The case of Mali
Improving financial access to healthcare through mutual health organisations?
Von Alexander Schulze & Samuel Frei / Novartis Foundation
Since the 1990s, mutual health organisations (MHOs) have been promoted in much of West and Central Africa to improve access to healthcare for disadvantaged people and to protect them from the financial risk of ill-health. A case study conducted in Mali provides additional evidence on the impact of MHOs.
Commonly known as mutuelles de santé in Francophone Africa, mutual health organisations (MHOs) as health financing mechanism mainly target rural populations and workers in the informal private sector who usually, when in need of healthcare, have no financial protection. A number of authors have presented evidence on the impoverishing effect of accessing healthcare when patients have to pay for their diagnosis and treatment entirely out-of-pocket (OOP). (Su, Kouyaté & Flessa 2006 / Kawabata, Xu & Carrin 2002). Yet, MHOs or other models are not only considered to facilitate access to healthcare and protect patients financially, they are also seen to help mobilise resources for financing healthcare provision. Since empirical evidence on the impact of MHOs on their members and healthcare provision is weak, a case study was conducted in Mali to provide additional evidence. In this article, study results are presented and recommendations for increasing the effectiveness of MHOs formulated.
Features of Malian MHOs and their position in the national health financing system
In Mali, MHOs are private, not-for-profit associations. They are own legal entities and usually separated from the healthcare provider although sometimes the latter initiates such schemes in order to mobilise resources for health services. Membership of individuals, households or other groups is voluntary and most schemes are run by unpaid management. Since the MHOs are associations, household heads are members and therefore not only insurance beneficiaries, but also insurance providers. For example, in the General Assembly they can vote for a new benefit package. In contrast to many other countries, Mali has developed a regulatory framework through one law, two decrees and two promulgations which guide the creation and management of such MHOs. Furthermore, the Malian State and the French Development Agency have built up a national technical support structure, the Union Technique de la Mutualité Malienne (UTM). On the one hand, as the umbrella organisation, the UTM develops strategies for the promotion of the MHO movement and defends its interest at a national level; on the other hand, it provides technical support for newly set-up schemes and guides them.
MHOs are only one model in the Malian health financing system. Despite the introduction of user fees in the beginning of the 1980s, a number of health services remain free of charge for specific segments of the population. For instance, malaria treatment and vaccinations are free for children under five and pregnant women. In case of birth complications, caesarean sections are also free of charge although, transport costs and necessary medical supplies must be borne by the patient. These free services are financed by the State, which is further supported by foreign aid through mechanisms such as the Global Fund for Malaria, TB and HIV/AIDS. MHOs also benefit from foreign aid. Donors either provide the UTM with funding in order to help set up new schemes and monitor their performance, or they directly support the set-up of new schemes, for instance in the frame of health projects. (Initiative Accès, Novartis Foundation 2008) In 2011, the Malian Government passed a new policy on community-based health insurance schemes. With the support of the World Bank, the Bill and Melinda Gates Foundation and the United States Agency for International Development (USAID), a policy was devised to ensure the creation of one mutuelle de santé per municipality or district and furthermore, to create umbrella organisations at district or regional levels. Moreover, 50% of all premiums are supposed to be subsidised by the Government (Healthsystems 20/20 2011).
Reforms are also underway with regard to health financing mechanisms for civil servants and employees in the formal, private sector. In 2009, the Government passed a law on the introduction of a compulsory health insurance (Assurance Maladie Obligatoire) which, once effectuated, will cover all civil servants, employees in the private, formal sector as well as members of parliament. Though not yet implemented, the compulsory health insurance puts the existing voluntary health insurance (Assurance Maladie Volontaire) at stake since some of their target groups – e.g. formal private sector employees – overlap. Voluntary health insurance can only be purchased by urban populations and is centrally managed by the UTM. It is distributed through a network of urban MHOs. One example is the Mutuelle Interprofessionnelle de Ségou which is one of the two case studies presented further below. In the nearer future, the voluntary health insurance scheme will however remain the only health financing mechanism open to urban, informal private sector workers. Other additional forms of health financing for employees in the formal, private sector are the Régimes d’Assurance Maladie et d’Entreprise. While the first are products provided by private, for profit insurance schemes, the second are corporate schemes which some companies offer to their employees. Last but not least, the reforms also include the implementation of a so-called Régime d’Assistance Maladie for people in need which is to be financed by the State and local municipalities. These reforms entrust the municipalities to set criteria for eligibility and to check requests. In total, an estimated 600’000-750’000 people are to be covered through this scheme.
