D+C Development and Cooperation (No. 6, November/December 2001, p. 4 - 5)

For most people living in developing countries "health insurance" is an unknown word. It is generally assumed that, with the exception of the upper classes, people cannot afford such type of social protection. Studies by the Bonn-based Center for Development Research (ZEF) in rural areas of Africa and Asia provide a more differentiated picture. An example from Senegal shows how even the poor can protect themselves against illness.

Health insurance schemes are an increasingly recognized tool to finance health care provision in low income countries. Given the high latent demand from people for health care services of a good quality and the extreme under-utilisation of health services in several countries, it has been argued that social health insurance may improve the access to health care of acceptable quality. Whereas alternative forms of health care financing and cost-recovery strategies like user fees have been heavily criticised, the option of insurance seems to be a promising alternative as it is a possibility to pool risk transferring, unforeseeable health care costs to fixed premiums. Recently, mainly in Sub-Saharan Africa but also in a variety of other countries, non-profit, mutual, community-based health insurance schemes have emerged. These schemes are characterised by an ethic of mutual aid, solidarity and the collective pooling of health risks. In several countries these schemes operate in conjunction with health care providers, mainly hospitals in the area.

Against this background the Center for Development Research (ZEF) analyses within his research programme on social security systems in rural areas the prospects and limitations of innovative health insurance schemes. In close collaboration with national research institutes empirical studies are currently being carried out in Ethiopia, China, Ghana, India, Senegal and Tanzania. The aim of these projects is to estimate demand for health care and health insurance, quantify economic and social impacts, as well as identifying factors of success and failure. The studies focus on rural areas because here the need for insurance is especially urgent, but private insurance markets do not exist and public measures often fail to reach their target population.

Recent study in Senegal finds
positive effects

That even the rural poor can insure themselves against the risk of illness was shown by a recent study, which was carried out in Senegal. In the region of Thiès in rural Senegal community based health insurance schemes, based on traditional forms of mutual assistance, have been developed in the last ten years. These "mutuelles de santés" (mutual health insurance schemes) have around 500 members on average and are deeply rooted in the local setting. The idea of the mutuelles is simple: low cost expenditures are covered by the family, but in case of hospitalisation the mutuals cover the costs. To do so, the community-based health insurance schemes have established a contract with the church run hospital St. Jean de Dieu in Thiès. This grants members of mutuelles substantially reduced fees for specific treatment and services.

Household survey

In May 2000 ZEF and the Institute for Health and Development in Dakar conducted a household survey to estimate the impact of the mutuelles on health care demand, costs and health outcomes. Roughly 350 households in four selected communities were interviewed, of which 60 per cent were members and 40 per cent non-members of a mutuelles. In light of the limited services provided - only hospitalisation, no primary health care - and the low financial capacities of the mutuelles, their impact is surprisingly impressive. Members of a mutual have better access to health care than non-members and benefit in other ways, too. Members can join the hospital more often than non-members in case of a serious illness and they pay only half the amount which non-members have to pay. Furthermore, during interviews members stressed that joining a mutual has led to a reduction in their worries. In case of serious illness they are no longer dependent on assistance from their social network, or money from the local money-lender. Another important result of the study is that the poorest of the poor do not participate in the mutuals. They lack the financial resources to pay regular membership contributions.

Determinants of viable health
insurance

Similar studies on the effect of community risk sharing in health care in Rwanda, India and Thailand have been undertaken by different research institutes. These studies as well as the Senegal case have been selected as part of the work of the Commission on Macroeconomics and Health of the World Health Organisation. This commission examines the interrelations among investment in health, economic growth, and poverty reduction. All the studies are based on household surveys on the effect of community financing schemes (CF schemes) and have used the same methodology for data analysis. The following points summarise important findings of the different studies with respect to the design of the schemes:

1. Flexibility in paying procedure

In Rwanda the households who could not afford to pay the premium in one bit, were allowed to pay in instalments to a tontine (small savings and loan association) before joining a pre-payment scheme. In addition, church based groups collected fees for the indigent, disabled, orphans etc. The paying of contributions by charitable organisations has also been reported in the Senegal study, which has given otherwise excluded people the chance to participate in the mutuals. Some of the mutuals even start collective activities from which they use some of the earnings to pay membership fees. Another example for a possible source of financing is organising a tombola or lottery. In conclusion there are various possibilities to adapt paying procedures to the local level requirements. In this context, the role of the state also needs to be explored e.g. the possibilities for demand targeted subsidies.

2. Experience in social protection
and community participation

Community financing schemes (CF schemes) are often set up by voluntary, non-profit-oriented organisations. These organisations act as an insurance broker between the interest of a health care provider and the expectations and needs of their members. To deal with these ambiguities is of major importance and requires trained personal. In this context it must be stressed that the administrative procedure for handling claims should be as simple and transparent as possible. The SEWA example (in India) but also other experiences (e.g. Grameen Bank in Bangladesh) shows that mutual insurance schemes are likely to perform better when they are linked to an organisation which already has experience in the field of financial services and social protection.

