Skip to navigation – Site map

HomeVolumes and IssuesSpecial Issue 8A public/private partnership expe...

A public/private partnership experiment in the area of social health protection in Tanzania

A public/private partnership experiment in the area of social health protection in Tanzania
Une expérimentation de partenariat public/privé dans le domaine de la  protection sociale contre la maladie en Tanzanie
Experimentación de una asociación pública/privada en el ámbito de la protección social frente a enfermedades en Tanzania
Bruno Galland, Josselin Guillebert and Alain Letourmy

Abstracts

An NGO, the Centre International de Développement et de Recherche (CIDR, International Centre for Development and Research), supported a network of mutual health organisations in the Mbozi district of the Mbeya region in Tanzania. The target population were households living in rural areas from the informal sector that had access to the services of a private religious hospital. The project was not in line with the national policy that, for this population category, promoted Community Health Funds (CHFs), voluntary health coverage organisations that only provide access to the district's public care structures and are therefore managed by district authorities. The results obtained by the network of mutual organisations provided an incentive for a partnership between public and private stakeholders. The partnership gave rise to the development of hybrid mutual and CHF organisations and allowed mutual organisations and the private hospital to benefit from subsidies reserved for CHFs. The partnership was replicated in the neighbouring district of Kyela, with the subsidising of contributions by the district and a private company (Biolands) that sells cocoa. The coverage by mutual organisations of people living with HIV/AIDS was organised thanks to the Elton John Foundation and associations of patients. The results of this extended partnership were even better than in the neighbouring district. The conditions for the success and sustainability of the experiment are discussed here.

Top of page

Author’s notes

These references do not take into account the progress reports of the CIDR or the MoUs signed in the context of the project

Full text

Introduction

1Partnerships between public and private stakeholders in the area of medical care are frequent in developing countries. They are rarer in the area of social health protection, where public and private organisations most often have their own areas of intervention. The experiment that was launched in Tanzania, in two districts in the Mbeya region, on the initiative of an NGO, the Centre International de Développement et de Recherche (CIDR, International Centre for Development and Research), was in this respect a valuable innovation. It involved:

  1. public stakeholders: district health authorities and the public care establishments that they manage, Community Health Funds (CHFs), which are voluntary community health insurance schemes that are nonetheless public initiatives and are managed by the district

  2. private stakeholders: private mutual-type schemes supported by the NGO, private non-profit care establishments, a private company that sells cocoa, a private foundation, and another NGO involved in the control of HIV-AIDS.

The experiment

Background and the initial project supported by the CIDR

2Tanzania began developing health protection schemes in the 1990s. Enrolment was made mandatory for civil servants and then for the State’s contract workers, who are affiliated with the National Health Insurance Fund (NHIF), which launched its services in 2001. This obligation was then broadened to the employees of private and semi-public companies, most of whom are affiliated with the Social Health Insurance Benefit under the National Social Security Fund (NSSF), which was launched in 2005. Mandatory insurance provides access to all of the country's approved public, private and charitable establishments. Approximately 12% of the population was enrolled in a mandatory scheme at the end of 2010.

3For the informal sector in rural areas, the government began promoting the CHF model, in which enrolment is voluntary, in 1996. CHFs offer modest coverage for a fairly low contribution: free access limited to the district’s public care establishments. The contributions, which vary between the districts, are collected in clinics and transferred to the district level where they are managed by the medical team. CHF funds are pooled and redistributed to primary structures to improve care quality. Representatives of the local population, appointed by the district’s administration, sit on monitoring committees which oversee the CHFs and care structures. Slightly less than 4% of the population was covered by CHFs at the end of 2010. For the urban informal sector, organisations similar to the CHFs, TIKAs, were promoted. In addition to CHFs and TIKAs, there are some mutual organisations and voluntary micro-health insurance schemes that are set up by charitable organisations and are often associated with private non-profit establishments. This category of organisation is linked to the Self Managed Health Insurance Schemes (SMHISs) that the CIDR launched in rural areas, in the Mbozi district of the Mbeya region, in the southwest part of the country.

