Alexander M. Clark

Meta-Analysis: Secondary Prevention Programs for Patients with Coronary Artery Disease

Cardiovascular disease remains the most common cause of office visits, hospitalizations, and death in the United States: More than 13 million Americans have documented coronary artery disease (CAD), and costs for CAD are expected to exceed

Socioeconomic status and cardiovascular disease: risks and implications for care

393 billion in 2005 (1). Control of the CAD epidemic requires a multifaceted strategy targeting the currently recognized modifiable risk factors for CAD that account for more than 90% of risk, regardless of sex, age, or region (2). This strategy should include primary prevention maneuvers (for the general population and high-risk individuals) and secondary prevention programs (for patients with established CAD). Despite the abundant evidence base for CAD prevention (3), health outcomes studies consistently demonstrate gaps in applying this evidence to clinical practice; these gaps contribute to suboptimal patient outcomes (4). Secondary prevention programs are often proposed as a way to improve management and outcomes. Although several reviews have shown that cardiac rehabilitation reduces mortality in survivors of myocardial infarction (MI) (5-8), these conclusions are informed largely by trials that tested supervised exercise programs versus no exercise postinfarction. Since exercise training confers substantial physiologic and clinical benefits and activity levels are inversely proportional to cardiovascular mortality (9), it is not surprising that trials of exercise programs found positive effects on survival. However, few trials included in these reviews evaluated secondary prevention programs that were not primarily exercise-based. In an earlier review (10), we identified 12 randomized trials (9803 patients) of nonexercise-based secondary prevention programs in patients with established CAD. We demonstrated improvements in risk factor profiles and processes of care (particularly the prescription of proven efficacious therapies) but indeterminate effect on rates of death or recurrent MIs (10). Because current guidelines recommend that secondary prevention programs should not be restricted to supervised exercise programs but should address the full range of modifiable risk factors (11), we conducted a systematic review to update earlier work and to determine the effects of different types of secondary prevention programs (particularly those with a structured exercise component versus those without). Methods Data Sources We searched MEDLINE (19662004); the Cochrane Central Register of Controlled Trials, Issue 4, 2004; EMBASE (19802004); CINAHL (19822004); SIGLE (19802004); and PubMed (January 2004December 2004). We also conducted a cited reference search for our previous review (10) in Web of Science (19992004). We based the searches on the following terms: case management, comprehensive health care, disease management, health services research, home care services, clinical protocols, patient care planning, quality of health care, rehabilitation, nurse led clinics, special clinics, and myocardial ischemia. We hand-searched reference lists of all identified studies, review articles, and references provided by the Centers for Medicare & Medicaid Services and content experts (search strategy available upon request). We limited our search to English-language publications (no abstracts), and the search extends from 1966 to 2004 (we completed the search on 16 December 2004). Study Selection Two investigators independently reviewed the titles and abstracts of all citations to identify studies reporting the effect of secondary prevention programs on mortality, MI, or hospitalization rates in patients with CAD. Both investigators obtained the full text of potentially relevant articles and reviewed them by using prestandardized data abstraction forms and eligibility criteria defined a priori. Any discrepancies were resolved by consensus. We excluded studies if they were not randomized, if they did not include a usual care group, if the outcomes for CAD patients were not reported or were not obtainable from the study investigators, if they evaluated single-modality interventions except exercise (such as telephone follow-up), if they tested interventions delivered to inpatients, or if the interventions were not provided by health professionals (for example, mailed reminders, self-help groups, or self-directed interventions). Data Extraction and Quality Assessment Two investigators extracted all outcome data independently, and a third investigator checked the data. We assigned outcomes according to the intention-to-treat principle and by using the definitions from the primary studies. When necessary, we contacted original investigators to clarify the data for any trial published in the past decade. Authors for 10 of these 21 studies provided further data. Two investigators independently assigned each intervention to 1 of 3 groups: 1) programs that incorporated education and counseling