Introduction
In the past few years, there have been increased movements by governments in low and middle-income countries (LMICs) to achieve universal health coverage (UHC) [
1,
2]. Under UHC, all people who need health services can receive them without undue financial hardship [
3]. UHC is a critical component of sustainable development and poverty reduction, and a key element of any effort to reduce social inequalities and enhance access to care [
4].
Many high-income countries that are either progressing towards or have achieved UHC have relied heavily on government or employer-based health insurance or a mix of both [
5]. However, in many LMIC, financing UHC has been difficult to achieve due to limited economic resources, modest economic growth, constraints on the public sector and weak institutional capacity of government [
6,
7].
Community-based health insurance (CBHI) has evolved as an alternative health financing mechanism to out of pocket payment in LMICs, particularly in areas where government or employer-based health insurance is minimal [
7‐
10]. CBHI operates by pooling risks and resources at the community level. In such schemes, individuals or households in a community voluntarily pay a predetermined amount of money in return for a benefit package consisting of health services [
11,
12].
CBHI aims to facilitate access to healthcare and increase financial protection against the cost of illness, particularly for underprivileged population [
13]. For instance, CBHI schemes have been implemented in low-income countries to insure rural population and informal workers that have been excluded from regular insurance schemes [
14,
15]. Evidence from systematic reviews indicate that CBHI schemes provide financial protection by reducing out-of-pocket expenditures and that such schemes improve resource mobilization and cost-recovery [
12,
13].
While CBHI schemes may hold strong potential to improve financial protection and enhance utilization among their enrolled populations, there is huge variation in the effects and coverage achieved [
13,
16]. This means that CBHI schemes are more likely to succeed under certain contexts and conditions [
12]. Thus, simply replicating an intervention from one setting to another is likely to fail without taking into consideration the factors critical to its implementation and sustainability [
17]. This, in turn, highlights a need to understand the contexts and conditions critical to the success of CBHI schemes.
Existing systematic reviews on implementation of CBHI schemes have focused on specific regions (i.e. South Asia) [
18] or on a subset of outcomes, primarily uptake of or willingness to pay for CBHI schemes [
19]. This systematic review adds to the extant reviews the following: given that our search includes studies published in all LMIC countries, we provide a much more global perspective than the South Asian alone. In addition, we identified all factors influencing implementation, enrollment, and sustainability of implemented CBHI schemes (and not proposed schemes), using an ecological perspective that takes into account the individual, interpersonal, community and systems level perspective. Findings from this systematic review can help inform the decisions of policymakers and stakeholders considering to implement CBHI within their own context.
Methods
Protocol and registration
We registered the review protocol in PROSPERO International prospective register of systematic reviews (ID = CRD42015019812).
Eligibility criteria
-
Study design: All studies that were eligible were peer reviewed publications or grey literature, published in English language and after year 1992. We included randomized trials, non-randomized studies (e.g., prospective studies, retrospective studies, before and after studies and cross-sectional studies), qualitative studies, process evaluation studies, policy analysis studies, and case studies.. We excluded editorials, commentaries, proposals, conferences, and systematic reviews. We also excluded policy analysis papers and case studies that lacked a clear methodology section.
-
Setting: low- and middle- income countries (as defined by the World Bank). The World Bank defines low- income economies as those with a Gross National Income (GNI) per capita of $1025, (U.S. dollars) and middle-income economies as those with GNI per capita between $1026 and $4035 [
20].
-
Interventions: community-based health insurance (CBHI) schemes. We excluded disease-specific schemes, vouchers, conditional cash transfer, social or national health insurance schemes or the extension of the latter two to the informal sector. We also excluded studies that looked at integration as opposed to implementation of specific programs. In addition, we excluded studies that focused on proposed CBHI schemes (i.e., the scheme was not implemented in an actual setting).