4The project that was launched by the NGO in 2002 initially took the form of participatory micro-insurance in a set of villages; the SMHISs were networked in view of their professional management. SMHISs are different from CHFs in three key respects. First, they are paying organisations whose contributions are calculated based on risk and they are independent of care providers. Depending on the village, contributions range from 3,000 to 5,000 TSh per person and per year (€1.5 to 2.5). On the other hand, CHFs are closely linked to public establishments, are financed irrespective of their members’ risk and do not pay care providers. In this respect, it is worth comparing CHFs to micro-insurance organisations. Enrolment in the district’s CHFs costs 10,000 TSh per family (6 members) and per year (€5). Second, SMHISs adhere to the mutual system principle of governance by members, while CHFs are in reality managed by the district's administration, since the monitoring committees have no real power. Lastly, SMHIS coverage includes access to a private religious hospital in Mbozi, the Moravian Church Mbozi Hospital (MCMH), and offers an ambulance service and the transportation of dead bodies, whereas CHFs only give access to the district’s public establishments, from clinics to the district public hospital.

5The distribution of CHFs across the country is unequal, and their attractiveness is generally low. Pregnancy-related care and care for children under the age of five years are free in clinics, which adults do not systematically frequent since they complain about the weak diagnostic capacities of the nurses. Furthermore, the low cost of enrolment serves little purpose, especially since drug shortages are frequent. In the Mbozi district, competition between SMHISs and CHFs turned in favour of the former. Access to the MCMH, which is more popular than the district public hospital due to the hospitality and quality of its staff and is better stocked with drugs, and the ambulance service were advantages that did not go unnoticed by the population. The population preferred the outpatient services of the MCMH over those of the clinics. The rate of enrolment in mutual organisations was higher than the rate of enrolment in local CHFs. In 2005-2006, SMHISs accounted for 11% of the total sum of payments due to the MCMH. The number of beneficiaries increased from 721 in 2003 with 4 mutual organisations to 8,000 in 2006 with 15 village-based mutual organisations and 2 school-based mutual organisations. The coverage rate in the area of intervention was 13%.

6The success of the private project in Mbozi and especially the low attractiveness of the CHFs sparked tension between the project and the district. It was said that the SMHISs robbed the district CHFs of their members and the fact that the MCMH, a private hospital, was more attractive than the district hospital, which was excluded from SMHIS coverage, did not make the project popular among local authorities.

The partnership between the network of mutual organisations and the district, the policy of subsidies to CHFs

7However, some SMHIS members wanted to include clinics close to their village in the mutual coverage, which motivated the project to establish a partnership with the district authorities. The first act of partnership took the form of a Memorandum of Understanding (MoU) between the District Council, the network of mutual organisations and the CIDR in 2007. It gave SMHIS members access to clinics for a fee of 500 TSh per person and per year, irrespective of use and with no copayment. Five thousand mutual organisation beneficiaries chose this option.

8At the same time, the mediocre results obtained by the CHFs in terms of enrolment led the Tanzanian government to analyse the reasons behind their low attractiveness to stimulate their development. It determined that the core problem was the quality of the care services offered with enrolment. Quality improvement was dependent upon the resources contributed to public establishments, and from this standpoint, the low contribution from the CHFs was a handicap. It would have been feasible to directly fund the care structures, but the policy was to subsidise demand to motivate the structures to stimulate CHF enrolments. Subsidisation occurred through a multiplier process: it was decided that for each enrolment in a CHF, the district would receive twice the paid amount, to provide it with more resources to improve care quality and set in motion a virtuous circle: this improvement would increase demand for enrolment and over time, the subsidy could be reduced. The government set up a matching fund to boost the CHFs, since clinics were encouraged to further promote the CHFs at the same time. The fund, which initially was directly financed by the World Bank, was then financed with the State budget.

