Final Report Summary - CBHI INDIA (Developing efficient and responsive community based micro health insurance in India)
Executive Summary:
The objective of this project was to examine a new model of community based health insurance (CBHI) that is affordable, responsive and inclusive, designed to increase equitable access to health care and to provide financial protection. For that purpose three CBHI schemes were set up in northern rural India. These schemes are characterized by the strong involvement of the target group in the design of the insurance package. The insurance was offered to a population of self-help group members (all women) and their households. These self-help groups already existed and were engaged in micro finance activities. Furthermore a community rated premium applied to all, and the schemes operated without any premium subsidies. Benefit packages varied across sites, reflecting local preferences. As a consequence also the premium differed between locations.
The CBHI schemes were launched in 2010 under the oversight of the Delhi-based Micro Insurance Academy, in partnership with three local NGOs in Kanpur Dehat and Pratapgarh districts in Uttar Pradesh and in Vaishali in Bihar. We used a randomized rollout of the schemes in three separate waves to identify their impact on health care utilization and financial protection. In each of the waves, one-third of the clusters (randomly drawn in each site) were offered the possibility of enrolling in the CBHI scheme. By the end of the project, the entire target population had been offered a chance to become insured.
During the first wave of implementation in 2011, from those offered insurance, 39% of the households had at least one individual who purchased insurance (23% of individuals), while the numbers for the second wave were 45% at the household level (24% of individuals). Dropout rates were quite considerable with 54% of the households (42% of individuals) who enrolled in the first wave renewing in the second, followed by a renewal of 25% of those originally enrolled households (16% of individuals) during the third wave. Enrolment and renewal in CBHI were independent of households’ socio-economic status (social equalizing effect). Furthermore we found no evidence of adverse selection, women were overrepresented among those enrolled and households with greater financial liabilities found insurance more attractive. Finally, making a claim and better knowledge of the insurance concept increased the likelihood of renewal.
For those who actually enrolled in the CBHI schemes the probability of seeking outpatient care from formal sources in Kanpur Dehat increased substantially, while the opposite was true for Pratapgarh where CBHI membership caused a drop in the probability of seeking care. No changes in the use of outpatient care were noted for members in Vaishali or for inpatient care in any of the three sites. In line with healthcare utilization, participation in CBHI had large negative impacts on out-of-pocket payments for members in Pratapgarh, driven by lower utilization, while the opposite pattern was found for Kanpur Dehat, though statistically insignificant. This resulted in a lower probability of hardship financing in Pratapgarh, driven by lower healthcare utilization. In sum, these results suggest limited success of the CBHI schemes on healthcare use and financial protection, with offsetting effects between two sites.
The negative effects in Pratapgarh are surprising and may be related to the capitation payment system applied under CBHI, which may incentivize Non-degree Allopathic Providers (NDAPs) to provide lower amounts of care or drugs as compared to a situation before CBHI when they were paid on a fee-for-service basis. Qualitative field work revealed that a number of insured respondents were dissatisfied with the NDAPs affiliated with the CBHI scheme as they had been providing insufficient care and medications as a result of which many had to pay additional fees or preferred to seek care elsewhere. This is unfortunate because NDAPs are the most accessible medical providers in the community and first contact points for acute illness episodes.
Qualitative work revealed that members perceived value in CBHI on financial aspects, access and quality of healthcare services, and “peace of mind” due to better social cohesion/solidarity. The CBHI schemes were viewed as providing three kinds of access: financial access, physical access (proximity of NDAPs and availability at all hours), cultural access (familiarity, respectful conduct and non-discrimination by providers).
In sum, the results of this project have shown that adapting the insurance product to meet local needs and raising awareness of the insurance concept, can indeed lead to relatively high and equitable uptake of a voluntary insurance scheme. But the results also suggest that in the absence of subsidised premiums, voluntary CBHI schemes operated by local NGOs offer limited benefit packages and face managerial challenges, especially related to the reimbursement of providers, which limits the impact of such schemes on access to care and financial protection. Furthermore, they need to be designed carefully to avoid negative effects on perceived quality of health care supply. Constant monitoring of effects and consequent adaptation of arrangements is necessary.
Project Context and Objectives:
1. Overall Objective and Design of the project
Examine a new model of CBHI that is affordable, responsive and inclusive, designed to increase
(i) Equitable access to healthcare and
(ii) Financial protection.
Specific objectives are:
1.Establish 3 CBHI schemes (each one covering about 20 villages) in 2 states in India.
2.Monitor and evaluate the impact of CBHI on
a.Equitable access to health care; this includes two aspects:
i. Income and age related equity of access
ii. Gender related equity of access.
b.Financial protection from catastrophic cost of illness.
c.The quality and quantity of health care supply.
d.The effect of CBHI on the understanding of the concepts of insurance, solidarity and social action among the insured
3.Field test and document (cost) efficient management of CBHI.
4.Develop and disseminate policy recommendations on implementing and up scaling Community Based Health Insurance.
Three CBHI schemes were set up in Northern rural India. To evaluate the effects of being insured, the enrolment followed step-wise clustered randomized control trials (RCTs) methodology. The CBHI schemes were introduced in 2010 under the oversight of the Delhi-based Micro Insurance Academy, in partnership with three local NGOs in Kanpur Dehat and Pratapgarh districts in Uttar Pradesh and in Vaishali in Bihar. These two states have been characterized as amongst India’s most populated and least educated with large gender disparities.
A new model of CBHI was implemented which was characterized by the strong involvement of the target group in the design of the insurance package and the creation of the ground structure of insurance from among the group of insured (Mutual-aid model). The insurance was offered to a population of self-help group members (all women) and their households. Each household could enroll if at least one female (who was also a member of a self-help group) joined; other household members could join, or not. Furthermore a community rated premium applied to all, and the schemes operated without any premium subsidies.
Benefit packages varied across sites, reflecting local preferences. The benefit-package selection process entailed four steps: narrowing down the options by community representatives, plus three rounds of facilitation of consensus (using the Choosing Healthplans All Together (CHAT) simulation game) conducted among female members of self-help groups. Disagreement on the benefit package decreased consistently over consecutive CHAT rounds, reaching zero (ie, 100% consensus) in two locations, and confirmed gradual adoption of consensus. Evaluation interviews revealed that the wish to be part of a consensus was dominant in all locations. The in-depth interviews indicated that people enjoyed the participatory deliberations, were satisfied with the selection, and that group decisions reflected a consensus rather than majority. The premium also differed between locations.
We used the randomized rollout of the schemes in three separate waves to identify their impact on health care utilization and financial protection, while distinguishing between outpatient care and hospitalizations. In each of the waves, one-third of the clusters (randomly drawn in each site) were offered the possibility of enrolling in the CBHI scheme. By the end of the project, the entire target population had been offered a chance to become insured.
The following published papers contain more details:
Doyle, Conor; Panda, Pradeep; Van de Poel, Ellen; Radermacher, Ralf and Dror, David M.: Reconciling Research and Implementation Needs in Cluster Randomised Controlled Trials: Lessons from 3 Micro Health Insurance Experiments in India. Trials, 12, 224,1-15 http://www.trialsjournal.com/content/12/1/224
Dror, David M.; Panda, Pradeep; May, Christina; Majumdar, Atanu and Ruth Koren (2014). ‘“One for all and all for one”; Consensus-building within communities in rural India on their health microinsurance package’. Risk Management and Healthcare Policy, 7, pp. 139-153, August.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4128598
Project Results:
2. WP 1 Quantitative Research
2.1 Healthcare Seeking Behaviour and the economic consequences of ill-health among Self-help Group Households in Rural Bihar and Uttar Pradesh
In recent years a number of community-based health insurance (CBHI) schemes have been functioning in rural India, supported by non-governmental organizations (NGOs). Such schemes design their benefit packages according to local priorities. We examined the healthcare seeking behaviour of the intended CBHI enrolees (before any enrolment into CBHI) and observed that most of the population does access some form of care. There appeared an overwhelming use of private care for both outpatient and inpatient services. Moreover, non-degree allopathic providers (NDAPs) accounted for a substantial proportion of private care (61 percent of total private use in acute cases and 35 percent in the case of chronic illnesses). Both the self-reported information on health-seeking choices and the econometric estimates confirmed that the main reason for using NDAPs is their proximity and convenience. Results also revealed that higher household consumption, non-SC/ST (scheduled caste and scheduled tribe) status and education were associated with an increase in the probability of using care. Additionally, there is evidence of gender differences in access to care, at least in the case of acute illnesses with men more likely to access care and to seek private care.
The household survey also contains information on various negative events that could affect households’ welfare. We examined the relative importance of health related adverse events (prior to any CBHI schemes operating) and found that these occured very frequently (34% of the households), second only to exposure to natural disasters (51% of the households). Crop and livestock disease and weddings affected about 8% of the households. Only a fourth of households reported to have recovered from illness and/or death in the family (by the time of the survey). Most of the economic risk of ill-health was reflected in healthcare expenditures, and indirect costs were also not negligible. Chronic conditions accounted for close to half of health expenditures. Households tried to cope with health-related expenditures mostly by dissaving, borrowing and selling assets. Few households reported having to reduce (food) consumption in response to ill-health. While most households seemed able to smooth consumption in the short term, coping strategies like selling assets and borrowing from money lenders are likely to have severe long term consequences.
In sum, analysis of our baseline data revealed health shocks to be an important threat to household welfare, and most of the economic risk from ill-health to be related to out of pocket spending. Introducing health insurance may therefore contribute significantly to alleviate economic hardship of families in rural India.
More recent analysis using the 2012-2013 panel data on the economic consequences of ill health indicates that declines in self-assessed health lead to a reduction in prime-ager labour supply and sharply rising OOP health expenditure, irrespective of health insurance enrolment. Despite these negative impacts, health shocks did not decrease non-health consumption in the period shortly after the shock. Instead, findings show that households compensate with a strong increase in debt. Results do not show evidence for short-term tapping of other coping strategies but speculate that heavy reliance on high-interest debt-financing leads to reductions in investment and thus a perpetuation of poverty in the longer run.
2.2 The impact of awareness campaign and preventive health campaign on the insured households
Prior to enrolment and scheme launch, the treatment population went through a structured insurance education program (awareness campaign), which imparted knowledge about and concepts of health insurance and CBHI, including operational aspects of the schemes. It was found that the treatment cohort demonstrated substantial and significantly higher understanding of insurance concepts compared to the control group, and that understanding of CBHI was a positive determinant for enrolment. Of the awareness-enhancing tools, the “Treasure-Pot” (an interactive game) was viewed by respondents as most effective, followed by household discussions.
