Impact of community-based support services on antiretroviral treatment programme delivery and outcomes in resource-limited countries: a synthetic review

To optimally capitalize on the relatively limited scientific information on the topic, the authors decided to apply a synthetic review design, a two-step approach which combines the strengths of a systematic review – gathering quantitative evidence on the effectiveness of community support – with additional attention for the social and behavioural mechanisms through which community support can help overcome the above-cited challenges – i.e. a realist evaluation approach [32].

Firstly, the current review aims to provide an exhaustive overview of the scientific evidence of the contribution of community support to ART programmes in resource-limited settings. To assess the achievements of community support programmes in producing favourable ART outcomes, we systematically reviewed published work (including e-publication ahead of print) for the period from November 2003 to December 2011 in accordance with the PRISMA guidelines [33]. We used the following medical subheading terms and text strings: “HIV” OR “AIDS” AND “sub-Saharan Africa” OR “sub-Sahara Africa” OR “Southern Africa” OR “resource-limited country” AND “antiretroviral therapy” OR “ART” OR “HAART” AND “community support” OR “community health worker” OR “community care giver” OR “lay health worker” OR “DOT-HAART” OR “treatment buddy” OR “adherence supporter” OR “peer health worker” OR “expert patient” to identify research articles and policy documents on the achievements of ART-related community support programmes in the ISI Web of Knowledge, Science Direct, BioMed Central, OVID Medline, PubMed, Social Services Abstracts, and Sociological Abstracts. Relevant literature was also identified by checking the websites of the WHO, the World Bank, UNAIDS, and Health Systems Trust. Government publications and institutional reports released by non-governmental organizations and academic research centres were gathered. Finally, we checked the reference lists of the above cited sources for potentially relevant books, reports and papers.

In order to merge systematic review standards with the realist evaluation approach in a second step, the latter should be applied after finishing the review in accordance with the PRISMA guidelines. In this second stage, we search for the pathways (‘why’- and ‘how’-questions) though which community support – as described in the selected studies – can impact ART programme delivery and outcomes. This requires in-depth analysis of the argumentation by the selected authors with regards to the mechanisms, contexts and outcomes of community support initiatives as part of ART programmes in resource-limited settings.

As the selected topic is innovative and the number of available scientific studies assessing the contribution of community support to ART programmes is thus rather limited, we opted to include studies applying a wide range of research designs and community support typologies.

We included all quantitative and qualitative (English-language) research papers on the selected topic, as different perspectives on a similar topic are often complementary. We included randomized controlled trials, studies utilizing a comparison group (including pre-test, post-test design), retrospective cohort studies, descriptive studies, and qualitative studies.

Secondly, with regard to the study population, we included all studies whose study population consisted of HIV/AIDS-patients enrolled in an ART programme in a resource-limited setting and not purposively selected for being on second-line treatment. ART was defined as treatment with at least three active antiretroviral medications, typically two nucleoside or nucleotide reverse transcriptase inhibitors (NTRI) plus a non-nucleoside reverse transcriptase inhibitor or a protease inhibitor or another NRTI called abacavir. We excluded articles which exclusively studied child or adolescent populations or articles reporting the accomplishments of ART programmes in resource-rich settings.

Thirdly, we opted for a broad perspective when assessing the types of community interventions included in the review. For inclusion, the studies had to comprise (1) lay members of the community (2) who enjoyed no or only limited training and (3) fulfilled a more or less organised role or function in the ART programme. This resulted in a wide range of community support initiatives (e.g. community health workers, community care workers, lay health workers, treatment buddies, field officers, peer educators/counsellors, adherence supporters, etc.). These broad criteria were chosen to capture all available evidence on the contribution of community support to ART delivery and outcomes in resource-limited settings. We did not include community-based ART programmes that lacked a clear community-based support component.

