This article has Open Peer Review reports available.
Assessing effectiveness of a community based health insurance in rural Burkina Faso
© Hounton et al.; licensee BioMed Central Ltd. 2012
Received: 28 December 2011
Accepted: 5 October 2012
Published: 19 October 2012
Financial barriers are a recognized major bottleneck of access and use of health services. The aim of this study was to assess effectiveness of a community based health insurance (CBHI) scheme on utilization of health services as well as on mortality and morbidity.
Data were collected from April to December 2007 from the Nouna’s Demographic Surveillance System on overall mortality, utilization of health services, household characteristics, distance to health facilities, membership in the Nouna CBHI. We analyzed differentials in overall mortality and selected maternal health process measures between members and non-members of the insurance scheme.
After adjusting for covariates there was no significant difference in overall mortality between households who could not have been members (because their area was yet to be covered by the stepped-wedged scheme), non-members but whose households could have been members (areas covered but not enrolled), and members of the insurance scheme. The risk of overall mortality increased significantly with distance to health facility (35% more outside Nouna town) and with education level (37% lower when at least primary school education achieved in households).
There was no statistically significant difference in overall mortality between members and non-members. The enrolment rates remain low, with selection bias. It is important that community based health insurances, exemptions fees policy and national health insurances be evaluated on prevention of deaths and severe morbidities instead of on drop-out rates, selection bias, adverse selection and catastrophic payments for health care only. Effective social protection will require national health insurance.
The surveys and data extraction were conducted from April to December May 2007.
Study site and populations
The Nouna community based health insurance (CBHI) scheme
Data collection and analysis
Data were extracted from longitudinal household surveys at the Nouna Demographic Surveillance Site. The household survey sample size was 1,504 households at the time of the study, half of which were from Nouna town and the other half from surrounding villages; data were extracted on household characteristics, births, deaths and age at death, distance from village to health centers and to Nouna district hospital. In addition, a prospective survey was administered to women with experience of delivery during the last 12 months prior to the survey (April – May 2007) and data were collected on place of delivery, membership in the Nouna Community Based Health Insurance, age of the mother, anemia , average distance from village to health centre, and assets ownership. Anemia was selected as this morbidity has not been investigated at the time the study was designed. Descriptive statistics and regression analyses (logistic regression and Poisson regression) were performed to assess the association of overall mortality, utilization of health services, and institutional delivery by membership to the Nouna CBHI adjusting for important covariates. We used a Poisson regression with computed person-time of all deaths (person-time of years spent before death in the DSS, since the DSS started) to investigate whether there is any mortality advantage (lower mortality risk) to any sub group and by any of the selected covariates, and assess whether there is an overall lower mortality risk among members versus non-members and by selected covariates. This method was used because the distribution of deaths is probably skewed and the person-time variable consists of non-negative integers. The dependent variable was the person-time of all deaths in the Nouna DSS and membership in the Nouna CBHI, educational level, age, asset ownership, place the explanatory variables. Dummy variables were created reflect the time spent in the scheme.
Study was approved by ethical review boards of Centre MURAZ and the Nouna Health Research Centre (Burkina Faso).
Descriptive analysis of the study populations
Descriptive characteristics of populations by enrolment status (from household survey, 1504 household)
Use of health services
- Did not use
- Have used
- At least primary school level
- Nouna town
- Nouna villages
- Most poor
- Second quartile
- Third quintile
- Fourth quintile
- Least poor
Effects on utilization of health services
Utilization of health services by enrolment status in Nouna CBHI scheme and by household characteristics (Household survey, 1504 household)
Did not use services
Education : None
- Non members
Education : At least primary level
- Non members
Place: Nouna town
- Non members
Place : Nouna villages
- Non members
Perceived quality of care: Low
- Non members
Perceived quality of care: High
- Non members
Asset ownership : Most poor
- Non members
Asset ownership : Least poor
- Non members
Logistic regression of utilization of health services by enrolment status and by covariates
95% Conf. Interval
Nouna town (1)
Effect of Nouna CBHI on mortality
Poisson regression of mortality status (person-time of all deaths) by enrolment status and by covariates (distance, place of residence, household size, educational level), Nouna DSS, Burkina Faso
95% Conf. Interval
Similarly, the risk of overall mortality among households whose head of household had at least primary school level education was 37 % lower than the risk of overall mortality among the reference group (no education level) and the risk of overall mortality among households whose head of household had at least secondary school level education was 88% lower than the risk of overall mortality among the reference group. Also, the risk of overall mortality within households covered by the scheme and members in the scheme was 88% lower than the risk of overall mortality among the reference group of households not members (because their areas were yet to be covered).