Effectiveness of Malian MHOs
In order to assess the effectiveness of health insurance schemes, different measures are used. Common indicators are the number of existing and functional schemes in a given country as well as the number of beneficiaries covered. However, available data is sketchy in many countries. A survey conducted in 2003 by the Concertation revealed 56 functioning schemes in Mali with 469.815 beneficiaries. These beneficiaries represented approximately 4% of the total population. Despite the lack of accurate data, the number of insured people in Mali has increased in the last decade. However, this is due to a higher number of new MHOs rather than due to existing insurance schemes growing in size. In 1997, there existed only 7 MHOs in Mali, in 2000 the number rose to 22 and reached 102 in 2006 (Ndiaye et al. 2007: 159; Concertation 2004: 12). Bearing this in mind, the number of beneficiaries still remains modest compared to the size of the overall population.
Other impact indicators measure the effects of insurance coverage on the beneficiaries and the contribution of insurance schemes to financing healthcare provision. Here, empirical evidence is even weaker. While a direct effect on beneficiaries is their financial protection through insurance coverage, an indirect effect is the beneficiaries’ improved access to health services. The contribution of insurance schemes to financing healthcare provision is usually labelled as their resource mobilisation (Ekman 2004: 253).
Financial protection of insured and non-insured people is measured by comparing their OOP payments (e.g. consultation fee, medicines, transport, food) in the event of an illness episode. Access is measured by comparing utilisation rates (e.g. for curative services) of insured and non-insured people. Utilisation rates are calculated as follows:
No. of new contacts of insured patients (non-insured patients)*
*New contact means consultations for a new illness episode. Several visits for the same illness episode are counted as one new contact.
Finally, resource mobilisation is measured through the Cost-Recovery Ratio (CRR). It is assessed as follows:
Total annual amount of payments made by the MHO + co-payments made by beneficiaries to healthcare providers
Ekman proposes the following benchmarks for different levels of the CRR (Ekman 2004: 254):
Ratio Effect on resource mobilization
Marginal positive effects
16% - 40%
Limited positive effects
41% - 60%
Substantial positive effects
61% - 100%
Monumental positive effects
Since empirical evidence with regard to these impact measures is largely missing for Mali (Miller Franco et al. 2008), a case study two MHOs (one rural, one urban) in the Region of Ségou was conducted. Only results on utilisation rates and resource mobilisation are presented since they are based on routine data from healthcare provider and insurance scheme registers. The data on OOP expenditures of insured and non-insured people was collected in the frame of a household survey; this data is still being analysed.
Results of a case study on two MHOs in the Région of Ségou
The degree of financial protection as well as of access and resource mobilisation depends heavily on the design of the benefit package, i.e. the covered services, the co-payment and the premium level. The two evaluated MHOs in the Region of Ségou have different benefit packages. In 2009, the urban Mutuelle Interprofessionnelle de Ségou (MUTISE) had 3.969 beneficiaries which accounted for 3.8% of the population of the City of Ségou (104.108 inhabitants). The rural Mutuelle de Santé de Cinzana (MUSAC) had 2.008 beneficiaries in 2009, equalling 13% of the target population. In 2010, the number increased to 2.135 beneficiaries, 13.5% of the target population.
While in 2009, the urban MUTISE charged a one-time enrolment fee of 2.500 FCFA (5.40 USD), the rural MUSAC took 1.000 FCFA. A member is classed as the head of the household, who enrols in the scheme. The annual premium of MUTISE amounted to 5.400 FCFA (11.70 USD), while MUSAC charged 1.150 FCFA. The higher membership fee and premium of the urban scheme reflect a more generous benefit package: 75% of costs with regard to consultations and medicines, normal deliveries, birth complications (e.g. caesarean section) and hospitalization (basic surgery) are covered. Beneficiaries can choose among five healthcare providers in the City of Ségou, including its regional hospital. Furthermore, they can access selected healthcare providers in other urban centres in Mali when they are on travel. Up to 2009, the rural MUSAC covered 60% of costs incurred from consultation fees, medicines and normal delivery. Yet, beneficiaries could only use the rural health centre in Cinzana. Only in the case of birth complications (e.g. caesarean section), were beneficiaries allowed to use services of the regional hospital in Ségou. In these emergency cases, 75% of the costs were covered. The results of the effectiveness analysis do not show an improved access for insured people when comparing their utilisation rates for curative healthcare with those of non-insured patients. In all observed health services, whether urban or rural, utilisation rates of non-insured people were higher than for the insured patients. In 2009, the annual average utilisation rates of the insured were 36% (43% in 2008) in the urban zone and 16% (17% in 2008) in the rural area. For the non-insured, they amounted in the urban zone to 86% (this high percentage is most probably biased since a lot of non-insured patients of one health centre in the outskirts of Ségou City come from surrounding rural areas and hence do not form part of the target population. Percentage in 2008: 55%). In the rural area they amounted to 33% (28% in 2008). This difference may be due to the fact that the insured apply more preventive health measures which result in less illness episodes. Again, both insured and non-insured people in the urban area used health services more often than those in the rural zone. This could be due to a generally higher socio-economic level of the urban population, a bigger choice of healthcare providers, shorter distances between home and health services or due to preferences for formal (“modern”) healthcare providers. During the rainy season, when e.g. malaria and respiratory diseases are more prevalent, utilisation rates for both groups in both localities increased. In addition, the results show that utilisation rates of insured people, especially in the rural area, already reached their peak in the third quarter of the year (during the rainy season) while those of the non-insured were highest in the fourth quarter (end of the rainy season), i.e. after the harvests. This suggests that some uninsured may delay their treatment due to lack of cash.