As the current review study is interested in both the impact of community support on programme outcomes and the mechanisms through which community support can overcome the five above-cited challenges to sustainably scaling-up ART in resource-limited settings, all studies that reported any type of outcome measure related to ART (including access, coverage, adherence, retention, virological and immunological outcomes, and survival), as well as all studies describing the context in which and the mechanisms through which community support can impact ART delivery and outcomes, were included.

In accordance with the PRISMA guidelines, we first excluded all duplicates from our total of 2344 selected records. Secondly, two researchers independently reviewed all titles of the identified research papers and reports (1831 titles on community support and ART), with any differences being discussed and resolved. Of the remaining articles and reports (242 abstracts), the abstracts were assessed. In the third and final step, the authors retrieved and independently reviewed the full-length paper (81 papers) to screen it according to criteria of content (providing relevant information the impact of community support on ART outcomes (see above)) and quality (i.e. are the study design, data collection methods, sampling strategy and analytic approach apparent and appropriate; is the context described sufficiently and the range of missing data acceptable and in accordance with the methodological requirements of a systematic review as specified by Mckee & Britton, Oxman, and Walsh? [34‐36] (see Figure 1)). This resulted in 29 full-text articles and one policy report, a total of 30 publications, assessing the contribution of community support initiatives to the health of HIV patients on ART in 18 different resource-limited countries (Table 1) [20, 21, 23, 27, 37‐62].

In accordance with the methodological literature [63], we performed a systematic content analysis to produce a concise summary (Table 1) of the overall effect of community support interventions on ART programmes outcomes. However, Van de Knaap et al. [32] and Forbes & Griffiths [63] have already indicated that such a systematic review only demonstrates whether a policy intervention works or not, and not why or how it works. Therefore, the current review also used a realist evaluation strategy to assess the context of the studies and the mechanisms through which these community support initiatives can address the above-cited ART programme challenges, and thus contribute to a sustainable ART scale-up in resource-limited countries [32, 63, 64].

The selected articles reported on the outcomes of 22 different programmes in 18 countries. Only eight of the 30 selected papers (27%) were published in 2007 or earlier. All other papers were published more recently, underlining the increasing relevance of and attention for this topic in health systems research.

As a consequence of the broad selection criteria – aiming to assess the impact of community support on ART programme outcomes as well as obtain a comprehensive overview of the potential contribution of these community support initiatives in addressing the five cited challenges to a sustainable ART scale-up – the 30 studies used a wide range of methodological designs: nine descriptive studies, five (cluster/nested-)randomised controlled trials, four quasi-experiments, five retrospective/observational cohort studies, and two qualitative studies. By offering relevant information on the sample size, research design, time frame and study limitations, Table 1 provides insight into the scientific and methodological quality of the included studies.

Table 1 further demonstrates that the selected studies represent a wide range of community support providers that are involved in ART-related activities. In addition, different studies often use alternative names for the same type of community support provider. Our review identified the following nine types. The first type of community support providers are community health workers (CHWs) (11 studies). CHWs can be defined as “non-professional cadres of health workers who undertake short course training and work within their own communities to complement and support the services provided by other health workers” [43, 65]. This definition also includes ‘accompagnateurs’, as developed by Partners In Health (PIH) [40, 50]. Secondly, two studies employed community care coordinators, which can be described as PLWHA who are trained to perform the tasks of CHWs [60, 61]. Very similarly, three studies assessed the contribution of peer health workers, defined as CHWs who are HIV positive and whose role it is to conduct adherence counselling and provide health education and psychosocial support [27, 44, 45]. Fourthly, a number of studies employed field officers as a means for community mobilisation: field officers can be defined as trained lay workers who support drug delivery and monitor patients. This is very similar to the definition of CHWs, but field officers generally enjoyed formal education in social sciences or education, enabling them to embrace a more holistic approach with sufficient attention for the psychosocial aspects of health and illness [38, 41, 48]. The fifth type of community support initiative, health extension workers, are recruited from the community and trained to manage operations of health posts, conduct home visits and outreach services to promote preventive health actions, refer cases to health centres and follow up on referrals [37, 66]. Four studies employed (peer) HIV/AIDS lay counsellors, whose tasks included managing HIV testing, providing pre-ART training and ART adherence support, searching for patients eligible for ART and tracking defaulting ART patients [39, 42, 54, 55]. The seventh type of community support strategy is Directly Observed Therapy (DOT) for ART, employing a community member close to the patient to directly monitor the daily medication intake – a strategy modelled after an effective method for controlling tuberculosis [47, 67]. The eighth group of community support providers, labelled adherence supporters, are trained to promote healthy behaviours, provide HIV/AIDS&ART-related knowledge and skills and promote adherence by providing psychosocial support [20, 21, 46, 49, 56, 57]. The final form of community support is labelled home-based care volunteers. These community members are trained to provide ART adherence counselling and perform a wide range of home-based care activities [62].