For financial sustainability of the scheme, as the premiums are low and not subsidized, the scheme should have a sizeable number of members and controlled claims ratio. This can be achieved by increasing the preventive health behaviour among the target population which might reduce the incidence rate of illnesses. A preventive health awareness campaign was conducted in the study locations. Analysis of data before and after the campaign shows a statistically significant increase both in awareness and in preventive practices related to airborne, vector-borne and water-borne diseases, suggesting that the awareness campaign was effective. Affiliation to CBHI had significant positive influence on awareness and on practice scores in the post-campaign period.
2.3 Enrolment in Community Based Health Insurance Schemes in Rural Bihar and Uttar Pradesh
During the first wave of implementation in 2011, from those offered insurance, 39% of the households had at least one individual who purchased insurance (23% of individuals), while the numbers for the second wave were 45% at the household level (24% of individuals). Dropout rates were quite considerable with 54% of the households (42% of individuals) who enrolled in the first wave renewing in the second, followed by a renewal of 25% of those originally enrolled households (16% of individuals) during the third wave.
The household’s socio-economic status did not appear to substantially inhibit uptake. In some cases scheduled caste/scheduled tribe households were more likely to enrol. Also females in the households that joined had higher education than in those that did not join. Second, households with greater financial liabilities found insurance more attractive. Third, access to the national hospital insurance scheme Rashtriya Swasthya Bima Yojana did not dampen CBHI uptake (but had some impact on the type of benefits chosen), suggesting that the potential for greater development of insurance markets and products beyond existing ones would respond to a need. Fourth, recent episodes of illness and self-assessed health status did not influence uptake. Fifth, insurance coverage is prioritized within households, with the household head, the spouse of the household head and both male and female children of the household head, more likely to be insured as compared with other relatives. Also females were overrepresented in households covered. Sixth, offering insurance through women’s SHGs appeared to mitigate concerns about the inclusiveness and sustainability of CBHI schemes.
Regarding renewal we observed that the economic status of the household did not appear to considerably impede renewal. Benefit packages of the schemes appeared to play a decisive role for retention. Receiving money from the scheme through the claim process in the past cycle and CBHI understanding turned out to have a positive and significant influence on renewal. Subscription to Rashtriya Swasthya Bima Yojana did not affect renewal in CBHI.
2.4 The impact of CBHI on access to health care and financial protection
The main objectives of CBHI are the provision of equitable access to healthcare and financial protection from expenses that deplete resources to the insured households. To measure the impact of the CBHI on the aforementioned outcomes, we distinguished between outpatient and inpatient care, where the outpatient care is further disaggregated to acute and chronic conditions. Using three-round panel data, we investigated the CBHI’s effects on healthcare utilization and financial protection at two levels: the impact of its offer (intention to treat effects, ITT) followed by the actual uptake (average treatment effect on the treated, ATT). So ITT investigates the effect on all persons offered CBHI and ATT on those that actually took it.
We began by estimating the effects (both ITT and ATT) of CBHI on the probability of seeking any care conditional on reporting an illness. Subsequently we investigated the effect of enrolment on whether or not care (for the first visit) is sought from a formal provider, defined as Non-degree Allopathic Providers (NDAPs) or qualified doctors (public/private general practitioners and specialists). Subsequently, we looked at the choice of formal provider (NDAPs or qualified doctors) conditional on use to explore whether the CBHI schemes precipitate changes in the choice of healthcare provider. For inpatient care, we first modelled the probability of being hospitalized followed by whether the individual seeks private or public hospital care (conditional on hospitalization).
Regarding financial protection, we looked at the effect of insurance on direct out of pocket healthcare costs (consultation fees, costs of medicine and lab/imaging tests) conditional on reporting an illness and on hospitalization. We defined hardship financing as an individual reporting having to borrow from high interest rate lenders, cutting back on essential costs or selling off assets to service the costs in case of an ill health event.
Using a conditional logistic model, the analysis of ITT effects showed a significant reduction in the probability of seeking formal care in Pratapgarh (driven by a drop in the utilization of GPs or specialists) while the opposite was found to be true for patients in Kanpur Dehat, driven by an increase in the utilizations of NDAPs. Intention to treat effects on healthcare spending and financial protection were limited.
Using the randomized offer of insurance as an instrument for the analysis, the effects of actual enrolment into the CBHI scheme were in line with this. While there was no discernible impact on healthcare utilization at the aggregate level, results revealed significant offsetting results with substantial magnitudes for Kanpur Dehat and Pratapgarh. While CBHI membership led to a surge in the probability of seeking outpatient care from formal sources in Kanpur Dehat, the opposite was true for Pratapgarh where CBHI membership caused a drop in the probability of seeking care. No changes in the use of outpatient care were noted for members in Vaishali or for inpatient care in any of the three sites. In line with healthcare utilization, results showed that participation had large negative impacts on out-of-pocket payments for members in Pratapgarh, driven by lower utilization, while the opposite pattern was found for Kanpur Dehat, though statistically insignificant. This resulted in a lower probability of hardship financing in Pratapgarh, driven by lower healthcare utilization. For inpatient care, the lack of tangible effects of the CBHI scheme is likely to be related to the small sample size given the infrequency of hospitalizations in the target population. Furthermore, coverage for inpatient care could be considered relatively shallow and physical accessibility to the healthcare facilities within the catchment area of this program is relatively low.
In sum, these results suggest limited success of the CBHI schemes on healthcare use and financial protection, with offsetting effects between two sites. The negative effects in Pratapgarh are surprising and may be related to several factors. Most importantly, the capitation payment system could incentivize NDAPs to provide lower amounts of care or drugs as compared to a situation before CBHI when they were paid on a fee-for-service basis. Qualitative field work (see section 4) revealed that a number of insured respondents were dissatisfied with the NDAPs affiliated with the CBHI scheme as they had been providing insufficient care and medications as a result of which many had to pay additional fees or preferred to seek care elsewhere. This suggests that lack of control of the quality of care provided and that the incentives created by the provider payment system might be contributing to the limited impact of CBHI, and low renewal rates. Furthermore, a low degree of competition between providers might exacerbate the incentives to underprovide in a capitation system. Indeed the density of NDAPs was lowest in Pratapgarh (e.g. in wave 2 the ratio of NDAP/village equaled 0.28 in Pratapgarh while it was 0.5 in the other two sites).
The overall results merit further investigation into the heterogeneity of impact of CBHI across gender and age groups. Preliminary results from ongoing work shows a strong decline in healthcare expenses for women whereas no changes were evident for the men. Outcomes such as hardship financing and healthcare utilization follow similar patterns across gender as it does with the whole sample. Heterogeneity across age was investigated by groups: children (0-13yrs), working aged adults (14-55yrs) and the elderly (above 56yrs). Results show that they are largely in line with that of the main results.
2.5 The impact of CBHI on health, wealth and coping behaviour.
Our first impact study shows that while CBHI increases healthcare utilization, impacts on financial protection are ambiguous and context specific. A second impact study uses an identical empirical approach to extend the analysis to impacts on
- Self-assessed health – improved health is a key indicator of health insurance effectiveness that is, however, rarely considered in insurance impact studies because of data constraints. We contribute to this literature by assessing impacts of CBHI on an EQ5D-based index, on individual’s self-ranking on a 5 category health scale, and on the reporting of acute and longer-lasting conditions.
- Non-healthcare consumption – spending on medical care and illness-induced income loss burden household finances. As a result, consumption opportunities reduce, and illness perpetuates poverty. CBHI can contribute to breaking this cycle by a) reducing OOP healthcare expenditure and b) avoiding/reducing illness-related income loss through better health. An extensive non-health consumption survey module permits us to test if CBHI is effective in protecting non-healthcare consumption. We hereby not only estimate impacts on total non-health consumption but also on items of intertemporal importance like food consumption and investments in agriculture and education.
- Debt levels and household assets – in the absence of social insurance, LMIC households revert to various strategies to cope with the negative economic consequences of illness. If increased OOP health expenditure and income loss cannot be compensated through cash income and savings, households with access to informal or formal sources of credit often borrow, or pawn productive assets. While such strategies may enable households to pay for medical costs and maintain consumption levels in the short-run, they likely have negative intertemporal consequences on household welfare. A key benefit of functional health insurance is to avoid harmful coping strategies. We are able to test if CBHI is effective in this regard through a variety of outcomes including current debt levels, land ownership and livestock holdings. Again, given the ambiguity of financial protection effects demonstrated in the first impact study, any positive impacts of CBHI on debt and assets would mainly be driven by improved health.
We are currently conducting data analysis for this project and plan to provide a first draft by the end of October 2014.
The following papers have been published or submitted for publication:
- Panda, Pradeep; Chakraborty, Arpita; Dror, David M. and Bedi, Arjun S. (2013). Enrollment in Community Based Health Insurance Schemes in Rural Bihar and Uttar Pradesh, India. Health Policy and Planning, doi: 10.1093/heapol/czt077.
- Dror, David M. and Firth, Lucy A. (2014). The Demand for (Micro) Health Insurance in the Informal Sector, Geneva Papers on Risk and Insurance, doi: 10.1057/gpp.2014.24
- Raza, Wameq A.; Van de Poel, Ellen; Panda, Pradeep; Dror, David M. and Bedi, Arjun S. (2013). Healthcare Seeking Behaviour among Self-help Group Households in Rural Bihar and Uttar Pradesh, India. Working Paper No. 575, International Institute of Social Studies, Erasmus University, Rotterdam, the Netherlands, December (ISSN 0921-0210).
- Quintussi, Marta; Van de Poel, Ellen; Panda, Pradeep and Rutten, Frans: Economic consequences of ill-health for households in northern rural India, submitted (invited to revise and resubmit at BMC Health Services Research).