With regard to resource mobilisation through MHOs for financing healthcare provision, in both 2008 and 2009 the average annual Cost Recovery Ratio (CRR) for the contracted urban healthcare providers were higher than for the rural provider. On average, the contracted urban providers covered 12.6% (2008) and 11.7% (2009) of their annual expenditure with the payments made by the insurance scheme and its beneficiaries. The only healthcare provider contracted by the rural scheme MUSAC covered 7.7% of its expenditures in 2008 with insurance contributions, and only 3.9% in 2009.
According to Ekman’s classification, these ratios indicate only marginal positive effects of the two analysed MHOs on resource mobilisation for financing healthcare provision. However, there are seasonal effects, with increases in the CRR during the rainy season (see graphs 1 and 2). Between July and October, the CRR for the urban scheme rose in both years to over 20%, indicating limited positive effects.
Conclusions and necessary steps to increase the impact of voluntary MHOs
In conclusion, one can observe several challenges which voluntary MHOs face: firstly, the enrolment and premium collection rates remain low. Furthermore, our data suggest that in terms of new contacts, insurance members in both urban and rural areas use health services less than non-insured people. All these factors translate into modest contributions of insurance schemes to financing healthcare provision. Finally, further data analysis will show whether in terms of OOP expenditure, their beneficiaries are financially better protected than non-insured when seeking healthcare.
In light of the fact that membership in Malian MHOs will remain voluntary in the foreseeable future, action is required at four levels in order to increase the impact of such schemes:
1. The State must fulfil its stewardship function and create an enabling institutional environment for the existing MHOs to grow;
2. On the supply side, healthcare providers must ensure adequate quality of health services, and the MHO must offer an attractive and affordable benefit package, without jeopardising the financial viability of the scheme;
3. On the demand side, better knowledge on population characteristics is required in order to target more efficiently potential members and increase enrolment rates. Furthermore, the resource base of the target population must be strengthened.
The new policy on MHOs is a step in the right direction in the sense that it is intended to establish one MHO per each municipality instead of one per each health zone. Hence the catchment area and the risk pool will be extended. However, the State must make sure that it does not only foster the creation of new (small) MHOs, but also encourage the existing ones to grow, even beyond the level of a municipality, in order to expand their risk pool. In addition, existing, functional schemes must be better linked. This would allow a continuous exchange of experiences and best practices, but eventually also a sharing of services or financial risks (e.g. through social re-insurance, see Dror & Preker 2002). Finally, within its policy to strengthen health systems, the State must link the promotion of insurance more explicitly with improvements in quality of health services. Without satisfactory quality, people are not likely to join an insurance scheme. Important elements are not only the availability of essential infrastructure, equipments, medicines and human resources. It also deals with improved health service management, compliance of the healthcare staff with treatment guidelines or better provider communication with patients.
In order to increase their attractiveness, the MHOs need to offer benefit packages that respond to people’s needs and expectations. However, increasing the benefits is only possible when the pool of insured people is big enough, unless the premium level is raised, which again might discourage potential members. Hence, adequate technical support is needed when balancing benefit package and willingness/ability to pay. MHOs should check on regular basis with their members whether the mix of primary (low-cost, high probability events) and secondary healthcare services (high-cost, low probability events) is adequate and whether the co-payment rates are appropriate. The rural scheme MUSAC improved its benefit package in 2010 with the support of the UTM. 75% of costs with regard to consultation fees, medicines and normal deliveries and 100% in the event of birth complications are now covered. Currently, the beneficiaries have the choice between 2 health centres, 2 more will be contracted by the insurance scheme in 2011. In return, the premium was raised. Both measures were approved by its General Assembly.