The wide range of tasks performed by community members, combined with the different descriptions for similar activities indicate that there is a need for further conceptual work in order to clearly establish a typology of these community support initiatives applicable in the field as well as in research studies on the topic. The current review study aims to explore the common ground between all these different types of community support in order to provide a comprehensive overview of the potential contribution of these different community support initiatives to successfully scaling-up ART in developing settings. However, it is vital to recognize – and take into account – this diversity in community support initiatives and its descriptions (displayed in Table 1) when interpreting their contribution to ART programme delivery and outcomes in high HIV-prevalence resource-limited settings.

The literature search indicated that there is not much scientific evidence of the explicit use of PLWHA in ART programmes. We found only 12 articles that reported the use of HIV-positive individuals in ART care [23, 27, 39, 42, 44‐46, 55, 58, 60‐62], of which, only seven specifically targeted PLWHA as providers of community support in ART programmes [42, 44, 45, 55, 58, 60, 61]. Despite this wide range of community support providers that are involved in ART-related activities, it is useful to search for common roles and functions in scaling up ART in resource-limited settings to obtain an evidence-based overview of their relative contribution.

A review of the literature (Table 1) on the association between community support and ART programme outcomes clearly indicates that the community can have a contributory role in scaling up ART in resource-limited settings [20, 23, 41, 42, 45, 47, 48, 50, 52, 53, 62, 68]. Almost all selected studies reported an unambiguous positive impact of community support on a wide range of aspects of the ART programme. First of all, a number of studies indicated that community support can aid in expanding the access to and increasing the coverage of ART programmes in resource-limited settings [23, 27, 37, 38, 40, 42, 43, 51, 55]. In addition, the services provided by these community support initiatives have been shown to increase adherence levels[41, 49‐51, 54‐58]. Thirdly, a number of studies demonstrated that the community component significantly improved the virological and immunological outcomes of the programme: patients with this kind of support achieved higher rates of virological suppression and immune restoration than patients lacking this support [20, 21, 23, 27, 38‐40, 43‐45, 47‐52, 54, 56, 59‐62]. Finally, the results displayed in Table 1 also indicate that community support initiatives can improve levels of patient retention[39, 42, 46, 48, 56, 59, 61, 62] and increase rates of survival in ART patients in high HIV-prevalence, resource-limited settings. [39, 41, 48, 53, 59, 62]. Unfortunately and although the general trend is evident, it was not possible to quantitatively estimate the mean effect of community support initiatives on the selected ART programme outcomes because of the fact that the selected studies were limited in number and displayed a wide range of programme and research designs.