- Panda, Pradeep; Chakraborty, Arpita; and Dror, David M. (2014). Building Awareness to Health Insurance among the Target population of Community-Based Health Insurance Schemes in Rural India, submitted
- Raza, Wameq A.; Van de Poel, Ellen.; Bedi, Arjun S. and Rutten, Frans (2014) Impact of community based health insurance on access to care and financial protection in rural India: evidence from three randomized control trials. submitted
- Panda, Pradeep; Chakraborty, Arpita and Dror, David M. (2014). Mobilizing Community-based Health Insurance to Enhance Awareness and Prevention of Airborne, Vector-borne and Waterborne Diseases in Rural India, submitted
Work in progress:
- Panda, Pradeep; Chakraborty, Arpita; Dror, David M.; Raza, Wameq A.and Bedi, Arjun S. (2014). Dynamic Process of Renewal in Community Based Health Insurance in Rural India
- Dror, David M.; Majumda, Atanu; Chakraborty, Arpita; Panda, Pradeep; Koren, Ruth (2014). The Impact of Membership in CBHI in Rural India on Healthcare Seeking and Financial Protection
- Neelsen, Sven; Raza, Wameq A.; Van de Poel, Ellen (2014) Effects of ill-health shocks on consumption and coping strategies.
- Neelsen, Sven; Raza, Wameq A.; Van de Poel, Ellen (2014) Impact of CBHI on health, wealth and coping behaviour
- Raza, Wameq A.; Panda, Pradeep; Van de Poel, Ellen (2014) Determinants of enrolment and the impact of RSBY.
3 WP 2 Provider survey
3.1 Spatial Access to Inpatient Health Care in Northern Rural India
Access to health care in rural areas is a major concern for local populations as well as for policy makers in developing countries. In order to measure spatial access, impedance-based competition using the Three-Step floating Catchment Area (3SFCA) method, a modification of the simple gravity model, was used. 3SFCA was chosen for the study of the districts of Pratapgarh and Kanpur Dehat in the Uttar Pradesh state and Vaishali in the Bihar state, two of India’s poorest states. This approach is based on discrete distance decay and also considers more parameters than other available methods, hence is believed to be a robust methodology.
It was found that Vaishali district has the highest spatial access to in-patient health care followed by Pratapgarh and Kanpur Dehat. There is serious lack of health care, in Pratapgarh and Kanpur Dehat with 40% and 90% of the villages having shortage of in-patient care facilities in these respective districts. We conclude that, firstly, serious spatial access disparities existed between two poorest states of India; secondly, distance to nearest urban centre is negatively associated with spatial access; and lastly equal spatial access to care requires needs-adjusted allocation of health care facilities.
3.2 Spatial Patterning of Rural Medical Practitioners: A Comparison of Three Northern Rural Areas in India
Many rural areas in developing countries lack access to qualified out-patient (OPD) care providers. The gap left by qualified doctors is filled by unqualified persons also known as Non-Degree Allopathic Practitioners (NDAP). In India, NDAPs are dominant providers in rural areas. Despite their dominance, no studies were made to quantitatively estimate their prominence and distribution, especially in our study area.
We used a mixed-method approach, combining exhaustive quantitative census of all health care providers along with Geographic Information System (GIS) to address the knowledge gaps. Longitudinal quantitative and spatial survey was conducted for three consecutive years (2010, 2012 & 2013) for three districts Pratapgarh, Kanpur Dehat in Uttar Pradesh and Vaishali in Bihar, two of India’s poorest states. This study reports that NDAPs locate their practice primarily to meet the health needs of the community that lacks access to formal health care providers.
The three sites studies vary in terms of spatial distribution of NDAPs, and in terms of their demography. After accounting for the demographic heterogeneity, the picture is similar in all three sites: that NDAPs are profit oriented, and more demand produces more NDAPs.
The following papers have been published or submitted for publication:
- Dixit, Shikha and Panda, Pradeep (2013). Spatial research methodology supplementing cluster randomized control trials: learning from a study of community-based health insurance schemes in India. International Journal of Geoinformatics. Vol. 9, No. 3, pp. 31-39.
- Ranga, Vikram and Panda, Pradeep (2014). Spatial Access to In-patient Health Care in Rural Northern India. Geospatial Health, 8(2), pp. 545-556.
Work in progress:
- Ranga, Vikram and Panda, Pradeep (2014). Spatial Patterning of Rural Medical Practitioners: A Comparison of Three Northern Rural Areas in India
4 WP 4 Qualitative survey
4.1 Non-degree allopathic practitioners as first contact points for acute illness episodes: insights from a qualitative study in rural Northern India,
In 2005, the Indian government launched the National Rural Health Mission (NRHM) to improve the quality of and access to rural public health care. Despite these efforts, recent evidence shows that the rural poor continue to consult primarily private non-degree allopathic practitioners (NDAPs) for acute illness episodes. To examine this phenomenon, we explored the perception of rural poor people and their utilization of the rural health care system, notably the role of NDAPs therein.
Our study is based on qualitative data from focus group discussions conducted in 2009/2010 in three rural districts in Bihar and Uttar Pradesh, two high-focus states of the NRHM in Northern India. Our study population consists of female members of self-help groups and their male household heads. We applied a directed content analysis and use a theoretical framework to differentiate between physical, financial and cultural access to care.
Our study population distinguishes between “home treatment” (informal self-care), “local treatment” (formally unqualified care) and “outside treatment” (formally qualified care). Local NDAPs are usually the first contact points for patients before turning to qualified practitioners; they are physically, financially and culturally accessible, because of their proximity, flexible payment options and familiarity with patients’ belief systems, among other things. They treat minor illnesses, provide first relief, refer patients to other providers and administer formally prescribed treatments.,
Our findings are similar for all three study sites and reinforce recent findings from southern and eastern India. The understanding among the poor, and their utilization patterns of the rural health system deviate from governmental ideas. NDAPs are the most accessible medical providers and first contact points for acute illness episodes because they are embedded in the community. Thus, they de-facto fulfill the role envisaged by the Indian government for accredited social health activists introduced as part of the NRHM.
Based on the findings of our study, we concluded that it is necessary for community-based health insurance schemes wishing to offer coverage for outpatient care to integrate these local, private non-degree allopathic practitioners into their insurance portfolio despite their lack of formal qualification, as not doing so would mean to ignore the reality of health care seeking behavior in rural Indian communities. On a more general level, we also concluded that the Indian government should regulate, qualify and integrate local, private non-degree allopathic practitioners as part of the existing public health care system.
4.2 Clients’ Perceived Value in Health Microinsurance – Some Insights from Community-Based Health Insurance Schemes in Rural Northern India,
CBHI is considered as an innovative risk protection tool for the poor. However, despite its potential benefits, demand for CBHI is still low. We tried to gain a deeper understanding of clients’ perceived value in CBHI schemes in rural northern India. We used longitudinal household case studies conducted in all three study sites over a 16-months period after the introduction of the CBHI schemes which included 41 households. This does include pre-purchase and post-purchase phases so that the dynamics of clients’ perceived values can be traced.
We found that many clients value financial protection from catastrophic cost of illness. This is not only of importance for them in financial terms, but also with regard to their state of mind. Financial protection is closely knit to peace of mind for many clients. Yet there is also a big part of clients who value financial protection from costs of minor, everyday illnesses at least as much as protection from catastrophic cost of major but rather rare illnesses. Reasons for this may be informal risk management strategies in case of catastrophic cost of illness, limitations in coverage through the scheme or low probability of illness that causes lower valuing.
We found that clients particularly value the quality and quantity of health care supply when experiencing these (or not), i.e. in the post-purchase phase. When being insured, the quality of health care supply (both by practitioners and as medicines) becomes of high value for the clients and is one of the most important factors for them to stay enrolled or drop out of the scheme.
Based on the findings on which effects are perceived as important by the clients and which are perceived as less important, policy recommendations on how to implement and sustain CBHI schemes can be given. In particular, due to observed dynamics in client value, we recommended assessing client value not only before the start of the scheme (i.e. in the planning phase), but to assess client value also while the scheme is running and possibly adopt scheme’s features to what is valued by the clients. Moreover, we recommended to focus not only on financial and functional dimensions of client value, but to also include social and emotional dimensions.
4.3 Accessing Local Outpatient Care in Community-Based Health Insurance Schemes – Insights From Rural India,
Based on longitudinal household case studies conducted in all three study sites over a 16-months period after the introduction of the CBHI schemes, we examined the access to outpatient care from local, non-degree allopathic practitioners (NDAPs) in three community-based health schemes in rural northern India. We identified three types of outpatient services utilization in the insurance scheme: no utilization, combined/integrated utilization and exclusive utilization.
Insured households appreciate having access to outpatient treatment without having to pay out-of-pocket and to be able to access treatment for outpatient care immediately. On the negative side, they perceive that the quality of care provided by CBHI-associated practitioners is low and not suitable for all illnesses, and that they are not always easily accessible. In some ways, the involvement of local, non-degree allopathic practitioners (NDAPs) in CBHI was a success, as they are situated closely and are familiar to their patients. When patients were not satisfied with their treatment, they were replaced, which was appreciated by insured households. In other ways, however, advantageous features of NDAPs were compromised by organizational arrangements within CBHI: it brings about a reduction of medicine dispensed, only limited access to services available and a restriction to selected practitioners within the insurance.
The insurance schemes – as external entities – have partly regulated and formalized the relationship between NDAPs and their communities, with sometimes negative effects. We argued that the involvement of NDAPs in CBHI is necessary due to their popularity and accessibility for the rural population. From the point of view of insured households, the CBHI schemes under study have been successful to some degree in providing improved financial access to these providers and in preserving their favorable characteristics in physical and cultural accessibility. However, not all positive features could be sustained and it became obvious that schemes need to be organized carefully to not compromise the accessibility of (perceived) quality care by NDAPs.
From the data presented in this paper, we derived that the following interventions would be necessary to improve access to quality outpatient care in the community-based health insurance schemes under study:
- Increase the medicines dispensed by associated providers, e.g. by increasing the payment to NDAPs, reducing the unit prices by coordinating the purchase of medicines in bulk or arranging outside funding.
- Increase the number of NDAPs involved in CBHI to (i) reduce the distance to the next CBHI-associated provider and (ii) provide alternatives if NDAPs are not available at the time of need.
- Thoroughly inform households about which services are covered under CBHI and which not and the reasons for this (such as the rise of premiums for more expensive services or the risk of inappropriate handling of injections).
The consideration of the above points could also help when implementing or up-scaling community-based health insurance schemes. On a more general note, we concluded that CBHI schemes can have positive effects on access to local outpatient care, but need to be designed carefully to avoid negative effects on perceived quality of health care supply. Constant monitoring of effects and adaptation of arrangements if needed is necessary.
The following papers have been published or submitted for publication:
- May, Christina; Roth, Katja and Panda, Pradeep (2014). ‘Non-Degree Allopathic Practitioners as First Contact Points for Acute Illness Episodes: Insights from a Qualitative Study in rural Northern India’. BMC Health Services Research, 14:182. doi:10.1186/1472-6963-14-182.
