Objective To identify the main mechanisms of heart failure (HF) disease management programmes based in hospitals, homes or the community.
Methods Systematic review of qualitative and quantitative studies using realist synthesis. The search strategy incorporated general and specific terms relevant to the research question: HF, self-care and programmes/interventions for HF patients. To be included, papers had to be published in English after 1995 (due to changes in HF care over recent years) to May 2014 and contain specific data related to mechanisms of effect of HF programmes. 10 databases were searched; grey literature was located via Proquest Dissertations and Theses, Google and publications from organisations focused on HF or self-care.
Results 33 studies (n=3355 participants, mean age: 65 years, 35% women) were identified (18 randomised controlled trials, three mixed methods studies, six pre-test post-test studies and six qualitative studies). The main mechanisms identified in the studies were associated with increased patient understanding of HF and its links to self-care, greater involvement of other people in this self-care, increased psychosocial well-being and support from health professionals to use technology.
Conclusion Future HF disease management programmes should seek to harness the main mechanisms through which programmes actually work to improve HF self-care and outcomes, rather than simply replicating components from other programmes. The most promising mechanisms to harness are associated with increased patient understanding and self-efficacy, involvement of other caregivers and health professionals and improving psychosocial well-being and technology use.
Statistics from Altmetric.com
Why are the effects of heart failure (HF) disease management programmes inconsistent? Understanding how these complex interventions achieve their outcomes is vital for designing interventions that are more consistently effective.1 However, despite over 60 published trials and other evaluations, very little research has examined the ways through which these programmes achieve their particular effects, that is, their mechanisms.2 ,3 Research to identify intervention mechanisms is common in the behavioural and biological sciences4 and current policy states that this explanatory step is essential for improving how interventions work.1
Better understanding of the mechanisms of HF disease management programmes is important because the inconsistent effects of these increasingly common interventions are not well explained.2 ,5 Programmes have a clear and important aim: to support effective HF self-care.6 However, variations in the programme effects have been downplayed to date.3 ,7 ,8 Inconsistent findings have been dismissed,2 ,7 ignored,8 simplified away9 or otherwise left unexplained.10 Yet, better knowledge of mechanisms is especially useful when programme effects vary across settings.11 Knowledge of the influence of mechanisms can then be used to inform intervention design and develop more consistently effective programmes in different contexts.11–13 This review thus has a timely and important purpose: to identify the main mechanisms of HF disease management programmes, using data from published trials and other studies.
To identify the main mechanisms of HF disease management programmes, we performed a systematic review with realist synthesis, a well-established approach in the social and health sciences to identify intervention mechanisms.13
While conventional meta-analysis seeks to increase statistical power via pooling intervention effect sizes from smaller studies together, realist review seeks to generate learning and insights into why interventions work when they do and why they have not worked when they do not and what explains these effects.13 The value of studies to the analysis is not attached to size or statistical power, but the degree to which research findings yield insights into what it is about interventions that works for whom, when and why.13 As data on mechanisms can be found in a wide range of different types of studies,13 the review included studies using both qualitative and quantitative data.
To be included, studies had to contain primary qualitative or quantitative data for HF populations plus data or themes pertaining to mechanisms of HF self-care interventions. HF self-care was defined throughout as ‘the decisions and strategies undertaken by the individual in order to maintain life, healthy functioning and wellbeing’,14 including medicines management, restriction of fluid, salt and alcohol, smoking cessation, regular physical activity and maintenance of psychosocial well-being.14 Studies had to be published as full papers in English and contain data, from patients, health professionals and/or family members/caregivers, extractable specifically for HF populations.
The search strategy was detailed and comprehensive, involving multiple databases and search terms (see online supplementary data). Complete details on the search strategy can be found online in the international prospective register of systematic reviews in health and social care: the PROSPERO database (Registration number: CRD42014009194). The search strategy incorporated general and specific terms in relation to the central concepts relevant to the research question: HF, self-care and programmes/interventions for HF patients. Papers included were published in English from 1995 (due to changes in HF care over recent years) until May 2014. Databases searched were Ovid MEDLINE (In Process and Other Non-Indexed Citations) and Ovid MEDLINE, Ovid EMBASE, Ovid PsycINFO, CSA Sociological Abstracts, EBSCO Academic Search Complete, EBSCO CINAHL, EBSCO SocINDEX, ISI Web of Science: Social Sciences Citation Index and Science Citation Index Expanded, Cochrane Central Register of Controlled Trials (CENTRAL) and Scopus. Grey literature was found via Proquest Dissertations and Theses, Google and publications from organisations focused on HF or self-care.