The current review aimed to collect the scientific evidence for the potential of community support initiatives to address the above-cited barriers to sustainable ART scale-up in a resource-limited setting, namely: (1) the lack of integration of ART services into the general health system; (2) the growing need for comprehensive care addressing the psychosocial and economic dynamics of HIV/AIDS; (3) the need to empower patients on ART towards self-management ; (4) the importance of defaulter tracing to improve retention in care; and (5) the crippling shortage in human resources for health in contrast to the growing caseload of people to be maintained on ART in the long term. To combat the HIV/AIDS epidemic effectively in high-HIV-prevalence, resource-limited countries, we thus need more care (i.e. integrated and comprehensive chronic disease care, patient empowerment and defaulter tracing) for more people (i.e. universal access) in a context of limited financial and human resources. How and why community care providers – according to the selected studies – can contribute to sustainably and successfully scaling up ART in resource-limited settings, while maintaining a high quality of care, is demonstrated below.

The vast majority of studies have indicated that community support providers have the potential to become essential players in the provision of HIV/AIDS care in particular and form part of a broader mobilisation of community and non-governmental participation in the health system at large [10]. However, the literature review did not only produce positive outcomes. Ford and colleagues [73] systematically assessed a series of randomised control trials that tested the outcomes of DOT-ART. Over half of the studies were performed in the United States, but five included data from resource-limited settings. The article showed no statistically significant benefit of DOT-ART over self-administered treatment, which questions the concurrent scientific results that indicated strong positive community support outcomes. However, the results of Ford’s review should not be viewed as evidence against community support, but rather as evidence that a particular paternalistic and disempowering approach to community support does not necessarily produce positive treatment outcomes. To capitalise optimally on the potential contribution of the community to a sustainable ART programme, community support should not render the patients incapable and thus in need of care, but rather support patients to take up responsibility. Our findings indicate that community support can be a crucial part of such a health-enabling community that is designed to empower the patients to self-manage their illness actively and durably.

A number of studies raised concerns about the quality and sustainability of the community support programmes [23, 39, 55, 61]. Most of the articles in our review focused their attention on showing that the use of lay health workers does not compromise the quality of HIV/AIDS care. The vast majority of reviewed studies concluded that this fear did not materialise during the observed time period, but these favourable outcomes do not warrant complacency [27, 44, 45, 61]. Hermann and colleagues [23] have reported that community support programmes often devote insufficient attention to CHW retention, quality supervision and continuous training, which jeopardises the quality of these programmes over time. Several programmes have indicated that CHWs cannot be retained in the long term if they do not receive adequate remuneration [55]. A more formal recognition of these initiatives as an integral part of the health systems is therefore urgently required. The real cost of community support programmes also includes the qualified human resources needed to supervise and train these new community health worker cadres [23]. The above indicates that community mobilisation is probably a necessary component of scaling up ART in sub-Saharan Africa, but not necessarily a cheap or an easy one.

To capitalise fully on the communities’ ability to play a contributory role in durably scaling up ART, a supportive policy environment and an enabling regulatory framework are needed to drive the implementation of task-shifting policies and to resource, guide and support these lay CHWs [55, 74]. Scott & Shanker (2010) reported in their article on an Indian CHW programme that the programme under study was institutionally limited by (1) the outcome-based remuneration structure; (2) inadequate institutional support; (3) the rigid hierarchical structure of the existing health system; and (4) a dearth of participation at the community level [75]. Hermann and colleagues [23] have mentioned five necessary conditions for all CHW programmes to be successful, namely: (1) careful selection based on motivation and belonging to the community; (2) initial training; (3) simple guidelines and standardised protocols; (4) good supervision, support and relationships with formal health services; and (5) adequate remuneration and career structure. An additional three conditions were identified to scale up these CHW initiatives successfully in resource-limited settings. Without (1) sufficient political support, (2) a broader strengthening of the health system, and (3) sufficient flexibility and dynamism of these community initiatives, it is highly unlikely that the above-cited positive impact on ART programme outcomes can be achieved in the long term [23]. The renewed attention to community mobilisation in high-HIV-prevalence, resource-limited settings is very welcome, but researchers and policy makers should continuously devote sufficient attention to creating an enabling context for these initiatives as well as guarding the necessary quality criteria.