Effects on anemia among women with recent deliveries
Description of levels of haemoglobin by antenatal care visits, and institutional delivery Nouna household cost survey
At least one antenatal care visit
No antenatal care visit
(Hg ≤ 7g/dl)
p value = 0.000
Hg › 7g/dl
(Hg ≤ 7g/dl)
P value = 0.008
Hg › 7g/dl
It was not clear to us whether people used services more because they were enrolled or they were enrolled because they tended to use services more. Our results point to a selection bias in enrolment in the Nouna CBHI. This was confirmed by the mortality analysis. Although one may argue that CBHI alone may not reduce mortality rates, we can reasonably assume that, everything being equal, reduction of financial barriers could significantly reduce the delays in accessing emergency care and therefore reduce the likelihood of mortality. Our study found no “dose–response” between the number of years of exposure to the scheme (enrolment in the scheme) and the risk of overall mortality, and no significant association between risk of overall mortality and areas covered by the scheme. Groups of individuals living close to Nouna town and in less crowded households, and individuals whose heads of household were better educated experienced a lower mortality risk compared to their counterparts living further away from the district hospital (more remote villages), staying in more crowded households, and whose heads of household had no education respectively.
The observed differences in overall mortality may be due to selection bias (ie people members in the scheme tend to be better off compared to non-members in terms of educational background, asset ownership, birth cohorts, household size, distance to health services, etc.), and a possible selection bias in the step-wedge design. In fact, one may hypothesize that people living in areas covered initially may be better off compared to people living in areas covered by subsequent waves of implementation. It is not unusual when launching an intervention to begin where there is a better likelihood of success before rolling the intervention out; but then one should be cautious about any observed outcome as the results may be confounded by who the recipients were and the context of the intervention instead of the intervention itself .
Regarding the data about anemia, one would expect that a greater utilization of health services may translate into a greater likelihood of early detection of anaemia among pregnant women during antenatal care visits and newly delivered women. The early detections will thus prompt care which will subsequently result in the lower prevalence of anaemia cases. These results point to fewer cases of severe anaemia among women who delivered in health facilities and this may have various interpretations. It may be that the anaemia condition triggered the visits to health facilities, hence the high prevalence (absolute numbers) of severe anaemia in the antenatal care visit category. It may well also be that compared to women who delivered at health facilities, an anaemia condition among those who delivered at home is less likely to be diagnosed and to be treated, and hence the observed lower levels of haemoglobin in the post-partum periods in which the survey took place. Another possible explanation is that low quality of care given by traditional birth attendants during delivery that could result in less control of haemorrhage during or after delivery and hence lower levels of haemoglobin in the post-partum periods for women who did not deliver in facilities. Regardless of the actual causal mechanisms, these findings confirm our assumption that a greater utilization of health services might translate into a greater likelihood of early detection of anaemia among pregnant and newly delivered women, and therefore better care and lower levels of anaemia. We could not find any difference in levels of haemoglobin by membership in the community based health insurance scheme. Finally, we could not find an association between delivery within an institution, antenatal care and membership of the community based health insurance scheme. This finding is puzzling because these associations are reasonable expectations using our conceptual framework and given that the Community Based Health Insurance scheme aimed to increase utilization of health services. However, the sample size (251 eligible women with delivery in the last 12 months before the survey), may well be too small to detect any effect, or maybe there are other factors that confound the relationship between enrolment in the Community Based Health Insurance scheme and utilization of health services.
The inferences made were based on data collected over 4 years ago, and the evolution of health outcomes may have changed. Also financial barrier is only one bottleneck in access and uptake of quality care, and the limited recall period since the launch of the Nouna CBHI (2–3 years) may not be sufficient to observe a change in outcomes. Nonetheless, change in health outcomes should the standard practice in evaluating any health intervention.
