Interventions to improve medication adherence in coronary disease patients: A systematic review and meta-analysis of randomised controlled trials

Abstract

Background

Adherence to multiple cardiovascular (CV) medications is a cornerstone of coronary heart disease (CHD) management and prevention, but it is sub-optimal worldwide. This review aimed to examine whether interventions improve adherence to multiple CV medications in a CHD population.

Design

This study was based on a systematic review and meta-analysis according to Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.

Methods

Randomised controlled trials were identified by searching multiple databases and reference lists. Studies were selected if they evaluated interventions aiming to improve adherence to multiple CV medications targeting a CHD population and if they provided an appropriate measure of adherence. Interventions were classified as complex or simple interventions. Odds ratios (ORs) were calculated and pooled for a meta-analysis. Risk of bias, heterogeneity and publication bias were also assessed.

Results

Sixteen studies (10,706 patients) were included. The mean age was 62 years (standard deviation (SD) 3.6) and 72% were male. In a pooled analysis, the interventions significantly improved medication adherence (OR 1.52; 95% confidence interval (CI) 1.25–1.86; p < 0.001) and there were no significant differences based on intervention type (complex vs simple), components categories and adherence method. There was moderate heterogeneity (I²= 61%) across the studies. After adjusting for publication bias, the effect size was attenuated but remained significant (OR 1.35; 95% CI 1.09–1.68).

Conclusion

Interventions to improve adherence to multiple CV medication in a CHD population significantly improved the odds of being adherent. Simple one-component interventions might be a promising way to improve medication adherence in a CHD population, as they would be easier to replicate in different settings and on a large scale.

Keywords

Medication adherence medication compliance coronary disease multiple medications polypharmacy systematic review meta-analysis randomised controlled trial

Introduction

Medication adherence is defined as the process by which the patients take their medication as prescribed by their healthcare professionals.^1,2 Although medication adherence is the preferred term,¹ other terms such as compliance or concordance have been commonly used in the literature. Adherence to a combination of long-term cardiovascular (CV) medications is a cornerstone of coronary heart disease (CHD) management and prevention. However, adherence to these medications is sub-optimal worldwide.^3–5 It is known that medication adherence decreases with time⁶ and a recent meta-analysis found that only 60% of nearly two million participants had an overall good adherence (adherence ≥ 80%) to CV medications.⁷ Therefore, to improve adherence to evidence-based CV medications and achieve the subsequent reduction in morbidity, mortality and costs, there is a need for interventions that improve the medication-taking behaviour of CHD patients.

CHD is a singular medical condition and it can have positive or negative effects in the patients’ life.⁸ While a fear of a heart attack may have a positive effect on medication adherence, the complexity of guideline-recommended CV medication regimens is a well-recognised barrier to adherence for patients with CHD.⁹ Previous reviews assessing interventions to improve medication adherence are limited in terms of addressing these specific details related to CHD. Reviews have tended to include a broad patient population with a variety of medical conditions^10,11 or patients with CV disease but with a focus on hypertension and heart failure.^12,13 In addition, previous reviews have focussed on one specific class of CV medication, such as lipid-lowering,¹⁴ or on one specific type of intervention such as multi-professional interventions¹⁵ or reminder packaging.¹⁶ To our knowledge, no previous review has evaluated interventions to improve adherence to multiple CV medications in a CHD-specific population.

Our primary aim was to examine whether interventions improved patients’ adherence to multiple CV medications, by reviewing randomised controlled trials (RCTs) that assessed any type of intervention aimed at improving adherence in a CHD population compared to usual care or no intervention. In addition, we aimed to determine which types of intervention components were present in the interventions; whether the interventions improved clinical outcomes; whether they were cost-effective and the degree of patient’s satisfaction with the interventions.

Methods

Protocol and registration

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines and followed the methods pre-specified and documented in a protocol. The review was registered in the International prospective register of systematic reviews (PROSPERO) Registry CRD42015016694.

Study inclusion criteria

Randomised controlled trials with individual or cluster randomisation were considered for inclusion if the study population comprised adults (over 18 years) with a diagnosis of CHD, including previous acute coronary syndrome (myocardial infarction and unstable angina), stable angina, percutaneous coronary intervention (PCI) or coronary artery bypass graft surgery (CABG). Studies were excluded if some participants in the study did not have an established diagnosis of CHD, for example studies that comprised a mixed population including participants with established CHD and participants with an estimated high risk of CV diseases (≥15%), and if they involved psychiatric and institutional patients (to avoid potential external controls over adherence). Studies were eligible if the main objective of the intervention was to improve patient’s adherence to multiple CV medications or if a broader intervention, such as in cardiac rehabilitation programmes, had a core component focused on improving medication adherence. Only studies with interventions targeting patients and consumers were included. Studies investigating interventions to improve adherence to only one specific class of CV medication, e.g. statins, were excluded as it is recommended that CHD patients take a combination of CV drugs concomitantly.^17–19