about coronary risk factors with a supervised exercise program (either in a group setting, per traditional comprehensive cardiac rehabilitation programs, or individually delivered), 2) programs that included education and counseling about coronary risk factors but had no exercise component (either delivered in a group setting or individually), and 3) supervised exercise programs only. Data Synthesis and Analysis We performed analyses by using RevMan 4.2 (Update Software, Oxford, United Kingdom). Our outcomes of interest were all-cause mortality and recurrent MIs. Because the outcomes were relatively common, we calculated risk ratios and used the I2 statistic to assess for heterogeneity in each outcome of interest. We combined studies by using the DerSimonian and Laird random-effects model. For the primary analysis, we used data from the longest follow-up period reported in each trial. In a priori sensitivity analyses, we pooled data for 3 follow-up periods (12, 24, and 60 months). To evaluate whether different types of secondary prevention programs had different effects, we calculated the summary risk ratio for each program type and used adjusted indirect comparisons to compare different types of interventions, according to the method of Song and colleagues (12). Role of the Funding Source An earlier version of this evidence report was produced under contract to the Agency for Healthcare Research and Quality, Rockville, Maryland (contract no. 290-02-0023). The funding source had no role in the collection, analysis, or interpretation of the data or in the decision to submit the manuscript for publication. Results Study Selection and Evaluation We identified 6345 citations from electronic databases (n= 6207), reference lists (n= 45), and the Centers for Medicare & Medicaid Services (n= 93). After the initial screening, we reviewed 254 full manuscripts and excluded 179 of these studies after detailed evaluation (Figure 1). Figure 1. Flow of trials through the selection process. Sixteen disagreements among the reviewers about eligibility of the studies occurred for a value of 0.81. All disagreements were resolved by consensus. Of the randomized trials that were eligible for inclusion (13-87), 9 were reported in more than 1 publication. The second publication reported different end points in 2 cases (13-16), results from different follow-up periods in 5 cases (13, 17-26), and results from the subgroup of patients with cardiac disease in 1 case (27, 28). The ninth trial (the World Health Organization [WHO] trial) (29) included 24 collaborating centers; however, the original investigators excluded 7 sites because of poor participant follow-up and 4 sites because of statistically significant differences between the intervention and control groups at baseline. We included the outcome data from the remaining 13 sites as 1 trial for our analysis, an approach validated by the nonsignificant test results for statistical heterogeneity for all-cause mortality and MI. While the 2 Finnish centers in the WHO trial published their results separately (and for several follow-up periods), we included only their 3-year outcome data with the other 11 WHO sites for consistency of data presentation (30-32). Studies Included in the Systematic Review Table 1 presents summary data from the 63 unique randomized trials that were eligible for our systematic review (13-87). Our search retrieved 51 trials that were not included in our previous systematic review (which was limited to literature published before 1999 and excluded any studies with exercise components) (10) and 26 trials that were not included in a more recent systematic review of cardiac rehabilitation (which was limited to literature published before 2003 and included few individual counseling programs) (8). Table 1. Description of Included Studies Qualitative Data Synthesis In all trials, patients who were randomly assigned to the control groups received usual care (this was generally undefined). Table 1 describes the types of secondary prevention programs; few trials described the intensity of the interventions. Almost all trials enrolled highly selected study samples: Forty-five trials recruited patients after acute MI or a coronary revascularization procedure. Thirty-five trials excluded elderly patients, and 19 trials excluded women (Table 1). Indeed, women constituted fewer than 50% of study participants in all but 2 trials. No trial was double-blind (which is not surprising, considering the nature of the intervention), and very few trials described randomization procedures or accounted for discrepancies between sample sizes at recruitment and follow-up. As a result, Jadad quality scores clustered around 2 (Table 2). Furthermore, only 15 (24%) trial reports described adequate allocation concealment. Table 2. Methodologic Quality of Included Studies No trial reported side effects with the secondary prevention programs beyond the adverse clinical outcomes described later. Quantitative Data Synthesis All-Cause Mortality Only 1 of the 40 trials reporting this outcome f