There was an independent and significant association between utilization of health services and membership of the scheme even after adjusting for education level, place of residence, and asset ownership, and even with the evidence of selection bias at both the design stage and in the membership. There was no observed association between an institutional delivery, antenatal care, levels of haemoglobin of women and membership of the community based health insurance scheme. Regarding mortality, even without actually being part of the Nouna CBHI scheme, households that could have been members (being located in areas covered by the scheme) experienced a 50% lower mortality rate compared to households not members because their areas were yet to be covered by the scheme, which point to a selection bias. We observed no significant association between memberships of the Nouna community based health insurance scheme and overall mortality within households. Given the usual low enrolment rates and the selection bias in most community based health insurance scheme, it is important for future assessments of community based health insurance schemes, exemptions fees schemes or national health insurance to focus on actual prevention of mortality and morbidities [23, 24] and not only on adverse selection, drop-out rates and catastrophic payments for health care or utilization of services.
Sennen Hounton is a medical epidemiologist with expertise in maternal and newborn health, health systems and economic evaluation. Sennen Hounton is an MD (Benin), MPH in Epidemiology (University of Oklahoma, USA) and a PhD in Public Health from University of Aberdeen, (Scotland, UK). He was a Senior Research Fellow with Immpact (Initiative for Maternal Mortality Program Assessment). He is currently Maternal Health Technical Adviser at the United Nations Population Fund (New–York), and serves as Scientific and Technical Advisor on the WHO Alliance for Health Policy and System Research Scientific and Technical Advisory Committee.
Peter Byass is Professor of Global Health, and Director of the Umeå Centre for Global Health Research (UCGHR) in Sweden. UCGHR is a leading research centre, relating human health to epidemiological transition, life-course interventions, primary care, gender and climate. Professor Peter Byass works with his team all around the world on research that tries to improve peoples’ health and lives.
Dr Bocar Kouyate is public health and health system research expert and Special Advisor to the Ministry of Health, Burkina Faso. He was the founding Director of the Nouna Health Research Center and the Director of the National Malaria Research Center in Burkina Faso. Dr Bocar Kouyate is a member of numerous scientific and advisory technical committees in health system research.
This work was supported by the international research programme – Immpact (University of Aberdeen) and United Nations Population Fund country office in Burkina Faso. The funders have no responsibility for the information provided or views expressed in this paper.
- Ouimet MJ, Fournier P, Diop I, Haddad S: Solidarity or financial sustainability: an analysis of the values of community-based health insurance subscribers and promoters in Senegal. Can J Public Health. 2007, 98 (4): 341-346.PubMedGoogle Scholar
- Jacobs B, Bigdeli M, van Pelt M, Ir P, Salze C, Criel B: Bridging community-based health insurance and social protection for health care–a step in the direction of universal coverage?. Trop Med Int Health. 2008, 13 (2): 140-143. 10.1111/j.1365-3156.2007.01983.x.View ArticlePubMedGoogle Scholar
- De Allegri M, Sauerborn R: Community based health insurance in developing countries. BMJ. 2007, 334 (7607): 1282-1283. 10.1136/bmj.39240.632963.80.View ArticlePubMedGoogle Scholar
- Drummond MF, Sculpher MJ, Torrance GW, Brien BJ O, Stoddart GL: Methods for the economic evaluation of health care programmes. 2005, Oxford University Press, New York, 3Google Scholar
- Dong H, De Allegri M, Gnawali D, Souares A, Sauerborn R: Drop-out analysis of community-based health insurance membership at Nouna, Burkina Faso. Health Policy. 2009, 92 (2–3): 174-179.View ArticlePubMedGoogle Scholar
- Dong H, Gbangou A, De Allegri M, Pokhrel S, Sauerborn R: The differences in characteristics between health-care users and non-users: implication for introducing community-based health insurance in Burkina Faso. Eur J Health Econ. 2008, 9 (1): 41-50. 10.1007/s10198-006-0031-4.View ArticlePubMedGoogle Scholar