Medication adherence is defined as the extent to which patients take medications as prescribed, therefore we only included studies that provided sufficient information about the extent to which patients were adhering to their drug regimen (e.g. whether there were any missed doses). We excluded studies that reported only prescription rates (e.g. whether the patient was prescribed a particular medication by their physician) and medication use rates (e.g. a Yes/No type of question about whether the patient is taking the prescribed medication). We included studies using any method of adherence, including indirect (e.g. pill count, prescription refill rate, electronic monitoring), direct (tracer substances in blood or urine) and subjective measures (e.g. self-report questionnaires). Other proxy measures for adherence included physiological indicators (e.g. cholesterol, blood pressure), health outcome indicators (e.g. quality of life, morbidity, mortality), adverse effects and cost evaluations, if reported in association with medication adherence outcomes in the main paper.

Search strategy

Five databases (Medline, CENTRAL, PsycINFO, EMBASE, CINAHL) were searched on 6 February 2015 and re-searched on 21 July 2015 with no language restriction. A date limit (1 January 1995 to year 2015) was applied to ensure that only contemporary evidence was reviewed. The database search strategy was developed in collaboration with a librarian. Search terms included patient compliance, treatment refusal, adherence, non-adherence, coronary disease, myocardial ischemia, myocardial infarction, percutaneous coronary intervention and coronary artery graft bypass (Supplementary Material, S1). Appropriate RCT filters were used to maximise identification of randomised controlled trials. Additional articles were obtained through manual search of reference lists and by contacting experts in the field.

Study selection

One author (KS) carried out the database searches. Two authors (KS, SK) independently screened all titles and abstracts. Full-text articles of potentially relevant studies were assessed independently by the same authors using a pre-determined eligibility assessment form. A study was included if it met all inclusion criteria. Any disagreement was discussed and resolved by consensus; and a third reviewer (JR) was consulted in case of an unresolved disagreement. A reference management system (Endnote X6 - Thompson Reuters) was used to identify and remove duplicate references, which was verified by manual search.

Assessment of risk of bias

Assessment of risk of bias of the included studies was done independently by two authors (KS, SK) using the Cochrane ‘Risk of Bias’ tool.²⁰ The main sources of systematic bias in trials were assessed: selection bias, performance bias, detection bias, attrition bias and reporting bias. Random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data and selective reporting were coded into low, high or unclear risk of bias. The method used to measure adherence was also considered, as some methods are more likely to be biased than others.

Data extraction

The relevant data from the included articles were extracted using a pre-designed electronic spreadsheet that was developed and test-piloted for this review. One author (KS) extracted the data and a second author (SK) checked it for accuracy. Extracted data included study characteristics (design, date, aim, sample size, country, recruitment, inclusion and exclusion criteria), participants’ characteristics (CHD diagnosis, age, gender), intervention characteristics (intervention and comparator descriptions, number of intervention groups, CV medication targeted, mode of delivery, professional delivering the intervention, number and time points of intervention delivery), adherence measures (number, type and definition of adherence outcome), adherence outcomes (outcome data and statistical significance, time point reported) and other outcomes (type and definition of outcome, outcome data and statistical significance). When study results were reported at more than one time point, the last time point results were recorded. When two or more articles reported results from the same study, both articles were considered together for more complete data extraction. Authors were contacted for missing, incomplete or unclear data.

Two authors (KS, SK) independently categorised the interventions as broader or specific, and complex or simple using the following definitions. A broader intervention was defined as an intervention with a concomitant focus on lifestyle improvement, such as smoking cessation, regular physical activity and healthy diet; while a specific intervention was focused only on improving medication adherence. An intervention was considered complex if it comprised two or more components, while a simple intervention comprised only one component. Intervention components were classified into categories adapted from previous medication adherence Cochrane reviews^14,21 (Table 1).

Table 1.

Categories of interventions’ components.

Component category	Definition
Patient education	Provision of any verbal, written or visual information and educational materials in either individual appointments or group sessions.
Counselling	Discussions between patients and healthcare professionals about importance of medication, side effects, and barriers to adherence including behavioural change techniques such as motivational interviewing and cognitive behaviour therapy.
Intensified patient care	Regular contact between patients and health professionals to reinforce and/or monitor adherence either through face-to-face consultations or phone calls.
Medication aids	Provision of medication cards (card containing information on medication name, dosage and time of intake), pillboxes and special pill packaging to patients.
Simplification of drug regimen	Decrease in number of doses or number of pills taken per day.
Reminders	Reminders to support and reinforce adherence to medication, including reminders delivered by mail, phone calls, automated voice messaging and text-messaging.
Financial incentives	Provision of financial rewards or reduction of co-payments for taking medication.
Collaborative care	Two or more health professionals communicating and collaborating in patient care (e.g. pharmacist making recommendations to clinicians).
Lay health mentoring	Individual or group support to patients provided by lay people.
Direct observation treatment	Administration of medication observed by a health professional or family members.