After initial screening of 15439 abstracts, 688 studies were reviewed via title and abstract review for inclusion–exclusion criteria, and finally for full review and data extraction (figure 1). After primary data extraction, the second reviewer reviewed the full paper for accuracy and agreement on inclusion: differences were resolved in discussion. Study quality was appraised using the appropriate validated quality screening tool from the Critical Appraisal Skills Programme.15
Matrices were used to record relevant qualitative and quantitative data alongside population details (sex, age, HF severity) and, when appropriate, details about the interventions were recorded. Descriptions of the interventions used a standardised taxonomy to ensure consistency.14 Mechanisms usually have to be inferred or interpreted from these data as opposed to being clearly labelled in studies.4
Current quality standards for realist synthesis16 were used to ensure rigour and quality of the review. The synthesis of the data was undertaken by AMC and KW in accordance with these standards when study data identified key elements of importance to the success or failure of an intervention in a certain context and identified prominent recurrent patterns…that explain(ed) these outcomes through the means (mechanisms) by which they occurred. In addition, in line with these quality standards, an additional member of the research team (KC) verified the final synthesis findings.
In deciding whether data or themes were pertinent to the synthesis, reviewers considered if the identified data offered an explanatory account of what was going on between the intervention(s) and its outcomes. For qualitative studies, data were derived from themes or data relating to mechanisms, while for quantitative studies, primary numerical data were extracted from each study that were interpreted as giving insight into underlying mechanisms. Data from different types of studies were triangulated and reviewers’ interpretations were applied to the extracted data. Higher order concepts were identified as codes through comparison of mechanisms found in original studies and reinterpreted in the context of other studies.
Thirty-three studies were identified for inclusion (see online supplementary table, figure 1). Of these studies, there were 18 randomised controlled trials (RCTs),17–34 three mixed methods,35–37 six pre-test/post-test38–43 and six qualitative studies.44–49 A total of 3355 participants were included with an average age of 65.14 years; 1171 (35%) were women (one trial21 did not report the sex of participants).
Data extraction from the included studies yielded 66 mechanisms, which fell into six primary groupings (see table 1). Discussed below, almost half the studies (n=16, 48%) indicated only one mechanism, while none found all six mechanisms. Over a third (n=12) of the studies revealed two or three mechanisms. Only three of the articles (10%) demonstrated four or five mechanisms.
Effective interventions mobilised real support from health professionals (n=19)
Almost 60% of the studies found that health professional support affected self-care. Specifically, greater health professional involvement was perceived to increase intervention effectiveness23 ,34 ,49 but only in some instances. This occurred particularly in contexts in which patients had a strong sense of receiving ‘professional follow-up’ care27 or help with basic management needs.37 ,41 ,49 Participants appeared to especially value support from a wide range of professional groups49 or specialists.30 ,32 ,42
Health professional support was effective in these ways providing that consultations were judged to be of ‘sufficient’ time duration,29 ,42 ,44 used research,35 incorporated patient goals into care,31 ,32 were proactive,43 provided rapid feedback or response to worsening symptoms,35 ,48 and consolidated existing patient–professional relationships.42
Intervention effectiveness was reduced when health professionals focused too much on simply imparting information during consultations,44 or placed excessive priority on treatment goals over the patient's goals/values18 ,47 or cultural beliefs and practices.47 Other barriers to effectiveness included a lack of time with professionals36 or excessive repetition.45
Effective interventions promoted psychological well-being (n=17)
Effective interventions improved recipients’ psychosocial well-being in a variety of ways. This occurred predominantly when interventions increased self-efficacy.19 ,20 ,28 ,33 ,38 ,46 Other psychological changes associated with effective interventions were increased personal confidence,22 ,24 improved hope38 and lower depression.36 ,38 ,40
The power of interventions to affect change was reduced when patients were depressed,37 when interventions focused excessively on knowledge over skills,27 ,37 ,44 ,49 did not induce sufficient fear of worsening HF49 or reduced personal confidence.22 ,30 ,48
Effective interventions increased insight and understanding (n=9)
Effective interventions promoted understanding of the nature and complexities of HF and its self-care. Interventions which functioned to emphasise or reinforce the many and complex links between symptoms and HF self-care tasks were perceived to be particularly valuable.17 ,25–27 ,37 ,47 ,49 Interventions that promoted this understanding included educational materials that incorporated audio-video techniques,17 demanded attention,17 were especially realistic17 or had content which allowed recipients to better understand links between the heart's function and the body.48 ,49 Such understanding went beyond mere knowledge, which was seen to be necessary but insufficient to promote effectiveness.27 ,44
Effective interventions increased others’ involvement (n=9)
Effective interventions focused on patients and mobilized other people to support the patient and their HF self-care. This was most commonly noted in relation to promoting involvement of family member caregivers17 ,18 ,26 ,28 ,36 ,37 ,39 ,45 ,47 but could extend to other similar HF patients who provided peer support.36 Interventions were also noted to be effective when they were perceived to foster in patients a subjective sense of receiving support or being cared for by other people.45
Effective interventions are individualised and responsive (n=7)
Interventions were seen to be more effective when they were responsive to individuals’ needs and preferences. This could be expressed in a variety of ways, including individualised reinforcement of particularly salient information,45 ,49 responsiveness to existing life routines in self-care regimen18 ,34 ,37 ,49 and their scheduling18 ,21—for example, by integrating medication management with an individual's personal household routines.49
Inconsistencies in HF disease management programmes persist in recent trials.50 ,51 To design more consistent HF disease management programmes, it is important to harness these same main mechanisms across different patients and settings. This is the first systematic review to identify that the main mechanisms of HF disease management programmes are associated with increased patient understanding of HF and its self-care, higher involvement of caregivers and family members in this self-care, enhanced self-efficacy and psychological well-being, increased support from health professionals and ease of use of technology.