Participating in cardiac rehabilitation: a systematic review and meta-synthesis of qualitative data

Socioeconomic status (SES) refers to an individuals social position relative to other members of a society. Low SES is associated with large increases in cardiovascular disease (CVD) risk in men and women. The inverse association between SES and CVD risk in high-income countries is the result of the high prevalence and compounding effects of multiple behavioral and psychosocial risk factors in people of low SES. However, strong and consistent evidence shows that parental SES, childhood and early-life factors, and inequalities in health services also contribute to elevated CVD risk in people of low SES who live in high-income countries. In addition, place of residence can affect CVD risk, although the data on the influence of wealth distribution and work-related factors are inconsistent. Studies on the effects of SES on CVD risk in low-income and middle-income countries is scarce, but evidence is emerging that the increasing wealth of these countries is beginning to lead to replication of the patterns seen in high-income countries. Clinicians should address the association between SES and CVD by incorporating SES into CVD risk calculations and screening tools, reducing behavioral and psychosocial risk factors via effective and equitable primary and secondary prevention, undertaking health equity audits to assess inequalities in care provision and outcomes, and by use of multidisciplinary teams to address risk factors over the life course.

A meta-analysis of randomized control trials of home-based secondary prevention programs for coronary artery disease.

Background: Participation in cardiac rehabilitation (CR) benefits patients with coronary heart disease (CHD), yet worldwide only some 15–30% of those eligible attend. To improve understanding of the reasons for poor participation we undertook a systematic review and meta-synthesis of the qualitative literature. Methods: Qualitative studies identifying patient barriers and enablers to attendance at CR were identified by searching multiple electronic databases, reference lists, relevant conference lists, grey literature, and keyword searching of the internet (1990–2010). Studies were selected if they included patients with CHD and reviewed experience or understanding about CR. Meta-synthesis was used to review the papers and to synthesize the data. Results: From 1165 papers, 34 unique studies were included after screening. These included 1213 patients from eight countries. Study methodology included interviews (n = 25), focus groups (n = 5), and mixed-methods (n = 4). Key reasons for not attending CR were physical barriers, such as lack of transport, or financial cost, and personal barriers, such as embarrassment about participation, or misunderstanding the reasons for onset of CHD or the purpose of CR. Conclusions: There is a vast amount of qualitative research which investigates patients’ reasons for non-attendance at CR. Key issues include system-level and patient-level barriers, which are potentially modifiable. Future research would best be directed at investigating strategies to overcome these barriers.

Patient and informal caregivers' knowledge of heart failure: necessary but insufficient for effective self-care.

Background A variety of different types of secondary prevention programs for coronary heart disease (CHD) exist. Home-based programs have become more common and may be more accessible or preferable to some patients. This review compared the benefits and costs of home-based programs with usual care and cardiac rehabilitation. Methods A meta-analysis following a systematic search of 19 databases, existing reviews, and references was designed. Studies evaluated home-based interventions that addressed more than one main CHD risk factor using a randomized trial with a usual care or cardiac rehabilitation comparison group with data extractable for CHD patients only and reported in English as a full article or thesis. Results Thirty-nine articles reporting 36 trials were reviewed. Compared with usual care, home-based interventions significantly improved quality of life [weighted mean difference: 0.23; 95% confidence interval (95% CI): 0.02-0.45], systolic blood pressure (weighted mean difference: − 4.36mmHg; 95% CI: − 6.50 to − 2.22), smoking cessation (difference in proportion: 14%; 95% CI: 0.02-0.26), total cholesterol (standardized mean difference: − 0.33; 95% CI: − 0.57 to − 0.08), and depression (standardized mean difference: − 0.33; 95% CI: − 0.59 to − 0.07). Effect sizes were small to moderate and trials were of low-to-moderate quality. Comparisons with cardiac rehabilitation could not be made because of the small number of trials and high levels of heterogeneity. Conclusion Home-based secondary prevention programs for CHD are an effective and relatively low-cost complement to hospital-based cardiac rehabilitation and should be considered for stable patients less likely to access or adhere to hospital-based services. Eur J Cardiovasc Prev Rehabil 17:261-270

An open letter to The BMJ editors on qualitative research

Self‐care of heart failure (HF) is a process that is important, complex, and challenging. Little is understood of the contextual factors influencing self‐care. We aimed to examine the individual and contextual factors perceived by patients and their informal caregivers’ to influence their willingness and capacity to undertake effective HF self‐care.

Secondary prevention programmes for coronary heart disease: a meta-regression showing the merits of shorter, generalist, primary care-based interventions

Seventy six senior academics from 11 countries invite The BMJ ’s editors to reconsider their policy of rejecting qualitative research on the grounds of low priority. They challenge the journal to develop a proactive, scholarly, and pluralist approach to research that aligns with its stated mission

A meta-review of evidence on heart failure disease management programs: the challenges of describing and synthesizing evidence on complex interventions.