Community participation matters, when it comes to the control of moral hazard behaviour and costs. The results of the studies in the four countries suggest, that the degree of community participation in the design and running of the CF schemes can vary widely and is usually greater if funds are owned and managed by the members themselves than if schemes are run by health facilities. If members can identify themselves with "their" schemes because they control the funds and have decision-making power, they will tend less to unnecessary use of health care services.

3. Existence of a viable health
care provider

The success or failure of health insurance schemes is largely dependent on the existence of viable health care providers, e.g. a hospital that offers services to the insured. Decisions taken by the health care provider have an impact on mobilizing demand for CF schemes as well as on the financial balance of the scheme. The Senegalese case study was enlightening in that respect: From the beginning of the mutual health organisation movement, it has been supported by the hospital St. Jean de Dieu. The administration of the hospital had recognized that their ultimate target group - the poor - couldn’t pay their fees, but it was also not possible for the hospital to allow for a general exception of fees for the poor. The creation of mutual health organisations allowed to directly target their clientele in a cost effective manner. Beside the financial support which the hospital gives to the mutuals an equally important point is the well recognized quality of care.

The delivery of services with high quality is a very important point for mobilizing demand in the medium to long term. In some settings it will even not be possible to set up a viable insurance scheme and mobilize demand before quality of care is not improved, because if people feel that they will get no "value for money" at the hospitals or health posts, they would be unwilling to pay premiums.

4. Community and household
characteristics

The demand for health insurance is a crucial factor if the benefits expected from community financing schemes are to be realized. The demand of households for health insurance depends not only on the quality of care offered by the health care provider, on the premium and benefit package, but also on socio-economic and cultural characteristics of households and communities. Widespread absolute poverty among potential members can be a serious obstacle to the implementation of insurance. This argument was frequently put forward from non-members in Senegal. If people are struggling for survival every day, they are less willing to pay insurance premiums in advance in order to use services at a later point in time. Social exclusion may persist even if barriers to access are reduced for part of the population, and exemption mechanisms for the poorest or sliding scales for premiums that might be a remedy are not easy to implement. After or before the introduction of health insurance, rising incomes, that may be brought about by development projects, can be necessary to attract members and realize the potential benefits of CF. SEWA’s activities in this direction are a good example.

The prevailing concepts of illness and risk are relevant to the decision of households whether to purchase health insurance or not. If people see illness as a somewhat random event that can hit anyone, they are surely more willing to purchase insurance than if they perceive it as punishment for misbehaviour by magic powers. Cultural habits in dealing with the risk of illness can influence the demand for insurance. In Senegal this has been frequently reported as one obstacle to buy health insurance as people were used to put money aside for unpredictable events like marriages and funerals, but they believed that saving money for eventual health care costs meant "wishing oneself the disease". If solidarity is strong, people will not worry so much if the benefits of the premiums they paid will accrue to themselves or other community members. For example, members of the Fandene scheme in Senegal expressed the opinion that if they would not need health care themselves, at least they had done something good for the community by contributing to the insurance fund. The degree of solidarity and mutual trust is probably higher in homogeneous, close-knit communities than in scattered and diverse populations comprising people of different ethnic origin, religion and culture.

Conclusions

For all to long, social security and risk management has been regarded from the supply side neglecting to a large extent what institutions, innovations and the variety of different actors can contribute in providing and financing services at the local level. Recently, the interest has shifted to the demand side and institutions like CF schemes gain importance, which help otherwise excluded people to better deal with day to day risks. Preliminary research suggests that health insurance for the poor is feasible under certain conditions. More important, it can have a positive effect on the economic and social situation of their members. Besides the spread of the financial risk of illness and better access to health care, indirect effects of health insurance can result such as better health outcomes and an increase in labor productivity. To what extent health insurance, or rather lack thereof, affects people’s willingness to undertake risky, but potentially profitable investments needs to be further investigated.

A successful inclusion of the poorest of the poor is, however, not an automatic outcome of community structures. Community involvement can be exclusionary as well as inclusionary. This suggests that certain community characteristics as well as scheme design and implementation features are important determinants to achieve a pro-poor targeting outcome. These determinants and the direct causality is not well explored with regards to health financing, and further investigation is warranted. In particular, the role of external financial support (such as government subsidies, donor funding, re-insurance needs) in encouraging social inclusion needs to be further explored. In this context, one should be aware that CF schemes cannot be the sole solution to manage health risks but represent an important element. Further research is needed, how these schemes can be scaled up and replicated as well as how to link them to social risk management and instruments, e.g. social funds to broaden the risk pool and increasing coverage rates.