These main mechanisms do not operate in isolation but require favourable contextual factors to be present. For example, for the benefits of mechanisms related to support from health professionals to occur, healthcare had to also be responsive, individualised, based on research and be of sufficient duration. As such, designing effective programmes requires understanding and harnessing of the mechanisms through which a programme exerts its effects and of the other wider factors that can modify the influence of these mechanisms.11
Further research is needed into the mechanisms of HF disease management programmes—interventions with puzzlingly and ongoing inconsistencies and small effect sizes for a syndrome for which such strong evidence exists for effective medicines and self-care strategies. Future trials and large-scale evaluations of HF disease management programmes should routinely incorporate an examination of intervention mechanisms. This expanded focus on mechanisms reduces the likelihood of future large and costly studies examining only if interventions work to the neglect of understanding why they work. While many factors can explain the negative findings of past large trials, such as variations in populations, interventions, fidelity and comparison groups, studies that both measure outcomes and examine mechanisms have more explanatory power and potential for improving future programmes.1 This research should incorporate mixed and qualitative methods to explore systematically which mechanisms are consistently associated with positive programme outcomes and how these are moderated by other intervention and contextual factors. Specific guidance on how to do this in research studies is available in a number of generic approaches4and in dedicated methodologies, notably process evaluation1 and realist evaluation.52
In terms of limitations, while this review has identified main mechanisms of programmes, these findings are best viewed as hypothesis generating. As with all reviews, the conclusions of this review are circumscribed by the availability of studies published in journals. While the studies included in this review contained data on intervention mechanisms, few overtly discussed doing so and none systematically compared programmes using different mechanisms. Additionally, the studies included were small in sample size and many published studies in this comparably large field of health services research did not include data on mechanisms. Consequently, the studies reviewed may not be representative. Future studies may not find those mechanisms identified in this review to be as influential.52
What is already known on this subject?
Heart failure is a common and burdensome syndrome for patients, their families and health systems.
People with heart failure can live longer lives of higher quality if they engage in effective self-care, including management of medicine and behaviours.
Heart failure disease management programmes are commonly used to support this self-care but have inconsistent effects that are poorly understood.
What might this study add?
This is the first review of studies with data on the mechanisms of heart failure disease management programmes.
Effective programmes contain components that increase patient understanding of heart failure and its self-care, foster greater self-efficacy and higher involvement of family and other caregivers, improve patient psychosocial well-being and provide support from health professionals and for use of technology.
How might this impact on clinical practice?
To be effective, future heart failure disease management interventions should harness these mechanisms via components that function to foster improved patient understanding of HF and self-efficacy in relation to self-care, mobilise psychosocial support from family and other caregivers, increase support from other health professionals and promote easy technology usage.
Programmes should be adapted for different context but nevertheless seek to use these mechanisms.
Twitter Follow Alexander Clark at @alexclark1944
Contributors The study was conceived by AMC. The search was coordinated by KW and AMC. Study screening and data extraction was undertaken by KW, DB, JK, LT, LMcL, AMC and KC. The analysis was led by AMC with comments provided by KW, DB, JK, LT, LMcL, AMC and KC. AMC wrote the first and final manuscript drafts with all author providing input.
Funding Funding for this review was provided by the Canadian Institutes of Health Research, Knowledge Synthesis Grant (number 124591).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Study extractions for all studies are available via email from AMC.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.