- Dror, David M.; Panda, Pradeep;, May, Christina; Majumdar, Atanu; and Koren, Ruth (2014). ‘“One for all and all for one”; Consensus-building within communities in rural India on their health microinsurance package’. Risk Management and Healthcare Policy, 7, pp. 139-153.
- Roth, Katja; Panda, Pradeep; May, Christina and Ghosh, Sudeshna (2014). ‘Clients’ Perceived Value in Health Microinsurance: Some Insights from Community-based Micro Health Insurance Schemes in Rural Northern India’, submitted.
Work in progress:
- Ghosh, Sudeshna; Panda, Pradeep; May, Christina and Roth, Katja (2014). ‘The role of health microinsurance in the context of risk management strategies in northern rural India: Insights from a qualitative study’.
- May, Christina; Panda, Pradeep; Roth, Katja and Ghosh, Sudeshna (2014). ‘Accessing Local Outpatient Care in Community-based Micro Health Insurance Schemes: Insights from Rural India’.
Potential Impact:
5 Dissemination and impact
Efforts to disseminate the main findings, insights and policy ramifications have taken several forms besides scientific publications, and have led to several notable results and recognitions.
Invitations to contribute chapters in reference books
• Microinsurance Compendium II 2012
• Encyclopaedia of Health Economics
Dissemination via internet and online knowledge networks
We posted regularly all published materials relating to our project and to the implementation model to several websites, notably
• Social Science Research Network (SSRN),
• Google Scholar
• Research Network
• Academia.edu
• Microinsuraneacademy.org
• Microinsurancenetwork.org
A partial count of browses, downloads of full texts and citations of David Dror’s publications is provided here as an indication of the huge dissemination effect achieved (on 15 August 2014, the count was as follows: 67,284 browses; 13,922 downloads of full text articles; and 1,000 citations by other researchers). The details are provided in the attached PDF.
The project and its findings have been presented at over 40 international events, a detailed list of events is uploaded in the online system.
Awards
• Dr. David M. Dror was bestowed the Insurance personality of the Year 2009 Award by the Asia Insurance industry
• MIA got Educational services provider of the year in 2010 and 2011
• Two months research scholarship award to Dr. Pradeep Panda by ICSSR-NWO in 2012.
• Dr Ellen van de Poel received a VENI-scholarship in The Netherlands in 2011
Exposure visits
•Meetings with other donors that support other CBHI schemes (Nepal, Orissa)
•In April 2011, a high level delegation, comprising researchers and policy makers from Nigeria, visited our project sites.
•In September 2011, three researchers from University of Cologne (Hans Jürgen Rösner, Katja Roth and Christina May) visited the project sites in India.
•As part of the yearly India Week in Cologne, the Department for Cooperative Studies presented the CBHI India project to professors, students and other interested public with a poster and other information material on the University’s India Day on June 08th 2011.
•From September 19th to October 3rd 2011, the Department for Cooperative Studies, University of Cologne, and the Micro Insurance Academy together with the Professional Centre of the University of Cologne organized a Summer School for students on “Micro Health Insurance in Developing Societies – Protection for the Poor?”.
•In February/March 2012, one researcher from Erasmus University Rotterdam (Wameq Raza) visited India and the project sites.
•In February/March 2012, four researchers from University of Cologne (Hans Jürgen Rösner, Katja Roth, Christina May and Marta Quintussi) re-visited India and the project sites.
•In April 2012, a high level delegation, comprising researchers and policy makers from Bangladesh, visited our project sites.
•In February - March 2013, one researcher from Erasmus University Rotterdam (Wameq Raza) re-visited project sites to take part in the training and quantitative field survey with the MIA quantitative researchers.
•In March 2013, an American journalist (Aimee Ginsburg) working for an Israeli newspaper, visited the project site in Vaishali, India and met the implementing partner and a group of insured people and the claim committee members to understand the operational aspects of the CBHI.
•In March 2013, one researcher from University of Cologne (Christina May) re-visited India. She met with some insured and non-insured households under the CBHI schemes for the qualitative data collection taking place during that time with the MIA researchers.
•In December 2013, Christina May (University of Cologne) visited India for her collaborative research work.
•In March 2014, journalist Mr. Benjamin Fernandez, visited the CBHI project site in Vaishali along with Dr. David Dror.
Systematic Review
The Micro Insurance Academy is currently leading a Systematic Review project entitled “What factors affect take-up of voluntary and community-based health insurance schemes in low- and –middle income countries?” Through this project we are synthesizing the experience and evidence gained from CBHI schemes not only in India but also in other countries. This project is funded by the 3ie (International Initiative for Impact Evaluation). This is an example of leveraging additional resources after the impact assessment and implementation project (CBHI India) funded by the European Union.
Socio-economic and societal impact of CBHI
Universal health coverage (UHC) has now become one of the flagship programs for post MDG development in low and middle income countries (LMIC). UHC is understood as a system that provides universal access to quality health care and financial protection in case of ill-health. Many LMICs would like to achieve UHC, but only very few have thus far been able to identify the methods or the means to make headway. Populous countries like India, Indonesia, Pakistan, Nigeria, or Bangladesh, are challenged by the need to adopt an implementation model that would be contributory (as the governments cannot subsidize the full cost of access to healthcare) yet would not be mandatory (as most of the population is in the informal sector where compliance with rules and obligations difficult).
The specific three CBHI schemes developed under this FP7 project have shown that it is possible to roll out benefit packages that will be complementary to any vertical government program that some people might be entitled to. In India, the government program in question is the Rashtriya Swasthya Bima Yojana (government’s hospital insurance scheme, RSBY); RSBY is rolled out in many locations, but not everywhere; and, it covers people with “BPL card”, which is available to only about 25% of the population. Hence, RSBY is not universal (either in terms of the geography of rollout, eligible persons, or in terms of healthcare covered). Yet, in CBHI locations where RSBY was implemented, the members were able to devise benefit packages that complemented RSBY rather than replicating free services. Creating complementarity that benefits the target population is itself an extremely important impact of CBHI, in an environment where there is almost no coordination between several vertical programs that are run by governments in “silos”.
Another important impact of the CBHI relates to a more refined understanding of the theory of change to deliver more and better access to healthcare. The prevailing models of commercial (or public-private) dissemination of health insurance are based on separation of roles of the underwriter from that of the agent/broker. The theory of change suggests that the way to scale coverage depends mainly on a broad network of agents that will be responsible for marketing and sales through one-to-one encounters with individuals. An adoption of this scenario through awareness creation in groups, consensus building among group members on the benefit package and premium, and enrolment inspired not by individual expected utility but by group interests, may be considered. This has considerable practical ramifications: in lieu of building the scaling strategy on agents selling to individuals (and getting a commission based on the premiums written by each agent) the scaling strategy must be built on broad insurance education, interactive awareness creation, consensus building and sufficient flexibility in terms of the package composition and price to allow the consensus to form locally.
The role of women as change-agents within communities has also been (re)demonstrated. The traditional role of women as caregivers gives them a pole position in the discussions with other women on which benefits should be prioritized within the package. And, we have seen that women often needed support in discussing with the menfolk the financial ramifications of joining the CBHI, and when such support was given, it was often possible to enrol both men and women. A related socio-economic impact is that as women are well represented not only as members but also in the committees that govern the CBHI schemes.
CBHI has also had an impact on better governance of pooled funds. The CBHI model that was followed in the FP7 project recognized that governance of the CBHI is in the hands of its members, through community-nominated trusted persons that are trained to perform the necessary skills. This component of the CBHI project made a real contribution to achieve not merely symbolic empowerment, but actually tangible involvement of grassroots groups in the decisions and responsibilities of implementing the scheme fairly and based on rules set by the group. This combination of capacity development and stronger adherence to a rule-based social institution is a very important societal impact of CBHI, which cannot be achieved merely by being enrolled in health insurance.
The CBHI project was limited in time; but the insights and societal impacts can be put to use by government and development agencies well beyond the life of this project. This project or the outcomes alone may not be sufficient to convince of the full potential of CBHI described above, notably because of its limited scope to three locations and to just a few years of actual implementation. At the end of the tenure of the programme, two of the three partner NGOs had ceased activities related to the project. Furthermore the impact on health care utilization and financial protection showed contradictory and rather disappointing results.
The disappointing impact of CBHI on access to care and financial protection, together with the high dropout rates raise concern about the sustainability of schemes such as those implemented in this project. While the localized design and community/NGO involvement have undoubtedly contributed to the initial high enrolment rates, they have also created managerial challenges, especially in terms of setting the appropriate provider incentives. Furthermore, it is questionable to which extent the care provided within the CBHI schemes, which is mostly given by NDAPs, is actually appropriate and of sufficient quality.
In sum, the results of this project have shown that adapting the insurance product to meet local needs and raising awareness of the insurance concept, can indeed lead to relatively high uptake of a voluntary insurance scheme. But at the same this project shows that the uptake of such insurance cover, at least of the type set up in the current experiment which offer limited cover, is unlikely to have a large effect on enhancing access to care and providing financial protection.
MIA has extended the search for robust conclusions on the determinants of enrolment and renewal decisions in CBHI by conducting the first (and only) Systematic Review of the literature on this topic (funded by the 3ie). The knowledge created, both on implementation and on impact evaluation of CBHI is now much more evolved, sophisticated, detailed and quality controlled (MIA training and implementation model is the only one that is compliant with ISO 9001:2008 standards). Those wishing to go forward with implementation and scaling are now much better positioned than we were in 2009, when this FP7 project on CBHI was launched. For that, the entire CBHI practitioners’ community is grateful to the EU-FP7 for having supported this project.
List of Websites:
Public website: http://www.microinsuranceacademy.org/project/uttar-pradesh-and-bihar-india/
Partners:
•Institute for Health Policy & Management, Erasmus University Rotterdam, Burgemeester Oudlaan 50, 3000DR Rotterdam (info@bmg.eur.nl)
•Department for Cooperative Studies, University of Cologne, Albertus-Magnus-Plat, 50923 Cologne, Germany
•Micro Insurance Academy, 52-B, 1st floor, Okhla Industrial Estate, Phase III, New Delhi 110020, India (info@mia.org.in)
•Bharatiya Agro Industries Foundation, BAIF Bhavan, National Highway 4,Pune 411058, India (baif@vsnl.com)
•Nidan, New Patliputra colony, Patna-800 013, (Bihar) India
•Shramik Bharti, Vikas Nagar, Lakhanpur 392, Kanpur 208024, India
The objective of this project was to examine a new model of community based health insurance (CBHI) that is affordable, responsive and inclusive, designed to increase equitable access to health care and to provide financial protection. For that purpose three CBHI schemes were set up in northern rural India. These schemes are characterized by the strong involvement of the target group in the design of the insurance package. The insurance was offered to a population of self-help group members (all women) and their households. These self-help groups already existed and were engaged in micro finance activities. Furthermore a community rated premium applied to all, and the schemes operated without any premium subsidies. Benefit packages varied across sites, reflecting local preferences. As a consequence also the premium differed between locations.