-
Outcome: barriers and facilitators to the implementation uptake and sustainability of CBHI schemes. We also included studies that described the process of implementation or assessed strategies to promote the implementation of CBHI schemes. Whenever available, we reported on interventions to overcome identified barriers. We excluded studies that assessed the impact of schemes on health and financial outcomes without considering factors contributing to the success or failure thereof. We also excluded studies that focused on payment methods or utilization of healthcare services in general without any linkage to CBHI schemes.
Search strategy
We searched the following electronic databases between December 2014 and January 2015: PubMed, MEDLINE, EMBASE, WHO Global Health Library, and Health Systems Evidence. We developed and validated the search strategy with the help of an information specialist. The strategies combined three different concepts: ‘health insurance scheme’, ‘barriers and facilitators’ and ‘low- and middle-income countries’. Additional file
1 provides the free text terms and MeSH (Medical Subject Headings) terms used to search the different electronic databases. We restricted searches to English language and from 1992 forward. We chose this start date because the concept of ‘health benefit packages’ took centre-stage in the debate when the 1993 World Development Report raised the question on how governments, especially in LMIC, should spend their limited health budgets [
21].
We complemented the electronic database searches with a variety of approaches to identify additional literature, including grey literature. We manually searched Google Scholar and the websites of relevant institutions like the World Health Organization (WHO) and the World Bank. We also screened the reference lists of included studies and relevant systematic reviews. In addition, we contacted the authors of conference proceedings that are of potential relevance.
Study selection
Prior to the selection process, and in order to enhance its reliability, all the reviewers participated in a calibration exercise using a randomly chosen sample of 150 citations. The selection process consisted of two stages, title and abstract screening and full text screening. Teams of two reviewers (RF, NH, RM, and CA) worked in duplicate and independently to screen the titles and abstracts of identified citations for potential eligibility. They obtained the full texts of citations judged as potentially eligible by at least one of the reviewers. Then, the teams of two reviewers screened the full texts independently and in duplicate. At this stage, the reviewers compared results and resolved disagreements by discussion or with the help of a third reviewer (FJ or EAA) if disagreement could not be resolved. They used standardized and pilot-tested screening forms. They documented the reason for study exclusion.
Data abstraction
We conducted calibration exercises on a randomly chosen sample to ensure adequate agreement. Teams of two reviewers (RF, NH, RM, CA and LH) abstracted data from eligible studies in duplicate and independently. They resolved disagreement by discussion or with the help of a third reviewer (if they could not reach an agreement).
They used a standardized data abstraction form to collect information on the following variables: study information (authors, year of publication, and study design), objective, methods (sample size and methods, timeframe, data collection, data analysis), population (sample population, setting), description of scheme (type of scheme, content of services covered, enrollment rate, unit of enrollment, source of fund, premium, cost-sharing, role of government, provider-payment method), socio-demographic factors, and reported barriers and facilitators.
Quality assessment
Two reviewers (RF, LH) assessed the quality of included studies in duplicate and independently. They resolved disagreement by discussion or with the help of a third reviewer.
We used Cochrane risk of bias tool to assess the risk of bias in randomized studies; a modified version of the Cochrane risk of bias tool, adapted from Alkhaled et al. (2014), to assess the risk of bias in non-randomized quantitative studies [
22]; the Critical Appraisal Skills Program (CASP) tool to assess the quality of qualitative studies; and a tool adopted from Niezen and Mathijssen (2014) to assess the methodological quality of mixed-methods studies that did not analyze quantitative and qualitative data separately [
23]. We did not exclude any study based on the results of the quality assessment. In this review, quality of primary studies is not as critical because we judged that every study may offer valuable insights on the various factors influencing CBHI [
24,
25].
Data analysis and synthesis
Given the heterogeneity in study design, settings, and outcome measures, we did not conduct meta-analysis. Instead, we synthesized the findings narratively, making use of both thematic [
26] and framework analysis [
27]. We used a slightly modified version of the Ecological Model framework to categorize emerging themes into the individual, interpersonal, community, and systems level [
28].