Analysis and synthesis

Overall study and participant characteristics were summarised as means or medians for continuous data and as frequencies and proportions for categorical data. We used a software program (Comprehensive Meta-analysis, Version 3.3.070) to pool the data for a meta-analysis. We calculated the odds ratio (OR) of being adherent with 95% confidence interval (CI) for each study and used a random-effects model to pool the estimates. We used the overall adherence to all CV medications measured in the study to calculate the OR. For studies that did not report an overall adherence, we combined the adherence to each individual drug to get an overall estimate of adherence, which is a recommended approach when multiple outcomes are reported within a study.²² For studies with multiple groups, we computed a summary effect for the groups with active interventions and analysed the data in a single pair-wise comparison (intervention vs control), which is also the recommended approach.²³ We assessed heterogeneity between the studies by calculating the I² statistic. An I² statistic exceeding 50% with p-value < 0.05 was interpreted as representing substantial heterogeneity.²⁴ We assessed publication bias by using the funnel plot symmetry and Egger regression intercept. An asymmetric funnel plot with an Egger test p-value < 0.05 was interpreted as indicative of publication bias. We used the trim-and-fill method by Duval and Tweedie to impute the missing studies.²⁵ To investigate heterogeneity, we performed subgroup analysis based on intervention type (complex vs simple), components categories (only for components that were present in more than half of the included studies) and type of adherence methods (self-report vs other more objective methods). We also performed sensitivity analysis excluding studies assessed with low quality (defined as having a high or unclear risk of bias in any of the source of bias criteria, excluding blinding of participants).

There is no gold standard for measuring medication adherence and defining a patient as adherent. Being adherent was defined as per the definition used in each individual trial. For studies that reported more than one type of adherence method, we used the most objective measure, according to the pre-defined hierarchy: medication event monitoring system (MEMS), prescription refills, pill count and self-report. If studies reported more than one self-report method, we used the measure that provided the percentage of doses taken. We used results presented in either continuous or dichotomous scales, as this is acceptable approach²⁶ and has been done by others.^27,28 For studies that reported the results in both types of scale, we selected the continuous scale over the dichotomised scale.

Results

The search retrieved 4766 references from all sources (Figure 1). After exclusion of duplicates, 3758 titles and abstracts were screened and 3670 were excluded. Eighty-eight full-text articles were assessed for eligibility and 71 articles were excluded for several reasons detailed in Figure 1. Seventeen articles reporting 16 studies that met all the eligibility criteria were included in this review.

Figure 1.

Flowchart of selection of included articles. CV: cardiovascular.

Characteristics of included studies

Study characteristics are presented in Table 2. All studies included in this review were published in English. The median sample size was 184 participants (range 15–5855 participants), with a total number of 10,706 participants included in this review. Fifteen trials included mostly males (72% of participants) and mean age was 62 years (SD 3.6). One trial did not report gender, age and other baseline characteristics.²⁹ One trial was a cluster RCT³⁰ and 15 trials were RCTs. Ten trials had an in-hospital recruitment,^29,31–39 while six trials recruited patients in an outpatient setting.^30,39–44 The median length of follow-up was 227 days (range 3–394 days).

Table 2.

Study characteristics and adherence measures.