The CBHI schemes were launched in 2010 under the oversight of the Delhi-based Micro Insurance Academy, in partnership with three local NGOs in Kanpur Dehat and Pratapgarh districts in Uttar Pradesh and in Vaishali in Bihar. We used a randomized rollout of the schemes in three separate waves to identify their impact on health care utilization and financial protection. In each of the waves, one-third of the clusters (randomly drawn in each site) were offered the possibility of enrolling in the CBHI scheme. By the end of the project, the entire target population had been offered a chance to become insured.
During the first wave of implementation in 2011, from those offered insurance, 39% of the households had at least one individual who purchased insurance (23% of individuals), while the numbers for the second wave were 45% at the household level (24% of individuals). Dropout rates were quite considerable with 54% of the households (42% of individuals) who enrolled in the first wave renewing in the second, followed by a renewal of 25% of those originally enrolled households (16% of individuals) during the third wave. Enrolment and renewal in CBHI were independent of households’ socio-economic status (social equalizing effect). Furthermore we found no evidence of adverse selection, women were overrepresented among those enrolled and households with greater financial liabilities found insurance more attractive. Finally, making a claim and better knowledge of the insurance concept increased the likelihood of renewal.
For those who actually enrolled in the CBHI schemes the probability of seeking outpatient care from formal sources in Kanpur Dehat increased substantially, while the opposite was true for Pratapgarh where CBHI membership caused a drop in the probability of seeking care. No changes in the use of outpatient care were noted for members in Vaishali or for inpatient care in any of the three sites. In line with healthcare utilization, participation in CBHI had large negative impacts on out-of-pocket payments for members in Pratapgarh, driven by lower utilization, while the opposite pattern was found for Kanpur Dehat, though statistically insignificant. This resulted in a lower probability of hardship financing in Pratapgarh, driven by lower healthcare utilization. In sum, these results suggest limited success of the CBHI schemes on healthcare use and financial protection, with offsetting effects between two sites.
The negative effects in Pratapgarh are surprising and may be related to the capitation payment system applied under CBHI, which may incentivize Non-degree Allopathic Providers (NDAPs) to provide lower amounts of care or drugs as compared to a situation before CBHI when they were paid on a fee-for-service basis. Qualitative field work revealed that a number of insured respondents were dissatisfied with the NDAPs affiliated with the CBHI scheme as they had been providing insufficient care and medications as a result of which many had to pay additional fees or preferred to seek care elsewhere. This is unfortunate because NDAPs are the most accessible medical providers in the community and first contact points for acute illness episodes.
Qualitative work revealed that members perceived value in CBHI on financial aspects, access and quality of healthcare services, and “peace of mind” due to better social cohesion/solidarity. The CBHI schemes were viewed as providing three kinds of access: financial access, physical access (proximity of NDAPs and availability at all hours), cultural access (familiarity, respectful conduct and non-discrimination by providers).
In sum, the results of this project have shown that adapting the insurance product to meet local needs and raising awareness of the insurance concept, can indeed lead to relatively high and equitable uptake of a voluntary insurance scheme. But the results also suggest that in the absence of subsidised premiums, voluntary CBHI schemes operated by local NGOs offer limited benefit packages and face managerial challenges, especially related to the reimbursement of providers, which limits the impact of such schemes on access to care and financial protection. Furthermore, they need to be designed carefully to avoid negative effects on perceived quality of health care supply. Constant monitoring of effects and consequent adaptation of arrangements is necessary.
Project Context and Objectives:
1. Overall Objective and Design of the project
Examine a new model of CBHI that is affordable, responsive and inclusive, designed to increase
(i) Equitable access to healthcare and
(ii) Financial protection.
Specific objectives are:
1.Establish 3 CBHI schemes (each one covering about 20 villages) in 2 states in India.
2.Monitor and evaluate the impact of CBHI on
a.Equitable access to health care; this includes two aspects:
i. Income and age related equity of access
ii. Gender related equity of access.
b.Financial protection from catastrophic cost of illness.
c.The quality and quantity of health care supply.
d.The effect of CBHI on the understanding of the concepts of insurance, solidarity and social action among the insured
3.Field test and document (cost) efficient management of CBHI.
4.Develop and disseminate policy recommendations on implementing and up scaling Community Based Health Insurance.
Three CBHI schemes were set up in Northern rural India. To evaluate the effects of being insured, the enrolment followed step-wise clustered randomized control trials (RCTs) methodology. The CBHI schemes were introduced in 2010 under the oversight of the Delhi-based Micro Insurance Academy, in partnership with three local NGOs in Kanpur Dehat and Pratapgarh districts in Uttar Pradesh and in Vaishali in Bihar. These two states have been characterized as amongst India’s most populated and least educated with large gender disparities.
A new model of CBHI was implemented which was characterized by the strong involvement of the target group in the design of the insurance package and the creation of the ground structure of insurance from among the group of insured (Mutual-aid model). The insurance was offered to a population of self-help group members (all women) and their households. Each household could enroll if at least one female (who was also a member of a self-help group) joined; other household members could join, or not. Furthermore a community rated premium applied to all, and the schemes operated without any premium subsidies.
Benefit packages varied across sites, reflecting local preferences. The benefit-package selection process entailed four steps: narrowing down the options by community representatives, plus three rounds of facilitation of consensus (using the Choosing Healthplans All Together (CHAT) simulation game) conducted among female members of self-help groups. Disagreement on the benefit package decreased consistently over consecutive CHAT rounds, reaching zero (ie, 100% consensus) in two locations, and confirmed gradual adoption of consensus. Evaluation interviews revealed that the wish to be part of a consensus was dominant in all locations. The in-depth interviews indicated that people enjoyed the participatory deliberations, were satisfied with the selection, and that group decisions reflected a consensus rather than majority. The premium also differed between locations.
We used the randomized rollout of the schemes in three separate waves to identify their impact on health care utilization and financial protection, while distinguishing between outpatient care and hospitalizations. In each of the waves, one-third of the clusters (randomly drawn in each site) were offered the possibility of enrolling in the CBHI scheme. By the end of the project, the entire target population had been offered a chance to become insured.
The following published papers contain more details:
Doyle, Conor; Panda, Pradeep; Van de Poel, Ellen; Radermacher, Ralf and Dror, David M.: Reconciling Research and Implementation Needs in Cluster Randomised Controlled Trials: Lessons from 3 Micro Health Insurance Experiments in India. Trials, 12, 224,1-15 http://www.trialsjournal.com/content/12/1/224
Dror, David M.; Panda, Pradeep; May, Christina; Majumdar, Atanu and Ruth Koren (2014). ‘“One for all and all for one”; Consensus-building within communities in rural India on their health microinsurance package’. Risk Management and Healthcare Policy, 7, pp. 139-153, August.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4128598
Project Results:
2. WP 1 Quantitative Research
2.1 Healthcare Seeking Behaviour and the economic consequences of ill-health among Self-help Group Households in Rural Bihar and Uttar Pradesh
In recent years a number of community-based health insurance (CBHI) schemes have been functioning in rural India, supported by non-governmental organizations (NGOs). Such schemes design their benefit packages according to local priorities. We examined the healthcare seeking behaviour of the intended CBHI enrolees (before any enrolment into CBHI) and observed that most of the population does access some form of care. There appeared an overwhelming use of private care for both outpatient and inpatient services. Moreover, non-degree allopathic providers (NDAPs) accounted for a substantial proportion of private care (61 percent of total private use in acute cases and 35 percent in the case of chronic illnesses). Both the self-reported information on health-seeking choices and the econometric estimates confirmed that the main reason for using NDAPs is their proximity and convenience. Results also revealed that higher household consumption, non-SC/ST (scheduled caste and scheduled tribe) status and education were associated with an increase in the probability of using care. Additionally, there is evidence of gender differences in access to care, at least in the case of acute illnesses with men more likely to access care and to seek private care.
The household survey also contains information on various negative events that could affect households’ welfare. We examined the relative importance of health related adverse events (prior to any CBHI schemes operating) and found that these occured very frequently (34% of the households), second only to exposure to natural disasters (51% of the households). Crop and livestock disease and weddings affected about 8% of the households. Only a fourth of households reported to have recovered from illness and/or death in the family (by the time of the survey). Most of the economic risk of ill-health was reflected in healthcare expenditures, and indirect costs were also not negligible. Chronic conditions accounted for close to half of health expenditures. Households tried to cope with health-related expenditures mostly by dissaving, borrowing and selling assets. Few households reported having to reduce (food) consumption in response to ill-health. While most households seemed able to smooth consumption in the short term, coping strategies like selling assets and borrowing from money lenders are likely to have severe long term consequences.
In sum, analysis of our baseline data revealed health shocks to be an important threat to household welfare, and most of the economic risk from ill-health to be related to out of pocket spending. Introducing health insurance may therefore contribute significantly to alleviate economic hardship of families in rural India.
More recent analysis using the 2012-2013 panel data on the economic consequences of ill health indicates that declines in self-assessed health lead to a reduction in prime-ager labour supply and sharply rising OOP health expenditure, irrespective of health insurance enrolment. Despite these negative impacts, health shocks did not decrease non-health consumption in the period shortly after the shock. Instead, findings show that households compensate with a strong increase in debt. Results do not show evidence for short-term tapping of other coping strategies but speculate that heavy reliance on high-interest debt-financing leads to reductions in investment and thus a perpetuation of poverty in the longer run.
2.2 The impact of awareness campaign and preventive health campaign on the insured households
Prior to enrolment and scheme launch, the treatment population went through a structured insurance education program (awareness campaign), which imparted knowledge about and concepts of health insurance and CBHI, including operational aspects of the schemes. It was found that the treatment cohort demonstrated substantial and significantly higher understanding of insurance concepts compared to the control group, and that understanding of CBHI was a positive determinant for enrolment. Of the awareness-enhancing tools, the “Treasure-Pot” (an interactive game) was viewed by respondents as most effective, followed by household discussions.