Data coding involved three phases: deduction (coding data and labeling each section), induction (screening data for new concepts or codes to emerge), and verification (verifying all coded data) [
27]. We reviewed the literature on CBHI schemes to generate an initial list of coding themes corresponding to each level of the ecological model (See Additional file
2). Then, the reviewers screened the “result” section of each included study and coded the findings under one of the predefined themes, while also allowing for new themes to emerge inductively. We iteratively updated the coding themes as we proceeded with data analysis [
29]. Throughout this process, team members with subject expertise were consulted to validate coding decisions and discuss emerging themes. We revisited and considered data in the context of any newly emergent theme. All studies were coded at least twice, once with the initial pre-defined list, and once with the finalized list of coding themes [
30]. We narratively present the main barriers to implementation, uptake, or sustainability of CHBI schemes and strategies that facilitated them, organized according to the Ecological Model framework into individual, interpersonal, community, and systems level.
Discussion
We identified 51 studies reporting on a range of barriers and facilitators to the implementation, uptake and sustainability of CBHI schemes across 22 countries. Many of the studies failed to meet methodological safeguards for protecting from bias, thus the findings should be interpreted with caution. Given the heterogeneity in quantitative study design and outcome measures, we could not conduct meta-analyses. Thus, we synthesized the findings narratively, and categorized according to the ecological model.
Although CBHI schemes have evolved rapidly in LMIC countries, many of these continue to be challenged by low uptake, coverage and sustainability. As evident from the findings of this review, there are a multitude of interrelated factors at the individual, interpersonal, community and systems level that drive the implementation and sustainability of CBHI schemes. These should be properly addressed in scheme design and implementation and harmonized across different levels of the ecological model to ensure proper attainment of scheme objectives and promote effective and equitable health systems. An overview of the factors influencing implementation, uptake and sustainability of CBHI schemes is presented in Fig.
2.
Two previously published systematic reviews focused on factors influencing CBHI enrollment: Bhageerathy et al. looked at the enrollment process, CBHI models, and health care seeking behavior in South Asia [
18], while Adebayo et al. focused on a subset of outcomes, specifically uptake of or willingness to pay for CBHI schemes in LMICs [
19]. Our systematic review provides a much more global perspective than the South Asian alone as well as attempts to identify all factors influencing implementation, enrollment, and sustainability of
already implemented CBHI schemes. Furthermore, we provide a conceptual framework of factors critical to the implementation, uptake and sustainability of CBHI schemes.
All three reviews pointed to the importance of involving the community in scheme development and implementation to increase enrollment and sustainability of schemes. In addition, they indicated that engaging the community in decision-making about the types of services, payment approach and service delivery increased satisfaction with services as these were tailored to the community needs. Our findings were also consistent with those by Adebayo et al. in terms of the negative influence of poor perceived quality of care, lack of trust, and lack of financial resources on CBHI uptake. However, unlike that review, we found a consistently negative correlation between long distance to health facility and enrollment or renewal of scheme membership (as reported in 17 studies). This could reflect the method used by Adebayo et al., whereby ‘willingness to pay’ was taken as a proxy indicator to enrollment. One critical area not covered by the findings of the two previous reviews was the role of government in CBHI schemes. Our review highlighted the important role of government in establishing the necessary legislative, technical and regulative support to ensure sustainability of CBHI schemes. Further, having a transparent, incorruptible, and honest governance were perceived as essential for trusting the scheme.
Implications for policy and practice
Policymakers and stakeholders interested in implementing CBHI schemes should first assess the specific characteristics and preferences of the community, including the approach to solidarity in the target population [
48]. This should be coupled with awareness and information campaigns on insurance concepts in general, and CBHI schemes in particular, to inform individuals about the scheme and promote its uptake. Policymakers and stakeholders could also consider creating opportunities for active participation of community members to enhance trust, accountability, and enrollment in scheme.