First author, year	Country and design	Sample size and duration	CHD diagnosis	Age in years (mean ± SD/median -IQR) and gender	Type of adherence method – adherence measure (adherence definition or questionnaire)
Bond et al., 2007⁴⁰	England RCT	1493 12 months	Stable angina, previous MI, PCI and/or CABG	IG: 68.7 ± 9.2 CG: 68.8 ± 9.1 M 68%	Self-report – median score (study questionnaire)
Calvert et al., 2012³¹	USA RCT	143 6 months	ACS, prior PCI or CABG, ≥50% stenosis on angiogram	IG: 63 (54–71) CG: 62 (52–70) M 64%	Self-report-% adherent patients (4-item MMAS) Prescription refills-% adherent patients (PDC^a ≥75%)
Castellano et al., 2014⁴¹	Argentina, Italy, Spain, Paraguay RCT	695 9 months	MI	64.01 ± 11.28 M 80%	Self-report + pill count-% of adherent patients (follow-up visit attendance + 4-item MMAS score = 20 + pill count^b > 80–110%)
Choudhry et al., 2011³⁰	USA Cluster-RCT	5855 394 days median duration	MI	IG: 53.6 ± 7.6 CG: 53.7 ± 7.6 M 75%	Prescription refills – mean MPR^c and % adherent patients (MPR^c >80%)
Gould et al., 2011²⁹	USA RCT	155 72 hours	ACS with PCI	Not provided	Self-report-% adherent patients (Machtinger and Bansberg Visual Analog Scale ≥ 90% of doses taken) and mean score (4-item MMAS)
Gujral et al., 2014³²	Australia RCT	200 12 months	MI	IG: 58.4 ± 11.3 CG: 60.4 ± 11.0 M: 79%	Self-report-% non-adherent patients (MARS) Prescription refills-% non-adherent patients (MPR^c <20%)
Ho et al., 2014³³	USA RCT	253 12 months	ACS	IG: 64 ± 8.57 CG: 63.8 ± 9.25 M 98%	Prescription refills-% adherent patients (PDC^a ≥80%)
Jiang et al., 2007³⁴	China RCT	167 6 months	Angina or MI	IG: 62.11 ± 7.44 CG: 61.37 ± 7.61 M: 71%	Self-report – mean score (study questionnaire)
Khonsari et al., 2015³⁵	Malaysia RCT	62 8 weeks	ACS	57.9 ± 12.64 M 86%	Self-report-% adherent patients (8-item MMAS score = 8)
Kripalani et al., 2012⁴²	USA RCT	440 12 months	Previous MI, PCI or CABG, CHD on angiogram or positive non-invasive stress test	63.7 ± 10.4 M 45%	Prescription refills-% adherent patients (CMG^d <0.20)
Lourenço et al., 2014⁴³	Brazil RCT	126 2 months	Angina or MI	62 ± 9 M: 65%	Self-report – mean % of doses taken (Percentage of adherence measure), % of adherent patients (global adherence evaluation and dose adequacy measure), mean score (4-item MMAS)
Ma et al., 2010³⁶	USA RCT	689 12 months	CHD defined as ≥50% stenosis on angiogram	IG: 60.38 ± 10.5 CG: 60.29 ± 10.4 M: 60%	Prescription refills – mean CMA^e
Ostbring et al., 2014⁴⁴	Sweden RCT	23 6 weeks	CHD (no clear definition)	IG: 70 CG: 62 M: 61%	Self-report-% of adherent patients (8-item MMAS = 8)
Park et al., 2014³⁷	USA RCT	90 1 month	MI and/or PCI	59.2 ± 9.4 M: 76%	MEMS – mean % of doses taken, mean doses taken, mean % of correct number of doses taken, mean % of doses taken on schedule Self-report – Mean score (8-item MMAS)
Polack et al., 2008³⁸	Canada RCT	15 6 weeks	MI	IG-Pre: 59.2 ± 10.71 IG-Post: 67.8 ± 9.98 CG: 65.4 ± 7.09 M: 73%	Self-report-% of non-adherent patients (4-item MMAS < 4)
Rinfret et al., 2013³⁹	Canada RCT	300 12 months	PCI	64 ± 10 M 73%	Prescription refills – median PDC^a

ACS: acute coronary syndrome; CABG: coronary artery bypass surgery; CG: control group; CHD: coronary heart disease; CMA: continuous multiple-interval; CMG: cumulative medication gap; IG: intervention group; IQR: interquartile range; M: male; MARS: Medication Adherence Report Scale; MEMS: medication event monitoring system; MI: myocardial infarction; MMAS: Morisky Medication-taking Adherence Scale; MPR: medication possession ratio; PCI: percutaneous coronary intervention; PDC: proportion of days covered; RCT: randomised controlled trial; SD: standard deviation.

Proportion of days covered ((number of days of medication supply available/total number of days during study period) × 100).

Pill count ((no. of pills dispensed–no. of pills returned)/no. of pills prescribed × 100).

Medication possession ratio ((number of days of medication supply available/days of eligibility for the medication) × 100).

Cumulative medication gap (number of days of medication supply not available/days of eligibility for the medication).

Continuous multiple-interval (number of days of medication supply obtained/total number of days between refill records).

In terms of types of interventions, two trials investigated a broader intervention with a concomitant focus on medication adherence and lifestyle improvement,^34,40 while the rest of the trials tested specific interventions only aimed at improving medication adherence. In terms of modes of delivery, the interventions were delivered by pharmacists,^{31,33,36,38,40,42,44} by nurses,^29,34,39 by researchers⁴³ and by two health professionals.^32,35 Four trials did not require dedicated personnel to deliver the intervention.^30,35,37,41 A detailed description of the interventions is presented in the Supplementary Material, S2. In terms of intervention components, the component categories delivered in each study are presented in Table 3. Thirteen trials were considered complex interventions comprising two or more components, while three trials had simple interventions with only one component.^30,35,41

Table 3.

Intervention components of included studies.