For financial sustainability of the scheme, as the premiums are low and not subsidized, the scheme should have a sizeable number of members and controlled claims ratio. This can be achieved by increasing the preventive health behaviour among the target population which might reduce the incidence rate of illnesses. A preventive health awareness campaign was conducted in the study locations. Analysis of data before and after the campaign shows a statistically significant increase both in awareness and in preventive practices related to airborne, vector-borne and water-borne diseases, suggesting that the awareness campaign was effective. Affiliation to CBHI had significant positive influence on awareness and on practice scores in the post-campaign period.
2.3 Enrolment in Community Based Health Insurance Schemes in Rural Bihar and Uttar Pradesh
During the first wave of implementation in 2011, from those offered insurance, 39% of the households had at least one individual who purchased insurance (23% of individuals), while the numbers for the second wave were 45% at the household level (24% of individuals). Dropout rates were quite considerable with 54% of the households (42% of individuals) who enrolled in the first wave renewing in the second, followed by a renewal of 25% of those originally enrolled households (16% of individuals) during the third wave.
The household’s socio-economic status did not appear to substantially inhibit uptake. In some cases scheduled caste/scheduled tribe households were more likely to enrol. Also females in the households that joined had higher education than in those that did not join. Second, households with greater financial liabilities found insurance more attractive. Third, access to the national hospital insurance scheme Rashtriya Swasthya Bima Yojana did not dampen CBHI uptake (but had some impact on the type of benefits chosen), suggesting that the potential for greater development of insurance markets and products beyond existing ones would respond to a need. Fourth, recent episodes of illness and self-assessed health status did not influence uptake. Fifth, insurance coverage is prioritized within households, with the household head, the spouse of the household head and both male and female children of the household head, more likely to be insured as compared with other relatives. Also females were overrepresented in households covered. Sixth, offering insurance through women’s SHGs appeared to mitigate concerns about the inclusiveness and sustainability of CBHI schemes.
Regarding renewal we observed that the economic status of the household did not appear to considerably impede renewal. Benefit packages of the schemes appeared to play a decisive role for retention. Receiving money from the scheme through the claim process in the past cycle and CBHI understanding turned out to have a positive and significant influence on renewal. Subscription to Rashtriya Swasthya Bima Yojana did not affect renewal in CBHI.
2.4 The impact of CBHI on access to health care and financial protection
The main objectives of CBHI are the provision of equitable access to healthcare and financial protection from expenses that deplete resources to the insured households. To measure the impact of the CBHI on the aforementioned outcomes, we distinguished between outpatient and inpatient care, where the outpatient care is further disaggregated to acute and chronic conditions. Using three-round panel data, we investigated the CBHI’s effects on healthcare utilization and financial protection at two levels: the impact of its offer (intention to treat effects, ITT) followed by the actual uptake (average treatment effect on the treated, ATT). So ITT investigates the effect on all persons offered CBHI and ATT on those that actually took it.
We began by estimating the effects (both ITT and ATT) of CBHI on the probability of seeking any care conditional on reporting an illness. Subsequently we investigated the effect of enrolment on whether or not care (for the first visit) is sought from a formal provider, defined as Non-degree Allopathic Providers (NDAPs) or qualified doctors (public/private general practitioners and specialists). Subsequently, we looked at the choice of formal provider (NDAPs or qualified doctors) conditional on use to explore whether the CBHI schemes precipitate changes in the choice of healthcare provider. For inpatient care, we first modelled the probability of being hospitalized followed by whether the individual seeks private or public hospital care (conditional on hospitalization).
Regarding financial protection, we looked at the effect of insurance on direct out of pocket healthcare costs (consultation fees, costs of medicine and lab/imaging tests) conditional on reporting an illness and on hospitalization. We defined hardship financing as an individual reporting having to borrow from high interest rate lenders, cutting back on essential costs or selling off assets to service the costs in case of an ill health event.
Using a conditional logistic model, the analysis of ITT effects showed a significant reduction in the probability of seeking formal care in Pratapgarh (driven by a drop in the utilization of GPs or specialists) while the opposite was found to be true for patients in Kanpur Dehat, driven by an increase in the utilizations of NDAPs. Intention to treat effects on healthcare spending and financial protection were limited.
Using the randomized offer of insurance as an instrument for the analysis, the effects of actual enrolment into the CBHI scheme were in line with this. While there was no discernible impact on healthcare utilization at the aggregate level, results revealed significant offsetting results with substantial magnitudes for Kanpur Dehat and Pratapgarh. While CBHI membership led to a surge in the probability of seeking outpatient care from formal sources in Kanpur Dehat, the opposite was true for Pratapgarh where CBHI membership caused a drop in the probability of seeking care. No changes in the use of outpatient care were noted for members in Vaishali or for inpatient care in any of the three sites. In line with healthcare utilization, results showed that participation had large negative impacts on out-of-pocket payments for members in Pratapgarh, driven by lower utilization, while the opposite pattern was found for Kanpur Dehat, though statistically insignificant. This resulted in a lower probability of hardship financing in Pratapgarh, driven by lower healthcare utilization. For inpatient care, the lack of tangible effects of the CBHI scheme is likely to be related to the small sample size given the infrequency of hospitalizations in the target population. Furthermore, coverage for inpatient care could be considered relatively shallow and physical accessibility to the healthcare facilities within the catchment area of this program is relatively low.
In sum, these results suggest limited success of the CBHI schemes on healthcare use and financial protection, with offsetting effects between two sites. The negative effects in Pratapgarh are surprising and may be related to several factors. Most importantly, the capitation payment system could incentivize NDAPs to provide lower amounts of care or drugs as compared to a situation before CBHI when they were paid on a fee-for-service basis. Qualitative field work (see section 4) revealed that a number of insured respondents were dissatisfied with the NDAPs affiliated with the CBHI scheme as they had been providing insufficient care and medications as a result of which many had to pay additional fees or preferred to seek care elsewhere. This suggests that lack of control of the quality of care provided and that the incentives created by the provider payment system might be contributing to the limited impact of CBHI, and low renewal rates. Furthermore, a low degree of competition between providers might exacerbate the incentives to underprovide in a capitation system. Indeed the density of NDAPs was lowest in Pratapgarh (e.g. in wave 2 the ratio of NDAP/village equaled 0.28 in Pratapgarh while it was 0.5 in the other two sites).
The overall results merit further investigation into the heterogeneity of impact of CBHI across gender and age groups. Preliminary results from ongoing work shows a strong decline in healthcare expenses for women whereas no changes were evident for the men. Outcomes such as hardship financing and healthcare utilization follow similar patterns across gender as it does with the whole sample. Heterogeneity across age was investigated by groups: children (0-13yrs), working aged adults (14-55yrs) and the elderly (above 56yrs). Results show that they are largely in line with that of the main results.
2.5 The impact of CBHI on health, wealth and coping behaviour.
Our first impact study shows that while CBHI increases healthcare utilization, impacts on financial protection are ambiguous and context specific. A second impact study uses an identical empirical approach to extend the analysis to impacts on
- Self-assessed health – improved health is a key indicator of health insurance effectiveness that is, however, rarely considered in insurance impact studies because of data constraints. We contribute to this literature by assessing impacts of CBHI on an EQ5D-based index, on individual’s self-ranking on a 5 category health scale, and on the reporting of acute and longer-lasting conditions.
- Non-healthcare consumption – spending on medical care and illness-induced income loss burden household finances. As a result, consumption opportunities reduce, and illness perpetuates poverty. CBHI can contribute to breaking this cycle by a) reducing OOP healthcare expenditure and b) avoiding/reducing illness-related income loss through better health. An extensive non-health consumption survey module permits us to test if CBHI is effective in protecting non-healthcare consumption. We hereby not only estimate impacts on total non-health consumption but also on items of intertemporal importance like food consumption and investments in agriculture and education.
- Debt levels and household assets – in the absence of social insurance, LMIC households revert to various strategies to cope with the negative economic consequences of illness. If increased OOP health expenditure and income loss cannot be compensated through cash income and savings, households with access to informal or formal sources of credit often borrow, or pawn productive assets. While such strategies may enable households to pay for medical costs and maintain consumption levels in the short-run, they likely have negative intertemporal consequences on household welfare. A key benefit of functional health insurance is to avoid harmful coping strategies. We are able to test if CBHI is effective in this regard through a variety of outcomes including current debt levels, land ownership and livestock holdings. Again, given the ambiguity of financial protection effects demonstrated in the first impact study, any positive impacts of CBHI on debt and assets would mainly be driven by improved health.
We are currently conducting data analysis for this project and plan to provide a first draft by the end of October 2014.
The following papers have been published or submitted for publication:
- Panda, Pradeep; Chakraborty, Arpita; Dror, David M. and Bedi, Arjun S. (2013). Enrollment in Community Based Health Insurance Schemes in Rural Bihar and Uttar Pradesh, India. Health Policy and Planning, doi: 10.1093/heapol/czt077.
- Dror, David M. and Firth, Lucy A. (2014). The Demand for (Micro) Health Insurance in the Informal Sector, Geneva Papers on Risk and Insurance, doi: 10.1057/gpp.2014.24
- Raza, Wameq A.; Van de Poel, Ellen; Panda, Pradeep; Dror, David M. and Bedi, Arjun S. (2013). Healthcare Seeking Behaviour among Self-help Group Households in Rural Bihar and Uttar Pradesh, India. Working Paper No. 575, International Institute of Social Studies, Erasmus University, Rotterdam, the Netherlands, December (ISSN 0921-0210).
- Quintussi, Marta; Van de Poel, Ellen; Panda, Pradeep and Rutten, Frans: Economic consequences of ill-health for households in northern rural India, submitted (invited to revise and resubmit at BMC Health Services Research).
- Panda, Pradeep; Chakraborty, Arpita; and Dror, David M. (2014). Building Awareness to Health Insurance among the Target population of Community-Based Health Insurance Schemes in Rural India, submitted
- Raza, Wameq A.; Van de Poel, Ellen.; Bedi, Arjun S. and Rutten, Frans (2014) Impact of community based health insurance on access to care and financial protection in rural India: evidence from three randomized control trials. submitted
- Panda, Pradeep; Chakraborty, Arpita and Dror, David M. (2014). Mobilizing Community-based Health Insurance to Enhance Awareness and Prevention of Airborne, Vector-borne and Waterborne Diseases in Rural India, submitted
Work in progress:
- Panda, Pradeep; Chakraborty, Arpita; Dror, David M.; Raza, Wameq A.and Bedi, Arjun S. (2014). Dynamic Process of Renewal in Community Based Health Insurance in Rural India
- Dror, David M.; Majumda, Atanu; Chakraborty, Arpita; Panda, Pradeep; Koren, Ruth (2014). The Impact of Membership in CBHI in Rural India on Healthcare Seeking and Financial Protection
- Neelsen, Sven; Raza, Wameq A.; Van de Poel, Ellen (2014) Effects of ill-health shocks on consumption and coping strategies.