Implementation of CBHI schemes should go hand in hand with ensuring the necessary institutional and regulatory environment to steer health care providers’ behaviors. It is important for policymakers and stakeholders to consider how the current payment methods of CBHI schemes influence provider performance, and how changes in the methods could improve performance and support for the scheme [
80]. Further, strengthening policymaker-implementer relations and promoting a common language across stakeholders could help minimize conflicts and facilitate the implementation process.
Policymakers and stakeholders should also invest in efforts to address potential inequities that may arise with CBHI schemes, specifically in terms of enrollment and access to services. Possible policy options include: exempting the poor and most vulnerable populations from premium payment; providing premium subsidies; differentiating contributions according to socio-economic groups; adjusting contribution rates to reflect changes in benefits, health costs and inflations; and making the timing and modalities of premium collection flexible and tailored to the context. Furthermore, addressing geographical coverage of health facilities in scheme design and implementation is critical given its central role in determining people’s access to care.
To enhance sustainability of CBHI schemes, it would be important to balance strategies promoting enrollment and access, with strategies that could help minimize adverse selection and moral hazards typically associated with CBHI schemes. Policy options include using ‘household’ as the unit of enrollment, defining a minimum percentage of individuals that would be required before providing insurance, imposing a waiting period before services could be utilized, or establishing strong referral systems across the different levels of care. Whatever mechanism is selected, it is important to ensure that it is flexible, adapted to reality, and clearly defined in order to avoid deterring individuals from enrolling.
Finally, if CBHI schemes are to contribute to UHC, it would be critical to involve the government to provide the necessary legislative, technical, financial, and regulative support to implement CBHI schemes. Establishing a policy framework could help legitimize the CBHI scheme and position it within the context of national health financing systems. Consideration could also be given to establishing an “umbrella organization” that would provide support in design, training and information services as well as involve government, non-government and academia, as an integral part of the development and implementation process [
48]. This is especially relevant in light of a resurgence in discussions about universal health coverage as a key component of health-related Sustainable Development Goals [
81].
Strengths and limitations
Strengths of our methodology include pre-publishing a protocol, using rigorous and transparent process, and following standard methods for reporting systematic reviews [
82]. In addition, we conducted a comprehensive search of the published and grey literature to avoid potential publication bias. Furthermore, the inclusion of all types of study design allowed for a more comprehensive understanding of the issue at hand [
21].
This review has several limitations. First, we acknowledge that there may be some areas of overlap in the categorization of themes according to the ecological model. Moreover, despite our attempt to report the findings by implementation, uptake, and sustainability, it is important to note that their interrelatedness brought up a few challenges. For instance, uptake was also reported to influence sustainability of the scheme in few studies. Also, and in few cases, the distinction between implementation and sustainability of scheme was not very clear. However, we attempted to minimize this through continuous input from team members with subject expertise on coding decisions and characterization of emerging themes. Second, our findings may be more generalizable to low-income countries, which were the focus of 35 (out of 51) studies. Third, we only included studies conducted in English, thus we may have missed out on relevant studies published in other languages. Also, despite our attempt to search the grey literature, we may have still missed potentially relevant studies published in other donor and governmental websites beyond the ones searched for this review. A final limitation is that the review does not incorporate studies that could have been published after the date of our search. However, it is unlikely that such studies would change the findings in a significant way.
Conclusion
There are a multitude of interrelated factors at the individual, interpersonal, community and systems levels that drive the implementation and sustainability of CBHI schemes. These should be properly addressed in scheme design and implementation and harmonized across the different levels to ensure attainment of scheme objectives. Future research efforts should be directed towards conducting well-designed primary studies with particular attention to recruitment strategy, use of validated tools, and control for potential confounding variables. Furthermore, more research is needed on how CBHI schemes could complement the broader health financing system to progress to UHC.