Study author, year	Patient education	Counselling	Intensified patient care	Medication aids	Simplification of drug regimen	Reminders	Financial incentives	Collaborative care
Bond et al., 2007⁴⁰	✓		✓					✓
Calvert et al., 2012³¹	✓	✓	✓	✓				✓
Castellano et al., 2014⁴¹					✓
Choudhry et al., 2011³⁰							✓
Gould et al., 2011²⁹	✓			✓
Gujral et al., 2014³²		✓	✓					✓
Ho et al., 2014³³	✓	✓	✓	✓		✓		✓
Jiang et al., 2007³⁴	✓	✓	✓
Khonsari et al., 2015³⁵						✓
Kripalani et al., 2012⁴²			✓	✓		✓
Lourenço et al., 2014⁴³		✓	✓
Ma et al., 2010³⁶	✓	✓	✓	✓				✓
Ostbring et al., 2014⁴⁴	✓	✓	✓					✓
Park et al., 2014³⁷	✓					✓
Polack et al., 2008³⁸	✓	✓						✓
Rinfret et al., 2013³⁹	✓	✓	✓

Lay health mentoring and direct observation treatment categories were not used in any intervention.

In terms of adherence measures, the majority of studies reported self-report measures, while seven used prescription refills, one used pill count³⁹ and one used MEMS (Table 2).³⁵ Trials reported adherence in either a continuous and dichotomous scale. The majority of the trials reported a threshold of 75–80% to be considered as adherent.³⁵ Half of the trials reported overall adherence,^{34,35,38,40-44} four trials reported adherence to individual classes of medications^32,36,37,39 and four trials reported both.^29–31,33

Effects of interventions

Medication adherence

In a pooled analysis of 10,706 patients, the delivery of an intervention, irrespective of its type, significantly improved medication adherence (OR 1.52; 95% CI 1.25–1.86; p < 0.001) (Figure 2). The interventions were similarly effective when the analysis was restricted to only high quality studies (OR 1.70; 95% CI 1.20–2.41; p = 0.003) (Supplementary Material, S3). There was moderate heterogeneity (I²= 61%; p = 0.001) across the studies. We investigated the heterogeneity by conducting subgroups analysis in which we did not find any significant differences based on intervention type, components categories and type of adherence method (Table 4). Publication bias was detected by funnel plot asymmetry (Supplementary Material, S4), however the Egger regression coefficient was not significant (one-tailed p = 0.06), possibly due to low power. Using the trim-and-fill imputation for missing studies, the adjusted effect size was attenuated but remained significant (OR 1.35; 95% CI 1.09–1.68).

Figure 2.

Meta-analysis of the effect of interventions on medication adherence. CI: confidence interval.

Table 4.

Effects of the interventions on medication adherence in sub-groups stratified by intervention type, components categories and type of adherence method.

Variable	No. of studies	Odds ratio (95% CI)	P-value for heterogeneity
Intervention type
Complex	13	1.50 (1.14–1.98)	0.60
Simple	3	1.74 (1.10–2.76)
Education component
Yes	10	1.56 (1.24–1.96)	0.81
No	6	1.51 (1.07–2.13)
Counselling component
Yes	9	1.64 (1.23–2.20)	0.48
No	7	1.42 (1.07–1.88)
Intensified patient care component
Yes	10	1.44 (1.09–1.91)	0.33
No	6	1.86 (1.22–2.84)
Adherence method
Self-report	7	1.61 (1.08–2.39)	0.82
Others*	9	1.52 (1.17–1.98)

Pill count, prescription refills and medication event monitoring system (MEMS).

Other outcomes

Few studies reported other outcomes. For clinical outcomes, there was no statistically significant difference in three trials,^33,36,41 while one trial reported greater reduction in total and low-density lipoprotein (LDL)-cholesterol.³⁴ In four trials, there was no significant difference in CV events, re-hospitalisation and death,^33,35,39,41 while one study showed significant reductions in major vascular events or revascularisation in the intervention group.³⁰ In terms of costs, two trials reported no statistical difference in total costs between groups,^30,33 while one trial reported higher costs due to pharmacists’ training in the intervention group.⁴⁰ In terms of patient’s satisfaction, two trials reported high satisfaction with pharmacist and nurse interventions,^29,40 while two trials reported that a text-messaging intervention was useful and helped the majority of intervention patients to take their medication.^35,37

Assessment of risk of bias

Overall, the majority of studies was classified as having low or unclear risk of bias across all main sources of bias, not taking into account blinding of participants and personnel as this was impractical due to the nature of the interventions (Figure 3 and Supplementary Material, S5). Eleven trials provided information about adequate random sequence generation and nine adequate allocation concealment. Eleven trials reported blinded outcome assessors or used objective measures of adherence (prescription refills and MEMS) and hence, were classified as having low risk for blinding of outcome assessment. In terms of incomplete outcome data, seven trials were classified as unclear risk, raising a concern about trial results being biased; while all included trials, except one, had low risk of bias in selective reporting.