Technical support also needs to include continuous refresher training for all organs of the MHO and their supportive supervision. With an increasing number of beneficiaries and a more complex benefit package, a professional management is required. This is hardly possible with an unpaid management. Therefore the rural scheme MUSAC followed the example of the urban scheme MUTISE and opted for the employment of an administrator who underwent nurse training. Next to regularly checking invoices from the contracted healthcare providers and the eligibility of the treated people, the administrator has to collect annual premiums and inform and sensitise the communities about the services of the insurance scheme in order to attract new members.
Furthermore, insurance schemes should develop strategies to better target potential members. On the one hand, group enrolment, i.e. cooperatives, women groups etc. should be fostered since these groups are already organised and offer services to their members; insurance could be one of these products. On the other hand, not only urban, but also rural populations are socially more differentiated than usually reflected in the development debate. Hence, more knowledge on household and family characteristics as well as on their decision- making patterns and risk protection strategies would enable the identification of households that are more likely to buy insurance (Schulze 2010).
Last but not least, MHOs should not only be promoted in combination with quality of health services, but also be linked to initiatives that aim to improve access to micro-credit for viable income generation and to strengthen entrepreneurial skills. Ultimately, increased income allows people to pay insurance premiums and other essential services. (See ACCESS Project 2008)
*Alexander Schulze is Access Program and Research Manager, Novartis Foundation for Sustainable Development, Basel, Switzerland. Contact: email@example.com
Samuel Frei is Study Coordinator, Novartis Foundation for Sustainable Development, Basel, Switzerland
Acknowledgement: We sincerely thank Dr. Aly Barry, Coordinator of the Initiative Accès, the consultancy agency GEFRAD and the UTM for their valuable support and advice. Without them, the design of the study, data collection and interpretation would not have been possible.
- ACCESS Project (2008): Concept Note for ACCESS II, 2008-2010. Improving Access to Effective Malaria Treatment and Care through Strengthened Healthcare Services and Patients’ Resources. Unpublished project document.
- Concertation entre les acteurs du développement des mutuelles de santé en Afrique (2004): Inventaire des systèmes d’assurance maladie en Afrique: Synthèse des travaux de recherche dans 11 pays.
- Dror, David M. & Alexander Preker (ed.) (2002): Social Re Insurance. A New Approach to Sustainable Community Health Financing. ILO & World Bank. Geneva / Washington D.C.
- Ekman, Björn (2004): Community-based health insurance in low-income countries: a systematic review of the evidence. In: Health Policy & Planning, 19 (5), p. 249-270.
- Health Systems 20/20 (2011): News: Mali Adopts National Community-based Health Insurance Policy. Downloaded on 04.04.2011 under http://www.healthsystems2020.org/content/news/detail/2827/
- Kawabata, Kei, Xu, Ke & Guy Carrin (2002): Preventing impoverishment through protection against catastrophic health expenditure. In: Bulletin of the WHO. 80 (2), p. 612.
- Miller Franco, Lynn et al. (2008): Effects of mutual health organizations on use of priority health-care services in urban and rural Mali: a case-control study. In: Bulletin of the WHO, Vol. 86, No 11.
- Ndiaye, Pascal, Soors, Werner & Bart Criel (2007): A view from beneath: Community Health Insurance in Africa. In: Tropical Medicine & International Health. Volume 12, No. 2, London: Blackwell, p. 157-161.
- Novartis Foundation (2008): Document de Référence - Période 2008-2010. Initiative pour un Meilleur Accès aux Services de Santé dans la Région de Ségou. Unpublished project document.
- Schulze, Alexander (2010): Krankenkassenmitgliedschaft im ländlichen Mali. Eine neue Sicherungsform im Kontext sozialer Differenzierung. In: Dilger, Hansjörg & Bernhard Hadolt (ed.): Medizin im Kontext. Gesundheit und Krankheit in einer vernetzten Welt. Frankfurt a. M.: Peter Lang- Verlag, p. 305-328.
- Su, Tin Tin, Kouyaté Bocar & Steffen Flessa (2006): Catastrophic household expenditure for health care in a low-income society: a study from Nouna District, Burkina Faso. In: Bulletin of the WHO. 84 (1), p. 21-27.