polypharmacy (Supplementary
a fixed-effects model as subsidiary analyses.
material online, Table S2).
Absolute risk difference associated with
Good adherence to individual poor adherence
cardiovascular therapy and adverse Using CVD estimates from the studies based on EU nations,21 the
outcomes corresponding absolute risk differences in the number of cases per
Among 33 studies reporting on cardiovascular outcomes (1 615 126 100 000 individuals per year associated with poor adherence were
participants and 135 627 CVD events), the combined RR (95% CI) for 13 cardiovascular cases for any CVD medication, 9 cardiovascular
good adherence compared with poor adherence for any CVD medi- cases for statins, and 13 cardiovascular cases for antihypertensive
cation was 0.80 (95% CI: 0.77– 0.84) (Figure 3), with an I 2 estimate of agents. Similar calculations for the US population are presented in
96.2% (Supplementary material online, Figure S3). Corresponding Supplementary material online, Appendix S5. Assuming the popula-
RRs were 0.85 (0.81–0.89), 0.81 (0.76–0.86), and 0.60 (0.31–1.16) tion prevalence of poor adherence to be 40% among patients who
for good adherence to statins, antihypertensive agents, and aspirin, were prescribed CVD medications, 9.1% of all CVD events were
respectively (Figure 3 and Supplementary material online, Figures S4 attributable to poor adherence.
and S5). Only one study was identified in the literature search that
assessed the association of clopidogrel with CVD independently of Assessment of heterogeneity
aspirin. The risk of CVD events was non-significant for both clopido- and publication bias
grel and aspirin in this study, but insufficient information was available The pooled estimates associated with good adherence to statin and
to include the results in this meta-analysis.25 In subsidiary assessments antihypertensive medications remained largely unchanged when all
Adherence to cardiovascular therapy 2945
Figure 4 Relative risks for all-cause mortality in good vs. poor adherence to major cardiovascular medications.
included studies were grouped by various adherence measurement (Pheterogeneity .0.05, Supplementary material online, Figures S10
methods (Pheterogeneity .0.05 for statins and antihypertensives, and S11). Additionally, estimates were generally similar for study
Figure 5). There was no material difference in the overall RRs characteristics such as age, location, source of participants (trial vs.
among the studies according to the analytical approaches (i.e. healthcare vs. insurance databases) and type of cohort (e.g.
studies that included socioeconomic status, polypharmacy, or co- primary vs. secondary vs. mixed, see Supplementary material
morbid conditions in the multivariate models, vs. the ones that online, Table S2). There was no evidence of publication bias except
did not) and the clinical cut-offs used to define adherence for the studies reporting on adherence to statins and any CVD
2946 R. Chowdhury et al.
outcomes (PBegg ,0.05), which indicated a possible lack of publica- non-adherence have been reported as no-fill of first prescription, ir-
tion of smaller studies reporting null or negative associations regular refills obtained, uncertainty about the effectiveness, prohibi-
between statin adherence and CVD outcomes (Supplementary tive costs, and serious adverse events.2 The proportion of study
material online, Figure S12). participants with good adherence to cardiovascular medications in
this review (based on prospective studies from affluent settings)
was very similar to a previous review,26 which involved principally
Discussion cross-sectional studies in resource-poor settings and reported
Overall, based on available prospective cohort studies, we found that this proportion as 57%. This perhaps reinforces the fact that sub-
a substantial proportion of participants do not adequately adhere to optimal adherence is a problem of global dimensions. The most
cardiovascular medications, and the prevalence of such suboptimal common predictors of poor adherence in this earlier report
adherence was high across all types of CVD medications. Using inci- included poor knowledge, negative perceptions about medication,
dence rates from the general EU population,21 the absolute risk dif- side effects, and cost.26
ference associated with poor medication adherence to CVD Favourable clinical consequences in those with good adherence
medication was 13 per 100 000 CVD deaths per year, and 9% of levels to cardiovascular medications observed in our review may
all CVD cases in the EU could be attributable to poor adherence. Fur- have several explanations. First, it is possible that adherence in
thermore, combining data from 2 million participants indicates that these studies may simply be a surrogate marker of unmeasured con-
good adherence to cardiac therapies could be associated with a 20% founders in these participants, with good adherence reflecting
lower risk of CVD and a 35% reduced risk of all-cause mortality, irre- healthy behaviours or, conversely, co-morbidities such as depression
spective of most clinically relevant patient and study characteristics. contributing to poor adherence in participants.27 – 29 This ‘healthy
Several factors may contribute to the low levels of good adher- adherer effect’ has been examined in a number of studies;
ence to cardiovascular medications observed among the partici- however, their results remain inconclusive. Rasmussen et al.30
pants of the studies that we reviewed. Factors which significantly reported that outcome benefits were mediated principally by cardio-
influenced adherence levels (in a subset of studies with relevant in- vascular medication effects, whereas Curtis et al. 31 found that high
formation) were low social status, low health literacy, existence of adherence in the placebo arm of the Women’s Health Initiative
co-morbid conditions, and polypharmacy (Supplementary material trial was associated with some favourable clinical outcomes including
online, Table S3). Other potential evidence-based promoters of MI and all-cause mortality, yet not CHD death. Our subgroup
Adherence to cardiovascular therapy 2947
analyses comparing studies that adjusted for socioeconomic variables of US participants in the component studies being identified from in-
and co-morbidities with studies that did not, also yielded no material surance registers, therefore, excluding those uninsured in the popu-
difference in estimates. Nonetheless, despite the argued role of be- lation who are expected to receive less timely management of
havioural (or other unmeasured) attributes—the potential clinical medical conditions. As the current review combined only summary
benefits of good adherence to efficacious vascular medications are level data from published studies, we were not able to assess
likely to be substantial and should not be underestimated. For trends in levels of adherence over time. Additionally, as the current
example, it has been estimated that over a period of 4 years of review is based on published reports, our results may have been
statin use, a reduction of 1 mmol/L (39 mg/dL) in the level of low- affected by selective underreporting of adherence, publication bias
density lipoprotein cholesterol translates into a 13% reduction in or inability to control for all relevant covariates in a consistent way.
Authors’ contributions 16. Naderi SH, Bestwick JP, Wald DS. Adherence to drugs that prevent cardiovascular
disease: meta-analysis on 376 162 patients. Am J Med 2012;125:882 –887.e1.
17. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic
O.H.F., H.K., and R.C. conceived the study. R.C., H.K., E.H., A.S., S.F., reviews and meta-analyses: the PRISMA statement. BMJ 2009;339:b2535.
and O.H.F. did the literature searches. H.K. and E.H. did the data ex- 18. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, Moher D,
traction and analyses. R.C., H.K., E.H., and O.H.F. wrote the manu- Becker BJ, Sipe TA, Thacker SB. Meta-analysis of observational studies in epidemi-
ology: a proposal for reporting. Meta-analysis of Observational Studies in Epidemi-
script. R.C., O.H.F., E.H., H.K., A.S., S.F., C.M., S.M., J.M., B.S., and ology (MOOSE) group. JAMA 2000;283:2008 –2012.
A.H. contributed to the critical revision of the manuscript before 19. Peterson AM, Nau DP, Cramer JA, Benner J, Gwadry-Sridhar F, Nichol M. A checklist
publication. R.C., H.K., and E.H. contributed equally to the study. for medication compliance and persistence studies using retrospective databases.
Value Health 2007;10:3 –12.
O.H.F. is the guarantor. 20. Chêne G, Thompson SG. Methods for summarizing the risk associations of quanti-