Figure 3.

Risk of bias assessment summary.

Discussion

In this review, we assessed 16 studies that investigated the effectiveness of interventions to improve adherence to multiple CV medications in a CHD population. We found that the interventions significantly improved medication adherence by increasing the odds of being adherent to the CV medications by 50%. This means that if we assume that 60% of the patients are adherent to CV medications⁷ then implementing an intervention, irrespective of its type, could increase the percentage of adherent patients to around 70%. We also found that the interventions were diverse and that the majority were complex, comprising two or more components. Due to this diversity, our review was unable to ascertain which types of interventions would definitely improve adherence, however the subgroup analysis suggest that there is no difference between complex and simple interventions, as well as between interventions that include patient education, counselling and intensified patient care components compared with those interventions without these components.

A previous Cochrane review also found that interventions to improve adherence to medications in any medical condition are mostly complex.¹⁰ The disadvantages of complex interventions are that they are difficult to compare and be replicated, thereby possibly increasing costs and making implementation in routine practice difficult. Our results showed that simple interventions might be as effective as more complex ones. In our review, three trials used simple interventions by delivering text-message reminders, simplifying the drug regimens using a polypill and by using co-payments to reduce the burden of medication costs in the patients’ life. Other reviews have found that text-messages^27,45,46 and polypills⁴⁷ can improve adherence in a wide range of diseases. An interesting common factor between these three interventions is that they do not require dedicated personnel to deliver them. Cutrona et al. have previously reported that person-independent interventions were successful in improving adherence to CV medication.⁴⁸ These findings highlight that simple interventions not requiring dedicated personnel to deliver them might be a promising way to improve medication adherence in a CHD population, as these interventions are easier to replicate in different settings and on a large scale. Future studies comparing the effects of complex and simple interventions on adherence will help confirm this hypothesis.

Importantly, medication adherence in a CHD population must be viewed within a broader context of CV risk management that includes a concomitant focus on lifestyle changes, such as smoking cessation, regular physical activity and a healthy diet. In this review, we included only two studies that investigated interventions classified as broader interventions. Although cardiac rehabilitation programmes generally aim to improve medication adherence in a CHD population, few studies report a core component of the programme dedicated to achieve this goal and the majority do not measure and report the adherence outcome appropriately. The results of this review emphasise that cardiac rehabilitation and prevention programmes should include a dedicated strategy for medication adherence promotion, meanwhile promoting lifestyle changes.

Our review has identified important limitations of medication adherence research in a CHD population. The lack of a gold standard method contributes to the lack of uniformity in measuring adherence, resulting in difficulty in assessing and comparing different interventions. In this review, studies reported different methods for measuring adherence, therefore to pool the data we had to combine these different methods, always considering the most objective measure reported according to our previously pre-specified hierarchy. In addition, international guidelines^17–19 recommend that patients with CHD adhere to multiple CV medication concomitantly. However, in our review, some studies only measured adherence to individual classes of CV medication, which were combined into an overall adherence to all CV medication in our meta-analysis. It is important to emphasise that future studies should aim to improve the overall adherence to all CV medication, as an overall improvement is more likely to be associated with clinical and health benefits than improvements in adherence to aspirin only, for example.

We also acknowledge that our results might have been limited by our rigorous inclusion criteria and the low quality of more than half of the trials. In addition, our review was not able to demonstrate that the significant improvement in medication adherence would be translated into concurrent benefits in clinical and health outcomes, as these outcomes were only reported in a few of the included studies. Therefore, we recommend that future medication adherence research always includes clinical measures to ensure this translation from medication adherence into clinical outcomes.

Conclusion

Interventions to improve adherence to multiple CV medication in a CHD population significantly improved the odds of being adherent. The interventions were diverse and mostly complex, however the complex interventions were found to have similar results when compared to simple one-component interventions. These simple interventions might be a promising way to improve medication adherence in a CHD population, as they would be easier to replicate in different settings and on a large scale. However, benefits in clinical outcomes and cost-effectiveness need to be determined by future research.

Footnotes

Acknowledgments

The authors acknowledge the role of Jeremy Cullis at the University of Sydney who contributed to the design of the search strategy, and Laurent Billot at The George Institute for Global Health who provided guidance and advice in the statistical analysis. Information about this work will be presented as a poster at the World Congress of Cardiology and Cardiovascular Health on 6 June 2016 in Mexico City, Mexico.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: KS is funded by a University of Sydney International Postgraduate Research Scholarship. JT is funded by a University of Sydney Australian Postgraduate Award. RW is funded by a National Heart Foundation Post-doctoral Fellowship (APP100493). CKC is funded by a Career Development Fellowship co-funded by the National Health and Medical Research Council (NHMRC) and National Heart Foundation, and a Sydney Medical Foundation Chapman Fellowship (APP1033478). JR is funded by a Career Development and Future Leader Fellowship co-funded by the NHMRC and the National Heart Foundation (APP1061793). JC is an investigator on NHMRC programme grant (ID1052555).