Background The aim of this study was to determine which programme characteristics influence the effectiveness of secondary prevention programmes for Coronary Heart Disease. Design The study follows a meta-regression design. Methods We conducted a meta-regression within a systematic review of randomized trials comparing secondary prevention programmes versus usual care. The primary outcome was all-cause mortality. Studies were identified by searching multiple electronic databases, bibliographies of published studies, contact with experts, and references provided by the United States Centers for Medicare and Medicaid Services. Primary authors of all relevant trials were surveyed for detailed information on programme characteristics. Forty-six unique trials were identified (18 821 patients). The pooled all-cause mortality risk ratio (RR) for programmes was 0.87 [95% confidence interval (CI) 0.79-0.97]. Programmes containing less than 10 h of patient contact with health professionals reduced all-cause mortality (RR 0.80, 95% CI 0.68-0.95) as effectively as programmes with more contact time. Programmes provided in general practice settings were effective at reducing all-cause mortality (RR 0.76, 95% CI 0.63-0.92) and compared favourably with the effectiveness of hospital-based programmes. Other characteristics, including specialist versus generalist provision, did not appreciably impact programme effectiveness. Conclusions Shorter secondary prevention programmes, those based in general practice, and those staffed by generalists are at least as effective in reducing all cause mortality in patients with coronary heart disease as longer programmes, hospital-based programmes, and programmes staffed by specialists.

What Is the Strength of Evidence for Heart Failure Disease-Management Programs?

BackgroundDespite favourable results from past meta-analyses, some recent large trials have not found Heart Failure (HF) disease management programs to be beneficial. To explore reasons for this, we evaluated evidence from existing meta-analyses.MethodsSystematic review incorporating meta-review was used. We selected meta-analyses of randomized controlled trials published after 1995 in English that examined the effects of HF disease management programs on key outcomes. Databases searched: MEDLINE, EMBASE, Cochrane Database of Systematic Reviews (CDSR), DARE, NHS EED, NHS HTA, Ageline, AMED, Scopus, Web of Science and CINAHL; cited references, experts and existing reviews were also searched.Results15 meta-analyses were identified containing a mean of 18.5 randomized trials of HF interventions +/- 10.1 (range: 6 to 36). Overall quality of the meta-analyses was very mixed (Mean AMSTAR Score = 6.4 +/- 1.9; range 2-9). Reporting inadequacies were widespread around populations, intervention components, settings and characteristics, comparison, and comparator groups. Heterogeneity (statistical, clinical, and methodological) was not taken into account sufficiently when drawing conclusions from pooled analyses.ConclusionsMeta-analyses of heart failure disease management programs have promising findings but often fail to report key characteristics of populations, interventions, and comparisons. Existing reviews are of mixed quality and do not adequately take account of program complexity and heterogeneity.

Complex critical realism: tenets and application in nursing research.

Heart failure (HF) disease-management programs are increasingly common. However, some large and recent trials of programs have not reported positive findings. There have also been parallel recent advances in reporting standards and theory around complex nonpharmacological interventions. These developments compel reconsideration in this Viewpoint of how research into HF-management programs should be evaluated, the quality, specificity, and usefulness of this evidence, and the recommendations for future research. Addressing the main determinants of intervention effectiveness by using the PICO (Patient, Intervention, Comparison, and Outcome) approach and the recent CONSORT (Consolidated Standards of Reporting Trials) statement on nonpharmacological trials, we will argue that in both current trials and meta-analyses, interventions and comparisons are not sufficiently well described; that complex programs have been excessively oversimplified; and that potentially salient differences in programs, populations, and settings are not incorporated into analyses. In preference to more general meta-analyses of programs, adequate descriptions are first needed of populations, interventions, comparisons, and outcomes in past and future trials. This could be achieved via a systematic survey of study authors based on the CONSORT statement. These more detailed data on studies should be incorporated into future meta-analyses of comparable trials and used with other techniques such as patient-based outcomes data and meta-regression. Although trials and meta-analyses continue to have potential to generate useful evidence, a more specific evidence base is needed to support the development of effective programs for different populations and settings.