- Neelsen, Sven; Raza, Wameq A.; Van de Poel, Ellen (2014) Impact of CBHI on health, wealth and coping behaviour
- Raza, Wameq A.; Panda, Pradeep; Van de Poel, Ellen (2014) Determinants of enrolment and the impact of RSBY.
3 WP 2 Provider survey
3.1 Spatial Access to Inpatient Health Care in Northern Rural India
Access to health care in rural areas is a major concern for local populations as well as for policy makers in developing countries. In order to measure spatial access, impedance-based competition using the Three-Step floating Catchment Area (3SFCA) method, a modification of the simple gravity model, was used. 3SFCA was chosen for the study of the districts of Pratapgarh and Kanpur Dehat in the Uttar Pradesh state and Vaishali in the Bihar state, two of India’s poorest states. This approach is based on discrete distance decay and also considers more parameters than other available methods, hence is believed to be a robust methodology.
It was found that Vaishali district has the highest spatial access to in-patient health care followed by Pratapgarh and Kanpur Dehat. There is serious lack of health care, in Pratapgarh and Kanpur Dehat with 40% and 90% of the villages having shortage of in-patient care facilities in these respective districts. We conclude that, firstly, serious spatial access disparities existed between two poorest states of India; secondly, distance to nearest urban centre is negatively associated with spatial access; and lastly equal spatial access to care requires needs-adjusted allocation of health care facilities.
3.2 Spatial Patterning of Rural Medical Practitioners: A Comparison of Three Northern Rural Areas in India
Many rural areas in developing countries lack access to qualified out-patient (OPD) care providers. The gap left by qualified doctors is filled by unqualified persons also known as Non-Degree Allopathic Practitioners (NDAP). In India, NDAPs are dominant providers in rural areas. Despite their dominance, no studies were made to quantitatively estimate their prominence and distribution, especially in our study area.
We used a mixed-method approach, combining exhaustive quantitative census of all health care providers along with Geographic Information System (GIS) to address the knowledge gaps. Longitudinal quantitative and spatial survey was conducted for three consecutive years (2010, 2012 & 2013) for three districts Pratapgarh, Kanpur Dehat in Uttar Pradesh and Vaishali in Bihar, two of India’s poorest states. This study reports that NDAPs locate their practice primarily to meet the health needs of the community that lacks access to formal health care providers.
The three sites studies vary in terms of spatial distribution of NDAPs, and in terms of their demography. After accounting for the demographic heterogeneity, the picture is similar in all three sites: that NDAPs are profit oriented, and more demand produces more NDAPs.
The following papers have been published or submitted for publication:
- Dixit, Shikha and Panda, Pradeep (2013). Spatial research methodology supplementing cluster randomized control trials: learning from a study of community-based health insurance schemes in India. International Journal of Geoinformatics. Vol. 9, No. 3, pp. 31-39.
- Ranga, Vikram and Panda, Pradeep (2014). Spatial Access to In-patient Health Care in Rural Northern India. Geospatial Health, 8(2), pp. 545-556.
Work in progress:
- Ranga, Vikram and Panda, Pradeep (2014). Spatial Patterning of Rural Medical Practitioners: A Comparison of Three Northern Rural Areas in India
4 WP 4 Qualitative survey
4.1 Non-degree allopathic practitioners as first contact points for acute illness episodes: insights from a qualitative study in rural Northern India,
In 2005, the Indian government launched the National Rural Health Mission (NRHM) to improve the quality of and access to rural public health care. Despite these efforts, recent evidence shows that the rural poor continue to consult primarily private non-degree allopathic practitioners (NDAPs) for acute illness episodes. To examine this phenomenon, we explored the perception of rural poor people and their utilization of the rural health care system, notably the role of NDAPs therein.
Our study is based on qualitative data from focus group discussions conducted in 2009/2010 in three rural districts in Bihar and Uttar Pradesh, two high-focus states of the NRHM in Northern India. Our study population consists of female members of self-help groups and their male household heads. We applied a directed content analysis and use a theoretical framework to differentiate between physical, financial and cultural access to care.
Our study population distinguishes between “home treatment” (informal self-care), “local treatment” (formally unqualified care) and “outside treatment” (formally qualified care). Local NDAPs are usually the first contact points for patients before turning to qualified practitioners; they are physically, financially and culturally accessible, because of their proximity, flexible payment options and familiarity with patients’ belief systems, among other things. They treat minor illnesses, provide first relief, refer patients to other providers and administer formally prescribed treatments.,
Our findings are similar for all three study sites and reinforce recent findings from southern and eastern India. The understanding among the poor, and their utilization patterns of the rural health system deviate from governmental ideas. NDAPs are the most accessible medical providers and first contact points for acute illness episodes because they are embedded in the community. Thus, they de-facto fulfill the role envisaged by the Indian government for accredited social health activists introduced as part of the NRHM.
Based on the findings of our study, we concluded that it is necessary for community-based health insurance schemes wishing to offer coverage for outpatient care to integrate these local, private non-degree allopathic practitioners into their insurance portfolio despite their lack of formal qualification, as not doing so would mean to ignore the reality of health care seeking behavior in rural Indian communities. On a more general level, we also concluded that the Indian government should regulate, qualify and integrate local, private non-degree allopathic practitioners as part of the existing public health care system.
4.2 Clients’ Perceived Value in Health Microinsurance – Some Insights from Community-Based Health Insurance Schemes in Rural Northern India,
CBHI is considered as an innovative risk protection tool for the poor. However, despite its potential benefits, demand for CBHI is still low. We tried to gain a deeper understanding of clients’ perceived value in CBHI schemes in rural northern India. We used longitudinal household case studies conducted in all three study sites over a 16-months period after the introduction of the CBHI schemes which included 41 households. This does include pre-purchase and post-purchase phases so that the dynamics of clients’ perceived values can be traced.
We found that many clients value financial protection from catastrophic cost of illness. This is not only of importance for them in financial terms, but also with regard to their state of mind. Financial protection is closely knit to peace of mind for many clients. Yet there is also a big part of clients who value financial protection from costs of minor, everyday illnesses at least as much as protection from catastrophic cost of major but rather rare illnesses. Reasons for this may be informal risk management strategies in case of catastrophic cost of illness, limitations in coverage through the scheme or low probability of illness that causes lower valuing.
We found that clients particularly value the quality and quantity of health care supply when experiencing these (or not), i.e. in the post-purchase phase. When being insured, the quality of health care supply (both by practitioners and as medicines) becomes of high value for the clients and is one of the most important factors for them to stay enrolled or drop out of the scheme.
Based on the findings on which effects are perceived as important by the clients and which are perceived as less important, policy recommendations on how to implement and sustain CBHI schemes can be given. In particular, due to observed dynamics in client value, we recommended assessing client value not only before the start of the scheme (i.e. in the planning phase), but to assess client value also while the scheme is running and possibly adopt scheme’s features to what is valued by the clients. Moreover, we recommended to focus not only on financial and functional dimensions of client value, but to also include social and emotional dimensions.
4.3 Accessing Local Outpatient Care in Community-Based Health Insurance Schemes – Insights From Rural India,
Based on longitudinal household case studies conducted in all three study sites over a 16-months period after the introduction of the CBHI schemes, we examined the access to outpatient care from local, non-degree allopathic practitioners (NDAPs) in three community-based health schemes in rural northern India. We identified three types of outpatient services utilization in the insurance scheme: no utilization, combined/integrated utilization and exclusive utilization.
Insured households appreciate having access to outpatient treatment without having to pay out-of-pocket and to be able to access treatment for outpatient care immediately. On the negative side, they perceive that the quality of care provided by CBHI-associated practitioners is low and not suitable for all illnesses, and that they are not always easily accessible. In some ways, the involvement of local, non-degree allopathic practitioners (NDAPs) in CBHI was a success, as they are situated closely and are familiar to their patients. When patients were not satisfied with their treatment, they were replaced, which was appreciated by insured households. In other ways, however, advantageous features of NDAPs were compromised by organizational arrangements within CBHI: it brings about a reduction of medicine dispensed, only limited access to services available and a restriction to selected practitioners within the insurance.
The insurance schemes – as external entities – have partly regulated and formalized the relationship between NDAPs and their communities, with sometimes negative effects. We argued that the involvement of NDAPs in CBHI is necessary due to their popularity and accessibility for the rural population. From the point of view of insured households, the CBHI schemes under study have been successful to some degree in providing improved financial access to these providers and in preserving their favorable characteristics in physical and cultural accessibility. However, not all positive features could be sustained and it became obvious that schemes need to be organized carefully to not compromise the accessibility of (perceived) quality care by NDAPs.
From the data presented in this paper, we derived that the following interventions would be necessary to improve access to quality outpatient care in the community-based health insurance schemes under study:
- Increase the medicines dispensed by associated providers, e.g. by increasing the payment to NDAPs, reducing the unit prices by coordinating the purchase of medicines in bulk or arranging outside funding.
- Increase the number of NDAPs involved in CBHI to (i) reduce the distance to the next CBHI-associated provider and (ii) provide alternatives if NDAPs are not available at the time of need.
- Thoroughly inform households about which services are covered under CBHI and which not and the reasons for this (such as the rise of premiums for more expensive services or the risk of inappropriate handling of injections).
The consideration of the above points could also help when implementing or up-scaling community-based health insurance schemes. On a more general note, we concluded that CBHI schemes can have positive effects on access to local outpatient care, but need to be designed carefully to avoid negative effects on perceived quality of health care supply. Constant monitoring of effects and adaptation of arrangements if needed is necessary.
The following papers have been published or submitted for publication:
- May, Christina; Roth, Katja and Panda, Pradeep (2014). ‘Non-Degree Allopathic Practitioners as First Contact Points for Acute Illness Episodes: Insights from a Qualitative Study in rural Northern India’. BMC Health Services Research, 14:182. doi:10.1186/1472-6963-14-182.