References

Vrijens

De Geest

Hughes

. A new taxonomy for describing and defining adherence to medications. Br J Clin Pharmacol 2012; 73: 691–705.

Osterberg

Blaschke

. Adherence to medication. N Engl J Med 2005; 353: 487–497.

Bowry

Shrank

Lee

. A systematic review of adherence to cardiovascular medications in resource-limited settings. J Gen Intern Med 2011; 26: 1479–1491.

Bryson

Rumsfeld

. Medication adherence: Its importance in cardiovascular outcomes. Circulation 2009; 119: 3028–3035.

Hamood

Green

. Determinants of adherence to evidence-based therapy after acute myocardial infarction. Eur J Prev Cardiol 2016; 23: 975–985.

Nieuwlaat

Schwalm

Khatib

. Why are we failing to implement effective therapies in cardiovascular disease? Eur Heart J 2013; 34: 1262–1269.

Chowdhury

Khan

Heydon

. Adherence to cardiovascular therapy: A meta-analysis of prevalence and clinical consequences. Eur Heart J 2013; 34: 2940–2948.

Smith

Fortune

Phillips

. UPBEAT study patients' perceptions of the effect of coronary heart disease on their lives: A cross-sectional sub-study. Int J Nurs Stud 2014; 51: 1500–1506.

Choudhry

Fischer

Avorn

. The implications of therapeutic complexity on adherence to cardiovascular medications. Arch Intern Med 2011; 171: 814–822.

10.

Nieuwlaat R, Wilczynski N, Navarro T, et al. Interventions for enhancing medication adherence. Cochrane Database Syst Rev 2014; (11): Art.No.: CD000011.

11.

Kripalani

Yao

Haynes

. Interventions to enhance medication adherence in chronic medical conditions: A systematic review. Arch Intern Med 2007; 167: 540–550.

12.

Van Dalem

Krass

Aslani

. Interventions promoting adherence to cardiovascular medicines. Intl J Clin Pharm 2012; 34: 295–311.

13.

Laba

Bleasel

Brien

. Strategies to improve adherence to medications for cardiovascular diseases in socioeconomically disadvantaged populations: A systematic review. Int J Cardiol 2013; 167: 2430–2440.

14.

Schedlbauer A, Davies P and Fahey T. Interventions to improve adherence to lipid lowering medication. Cochrane Database Syst Rev 2010; (3): Art.No.: CD004371.

15.

Mansoor

Krass

Aslani

. Multiprofessional interventions to improve patient adherence to cardiovascular medications. J Cardiovasc Pharmacol Ther 2013; 18: 19–30.

16.

Mahtani KR, Heneghan CJ, Glasziou PP, et al. Reminder packaging for improving adherence to self-administered long-term medications. Cochrane Database Syst Rev 2011; (9): Art. No.: CD005025.

17.

Smith

Benjamin

Bonow

. AHA/ACCF Secondary prevention and risk reduction therapy for patients with coronary and other atherosclerotic vascular disease: 2011 Update: A guideline from the American Heart Association and American College of Cardiology Foundation. Circulation 2011; 124: 2458–2473.

18.

Anderson

Adams

Antman

. 2011 ACCF/AHA Focused update incorporated into the ACC/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-elevation myocardial infarction: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation 2011; 123: e426–e579.

19.

Montalescot

Sechtem

Achenbach

. 2013 ESC guidelines on the management of stable coronary artery disease. Eur Heart J 2013; 34: 2949–3003.

20.

Higgins

JPT

Altman

Gøtzsche

. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. Br Med J 2011; 343: d5928–d5928.

21.

Haynes RB, Ackloo E, Sahota N, et al. Interventions for enhancing medication adherence. Cochrane Database Syst Rev 2008; (2): Art.No.: CD000011.

22.

Borenstein M, Hedges LV, Higgins JPT, et al. Multiple Outcomes or Time-Points within a Study. In: Introduction to Meta-Analysis. Chichester, UK: John Wiley & Sons, Ltd, 2009. DOI: 10.1002/9780470743386.ch25.

23.

Higgins JPT, Deeks JJ and Altman DG (eds) Special topics in statistics. In: Higgins JPT, Green S (eds) Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available at: www.cochrane-handbook.org.

24.