- De Allegri M, Kouyaté B, Becher H, Gbangou A, Pokhrel S, Sanon M, Sauerborn R: Understanding enrolment in community health insurance in sub-Saharan Africa: a population-based case-control study in rural Burkina Faso. Bull World Health Organ. 2006, 84 (11): 852-858.PubMedPubMed CentralGoogle Scholar
- Ndiaye P, Kaba S, Kourouma M, et al: MURIGA in Guinea: an experience of community health insurance focused on obstetric risks. in Reducing financial barriers to obstetric care in low-income countries, Studies in Health Services Organization & Policy. 2008, 117-148. N° 24Google Scholar
- Carrin G, Waelkens MP, Criel B: Community-based health insurance in developing countries: a study of its contribution to the performance of health financing systems. Trop Med Int Health. 2005, 10 (8): 799-811. 10.1111/j.1365-3156.2005.01455.x.View ArticlePubMedGoogle Scholar
- Ekman B: Community-based health insurance in low-income countries: a systematic review of the evidence. Health Policy Plan. 2004, 19 (5): 249-270. 10.1093/heapol/czh031.View ArticlePubMedGoogle Scholar
- Morris SS, Ranson MK, Sinha T, Mills AJ: Measuring improved targeting of health interventions to the poor in the context of a community-randomised trial in rural India. Contemp Clin Trials. 2007, 28 (4): 382-390. 10.1016/j.cct.2006.10.008.View ArticlePubMedGoogle Scholar
- Ranson MK, Sinha T, Chatterjee M, Gandhi F, Jayswal R, Patel F, Morris SS, Mills AJ: Equitable utilisation of Indian community based health insurance schemes among its rural membership: cluster randomised controlled trial. BMJ. 2007, 23: 334(7607).Google Scholar
- Ensor T, Cooper S: Overcoming barriers to health service access: influencing the demand side. Health Policy Plan. 2004, 19: 69-79. 10.1093/heapol/czh009.View ArticlePubMedGoogle Scholar
- Criel B, Waelkens MP: Declining subscriptions to the Maliando Mutual Health Organisation in Guinea-Conakry (West Africa): what is going wrong?. Soc Sci Med. 2003, 57: 1205-1219. 10.1016/S0277-9536(02)00495-1.View ArticlePubMedGoogle Scholar
- Jütting J: Do community-based health insurance schemes improve poor people's access to health care? Evidence from rural Senegal. World Dev. 2004, 2: 273-288.View ArticleGoogle Scholar
- Bennett S: The role of community-based health insurance within the health care financing system: a framework for analysis. Health Policy Plan. 2004, 19 (3): 147-158. 10.1093/heapol/czh018.View ArticlePubMedGoogle Scholar
- Su TT, Kouyate B, Flessa S: Catastrophic household expenditure for health care in low-income society: a study from Nouna district, Burkina Faso. Bull World Health Organ. 2006, 84 (1): 21-27.View ArticlePubMedPubMed CentralGoogle Scholar
- Dong H, Kouyate B, Cairns J, Sauerborn R: Willingness-to-pay for community-based insurance in Burkina Faso. Health Econ. 2003, 12 (10): 849-862. 10.1002/hec.771.View ArticlePubMedGoogle Scholar
- Dong H, Kouyate B, Cairns J, Sauerborn R: Differential willingness of household heads to pay community-based health insurance premia for themselves and other household members. Health Policy Plan. 2004, 19 (2): 120-126. 10.1093/heapol/czh014.View ArticlePubMedGoogle Scholar
- Dong H, Mugisha F, Gbangou A, Kouyate B, Sauerborn R: The feasibility of community-based health insurance in Burkina Faso. Health Policy. 2004, 69 (1): 45-53. 10.1016/j.healthpol.2003.12.001.View ArticlePubMedGoogle Scholar
- Mendrone A, Sabino EC, Sampaio L, Neto CA, Schreiber GB, Chamone Dde A, Dorlhiac-Llacer PE: Anemia screening in potential female blood donors: comparison of two different quantitative methods. Transfusion. 2009, 49 (4): 662-668. 10.1111/j.1537-2995.2008.02023.x.View ArticlePubMedGoogle Scholar
- De Allegri M, Sauerborn R, Kouyaté B, Flessa S: Community health insurance in sub-Saharan Africa: what operational difficulties hamper its successful development?. Trop Med Int Health. 2009, 14 (5): 586-596. 10.1111/j.1365-3156.2009.02262.x.View ArticlePubMedGoogle Scholar
- Eriksen J, Mujinja P, Warsame M, Nsimba S, Kouyaté B, Gustafsson LL, Jahn A, Müller O, Sauerborn R, Tomson G: Effectiveness of a community intervention on malaria in rural Tanzania - a randomised controlled trial. Afr Health Sci. 2010, 10 (4): 332-340.PubMedPubMed CentralGoogle Scholar
- Kouyaté B, Somé F, Jahn A, Coulibaly B, Eriksen J, Sauerborn R, Gustafsson L, Tomson G, Becher H, Mueller O: Process and effects of a community intervention on malaria in rural Burkina Faso: randomized controlled trial. Malar J. 2008, 7 (25): 50.View ArticlePubMedPubMed CentralGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1472-6963/12/363/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.