- Dror, David M.; Panda, Pradeep;, May, Christina; Majumdar, Atanu; and Koren, Ruth (2014). ‘“One for all and all for one”; Consensus-building within communities in rural India on their health microinsurance package’. Risk Management and Healthcare Policy, 7, pp. 139-153.
- Roth, Katja; Panda, Pradeep; May, Christina and Ghosh, Sudeshna (2014). ‘Clients’ Perceived Value in Health Microinsurance: Some Insights from Community-based Micro Health Insurance Schemes in Rural Northern India’, submitted.
Work in progress:
- Ghosh, Sudeshna; Panda, Pradeep; May, Christina and Roth, Katja (2014). ‘The role of health microinsurance in the context of risk management strategies in northern rural India: Insights from a qualitative study’.
- May, Christina; Panda, Pradeep; Roth, Katja and Ghosh, Sudeshna (2014). ‘Accessing Local Outpatient Care in Community-based Micro Health Insurance Schemes: Insights from Rural India’.
Potential Impact:
5 Dissemination and impact
Efforts to disseminate the main findings, insights and policy ramifications have taken several forms besides scientific publications, and have led to several notable results and recognitions.
Invitations to contribute chapters in reference books
• Microinsurance Compendium II 2012
• Encyclopaedia of Health Economics
Dissemination via internet and online knowledge networks
We posted regularly all published materials relating to our project and to the implementation model to several websites, notably
• Social Science Research Network (SSRN),
• Google Scholar
• Research Network
• Academia.edu
• Microinsuraneacademy.org
• Microinsurancenetwork.org
A partial count of browses, downloads of full texts and citations of David Dror’s publications is provided here as an indication of the huge dissemination effect achieved (on 15 August 2014, the count was as follows: 67,284 browses; 13,922 downloads of full text articles; and 1,000 citations by other researchers). The details are provided in the attached PDF.
The project and its findings have been presented at over 40 international events, a detailed list of events is uploaded in the online system.
Awards
• Dr. David M. Dror was bestowed the Insurance personality of the Year 2009 Award by the Asia Insurance industry
• MIA got Educational services provider of the year in 2010 and 2011
• Two months research scholarship award to Dr. Pradeep Panda by ICSSR-NWO in 2012.
• Dr Ellen van de Poel received a VENI-scholarship in The Netherlands in 2011
Exposure visits
•Meetings with other donors that support other CBHI schemes (Nepal, Orissa)
•In April 2011, a high level delegation, comprising researchers and policy makers from Nigeria, visited our project sites.
•In September 2011, three researchers from University of Cologne (Hans Jürgen Rösner, Katja Roth and Christina May) visited the project sites in India.
•As part of the yearly India Week in Cologne, the Department for Cooperative Studies presented the CBHI India project to professors, students and other interested public with a poster and other information material on the University’s India Day on June 08th 2011.
•From September 19th to October 3rd 2011, the Department for Cooperative Studies, University of Cologne, and the Micro Insurance Academy together with the Professional Centre of the University of Cologne organized a Summer School for students on “Micro Health Insurance in Developing Societies – Protection for the Poor?”.
•In February/March 2012, one researcher from Erasmus University Rotterdam (Wameq Raza) visited India and the project sites.
•In February/March 2012, four researchers from University of Cologne (Hans Jürgen Rösner, Katja Roth, Christina May and Marta Quintussi) re-visited India and the project sites.
•In April 2012, a high level delegation, comprising researchers and policy makers from Bangladesh, visited our project sites.
•In February - March 2013, one researcher from Erasmus University Rotterdam (Wameq Raza) re-visited project sites to take part in the training and quantitative field survey with the MIA quantitative researchers.
•In March 2013, an American journalist (Aimee Ginsburg) working for an Israeli newspaper, visited the project site in Vaishali, India and met the implementing partner and a group of insured people and the claim committee members to understand the operational aspects of the CBHI.
•In March 2013, one researcher from University of Cologne (Christina May) re-visited India. She met with some insured and non-insured households under the CBHI schemes for the qualitative data collection taking place during that time with the MIA researchers.
•In December 2013, Christina May (University of Cologne) visited India for her collaborative research work.
•In March 2014, journalist Mr. Benjamin Fernandez, visited the CBHI project site in Vaishali along with Dr. David Dror.
Systematic Review
The Micro Insurance Academy is currently leading a Systematic Review project entitled “What factors affect take-up of voluntary and community-based health insurance schemes in low- and –middle income countries?” Through this project we are synthesizing the experience and evidence gained from CBHI schemes not only in India but also in other countries. This project is funded by the 3ie (International Initiative for Impact Evaluation). This is an example of leveraging additional resources after the impact assessment and implementation project (CBHI India) funded by the European Union.
Socio-economic and societal impact of CBHI
Universal health coverage (UHC) has now become one of the flagship programs for post MDG development in low and middle income countries (LMIC). UHC is understood as a system that provides universal access to quality health care and financial protection in case of ill-health. Many LMICs would like to achieve UHC, but only very few have thus far been able to identify the methods or the means to make headway. Populous countries like India, Indonesia, Pakistan, Nigeria, or Bangladesh, are challenged by the need to adopt an implementation model that would be contributory (as the governments cannot subsidize the full cost of access to healthcare) yet would not be mandatory (as most of the population is in the informal sector where compliance with rules and obligations difficult).
The specific three CBHI schemes developed under this FP7 project have shown that it is possible to roll out benefit packages that will be complementary to any vertical government program that some people might be entitled to. In India, the government program in question is the Rashtriya Swasthya Bima Yojana (government’s hospital insurance scheme, RSBY); RSBY is rolled out in many locations, but not everywhere; and, it covers people with “BPL card”, which is available to only about 25% of the population. Hence, RSBY is not universal (either in terms of the geography of rollout, eligible persons, or in terms of healthcare covered). Yet, in CBHI locations where RSBY was implemented, the members were able to devise benefit packages that complemented RSBY rather than replicating free services. Creating complementarity that benefits the target population is itself an extremely important impact of CBHI, in an environment where there is almost no coordination between several vertical programs that are run by governments in “silos”.
Another important impact of the CBHI relates to a more refined understanding of the theory of change to deliver more and better access to healthcare. The prevailing models of commercial (or public-private) dissemination of health insurance are based on separation of roles of the underwriter from that of the agent/broker. The theory of change suggests that the way to scale coverage depends mainly on a broad network of agents that will be responsible for marketing and sales through one-to-one encounters with individuals. An adoption of this scenario through awareness creation in groups, consensus building among group members on the benefit package and premium, and enrolment inspired not by individual expected utility but by group interests, may be considered. This has considerable practical ramifications: in lieu of building the scaling strategy on agents selling to individuals (and getting a commission based on the premiums written by each agent) the scaling strategy must be built on broad insurance education, interactive awareness creation, consensus building and sufficient flexibility in terms of the package composition and price to allow the consensus to form locally.
The role of women as change-agents within communities has also been (re)demonstrated. The traditional role of women as caregivers gives them a pole position in the discussions with other women on which benefits should be prioritized within the package. And, we have seen that women often needed support in discussing with the menfolk the financial ramifications of joining the CBHI, and when such support was given, it was often possible to enrol both men and women. A related socio-economic impact is that as women are well represented not only as members but also in the committees that govern the CBHI schemes.
CBHI has also had an impact on better governance of pooled funds. The CBHI model that was followed in the FP7 project recognized that governance of the CBHI is in the hands of its members, through community-nominated trusted persons that are trained to perform the necessary skills. This component of the CBHI project made a real contribution to achieve not merely symbolic empowerment, but actually tangible involvement of grassroots groups in the decisions and responsibilities of implementing the scheme fairly and based on rules set by the group. This combination of capacity development and stronger adherence to a rule-based social institution is a very important societal impact of CBHI, which cannot be achieved merely by being enrolled in health insurance.
The CBHI project was limited in time; but the insights and societal impacts can be put to use by government and development agencies well beyond the life of this project. This project or the outcomes alone may not be sufficient to convince of the full potential of CBHI described above, notably because of its limited scope to three locations and to just a few years of actual implementation. At the end of the tenure of the programme, two of the three partner NGOs had ceased activities related to the project. Furthermore the impact on health care utilization and financial protection showed contradictory and rather disappointing results.
The disappointing impact of CBHI on access to care and financial protection, together with the high dropout rates raise concern about the sustainability of schemes such as those implemented in this project. While the localized design and community/NGO involvement have undoubtedly contributed to the initial high enrolment rates, they have also created managerial challenges, especially in terms of setting the appropriate provider incentives. Furthermore, it is questionable to which extent the care provided within the CBHI schemes, which is mostly given by NDAPs, is actually appropriate and of sufficient quality.
In sum, the results of this project have shown that adapting the insurance product to meet local needs and raising awareness of the insurance concept, can indeed lead to relatively high uptake of a voluntary insurance scheme. But at the same this project shows that the uptake of such insurance cover, at least of the type set up in the current experiment which offer limited cover, is unlikely to have a large effect on enhancing access to care and providing financial protection.
MIA has extended the search for robust conclusions on the determinants of enrolment and renewal decisions in CBHI by conducting the first (and only) Systematic Review of the literature on this topic (funded by the 3ie). The knowledge created, both on implementation and on impact evaluation of CBHI is now much more evolved, sophisticated, detailed and quality controlled (MIA training and implementation model is the only one that is compliant with ISO 9001:2008 standards). Those wishing to go forward with implementation and scaling are now much better positioned than we were in 2009, when this FP7 project on CBHI was launched. For that, the entire CBHI practitioners’ community is grateful to the EU-FP7 for having supported this project.
List of Websites:
Public website: http://www.microinsuranceacademy.org/project/uttar-pradesh-and-bihar-india/
Partners:
•Institute for Health Policy & Management, Erasmus University Rotterdam, Burgemeester Oudlaan 50, 3000DR Rotterdam (info@bmg.eur.nl)
•Department for Cooperative Studies, University of Cologne, Albertus-Magnus-Plat, 50923 Cologne, Germany
•Micro Insurance Academy, 52-B, 1st floor, Okhla Industrial Estate, Phase III, New Delhi 110020, India (info@mia.org.in)
•Bharatiya Agro Industries Foundation, BAIF Bhavan, National Highway 4,Pune 411058, India (baif@vsnl.com)
•Nidan, New Patliputra colony, Patna-800 013, (Bihar) India
•Shramik Bharti, Vikas Nagar, Lakhanpur 392, Kanpur 208024, India