Deeks JJ, Higgins JPT and Altman DG (eds) Analysing data and undertaking meta-analyses. In: Higgins JPT and Green S (eds) Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available at: www.cochrane-handbook.org.

25.

Duval

Tweedie

. Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 2000; 56: 455–463.

26.

27.

Thakkar J, Kurup R, Laba T, et al. Mobile telephone text messaging for medication adherence in chronic disease: A meta-analysis. JAMA Intern Med 2016. Epub ahead of print 1 February 2016. DOI: 10.1001/jamainternmed.2015.7667.

28.

Tao

Xie

Wang

. A meta-analysis of the use of electronic reminders for patient adherence to medication in chronic disease care. J Telemed Telecare 2015; 21: 3–13.

29.

Gould KA. A Randomized Controlled Trial of a Discharge Nursing Intervention to Promote Self-Regulation of Care for Early Discharge Interventional Cardiology Patients.

30.

Choudhry

Avorn

Glynn

. Full coverage for preventive medications after myocardial infarction. N Engl J Med 2011; 365: 2088–2097.

31.

Calvert

Kramer

Anstrom

. Patient-focused intervention to improve long-term adherence to evidence-based medications: A randomized trial. Am Heart J 2012; 163: 657–65.e1–657–65.e1.

32.

Gujral

Winckel

Nissen

. Impact of community pharmacist intervention discussing patients' beliefs to improve medication adherence. Int J Clin Pharm 2014; 36: 1048–1058.

33.

Lambert-Kerzner

Carey

. Multifaceted intervention to improve medication adherence and secondary prevention measures after acute coronary syndrome hospital discharge: A randomized clinical trial. JAMA Intern Med 2014; 174: 186–193.

34.

Jiang

Sit

Wong

. A nurse-led cardiac rehabilitation programme improves health behaviours and cardiac physiological risk parameters: Evidence from Chengdu, China. J Clin Nurs 2007; 16: 1886–1897.

35.

Khonsari

Subramanian

Chinna

. Effect of a reminder system using an automated short message service on medication adherence following acute coronary syndrome. Eur J Cardiovasc Nurs 2015; 14: 170–179.

36.

Ockene

Rosal

. Randomized trial of a pharmacist-delivered intervention for improving lipid-lowering medication adherence among patients with coronary heart disease. Cholesterol 2010; 2010: 383281. DOI: 10.1155/2010/383281–383281. DOI: 10.1155/2010/383281.

37.

Park

Howie-Esquivel

Chung

. A text messaging intervention to promote medication adherence for patients with coronary heart disease: A randomized controlled trial. Patient Educ Couns 2014; 94: 261–268.

38.

Polack

Jorgenson

Robertson

. Evaluation of different methods of providing medication-related education to patients following myocardial infarction. Can Pharm J (Ott) 2008; 141: 241–247.

39.

Rinfret

Rodes-Cabau

Bagur

. Telephone contact to improve adherence to dual antiplatelet therapy after drug-eluting stent implantation. Heart 2013; 99: 562–569.

40.

Bond

. The MEDMAN study: A randomized controlled trial of community pharmacy-led medicines management for patients with coronary heart disease. Fam Pract 2007; 24: 189–200.

41.

Castellano

Sanz

Penalvo

. A polypill strategy to improve adherence: Results from the FOCUS project. J Am Coll Cardiol 2014; 64: 2071–2082.

42.

Kripalani

Schmotzer

Jacobson

. Improving Medication Adherence through Graphically Enhanced Interventions in Coronary Heart Disease (IMAGE-CHD): A randomized controlled trial. J Gen Intern Med 2012; 27: 1609–1617.

43.

Lourenço

Rodrigues

Ciol

. A randomized controlled trial of the effectiveness of planning strategies in the adherence to medication for coronary artery disease. J Adv Nurs 2014; 70: 1616–1628.

44.

Ostbring

Eriksson

Petersson

. Medication beliefs and self-reported adherence- results of a pharmacist's consultation: A pilot study. Eur J Hosp Pharm Sci Pract 2014; 21: 102–107.

45.

Anglada-Martinez

Riu-Viladoms

Martin-Conde

. Does mHealth increase adherence to medication? Results of a systematic review. Int J Clin Pract 2015; 69: 9–32.

46.

Pfaeffli Dale

Dobson

Whittaker

. The effectiveness of mobile-health behaviour change interventions for cardiovascular disease self-management: A systematic review. Eur J Prev Cardiol 2016; 23: 801–817.

47.

Connor

Rafter

Rodgers

. Do fixed-dose combination pills or unit-of-use packaging improve adherence? A systematic review. Bull World Health Organ 2004; 82: 935–939.

48.

Cutrona

Choudhry

Fischer

. Modes of delivery for interventions to improve cardiovascular medication adherence. Am J Manag Care 2010; 16: 929–942.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.17 MB