Extension of the partnership

9For the SMHISs, the matching fund was a valuable opportunity, provided that they could benefit from it, which was not evident in that these organisations were not CHFs.

10The partnership between public and private stakeholders in the Mbozi district nonetheless led to a new MoU in 2008 that extended the bases of the public/private partnership, by incorporating the MCMH in the partnership, by including the district public hospital in SMHIS coverage and especially by allowing the SMHISs to partially benefit from the matching fund. The MoU also provided for the launch of a hybrid organisation in two villages: the Self Managed Community Health Fund (SMCHF), which introduced the principles of mutual organisations into CHFs as discussed below.

11Access to the matching fund for mutual organisations was negotiated as follows: a) SMHISs would transfer the total amount of contributions to the district level, b) the district would add these contributions to those collected by the CHFs and ask the fund to double the total amount, c) once this amount was obtained, the district would return the SMHIS contributions and share the matched amount received from the fund: 25% for the SMHISs, 50% for the district, 25% for the MCMH. Under this process, the district, the network of mutual organisations and the MCMH were therefore all winners and received resources to achieve their own objectives: improve public care for the district, obtain cash for the MCMH, indirectly subsidise contributions for the network. In 2009, 32 million TSh (€16 million) were transferred by the network of mutual organisations. In return, in 2010, it received 8 million TSh (€4 million) in addition to contributions, and the MCMH also received 8 million TSh. In 2010, 39.5 million TSh (€20 million) were transferred and in 2011, nearly 55 million TSh were transferred (€27.5 million).

12Another positive step was taken in 2011, with a new MoU: the institutionalisation of the SMHIS network and the unification of the SMHIS and CHF models. The SMHIS principles will gradually be applied to the district CHFs, which will be co-managed by the SMHIS network and the Council Health Service Board. The new Self Managed Community Health Fund (SMCHF) model was set up in 5 villages in 2011, with 288 members and 1,051 beneficiaries. This time the MoU precisely described the organisational principles of this hybrid organisation:

  1. Combine the comparative advantages of the two models (SMHIS/CHF)

  2. Enhance the product’s attractiveness

  3. Separate supply and demand for care so providers may be paid based on the actual cost of care given to beneficiaries

  4. Set up a member-based CHF association with participatory governance open solely to members

  5. Organise the joint management and supervision by district authorities of collected funds

  6. Introduce a system for the professional management of information

  7. Improve coverage without increasing contributions

  8. Allocate funds (contributions and the matching fund) in a way that promptly improves the quality of care

  9. Increase the number of qualified healthcare personnel in health centres and clinics

  10. Develop an efficient transport system

13The extension of the partnership is an upshot of the results obtained by the SMHISs (Table 1).

Table 1: growth of the number of mutual organisations in the Mbozi district

Fiscal Year

Beneficiaries

Number of mutual organisations

2003-2004

721

4

2004-2005

3,589

7

2005-2006

6,638

15

2006-2007

7,978

17

2007-2008

9,792

20

2008-2009

11,228

24

2009-2010

10,837

28

2010-2011

13,643

30

14The extension is also a result of the success achieved in the neighbouring Kyela district, where the project directly developed a hybrid Community Health Insurance Fund (CHIF) model.

The project’s extension to Kyela

15Three years after the project was launched in the Mbozi district, a feasibility study was undertaken by the CIDR in the neighbouring district of Kyela, in view of creating a second SMHIS network in an area where the population mainly lives off of cocoa farming. The plan in Kyela was to launch a few SMHISs and enter into an agreement with the private religious hospital of Matema, similar to the MCMH in Mbozi. The study identified two avenues for work, one with the Kyela district, and the other with the company Biolands, the world’s largest cocoa exporter, which sells a percentage of the cocoa farmed in the area.

16The relations developed with the district were initially tense, on account of the future competition between the SMHISs and the district CHFs. They then improved, partly thanks to the proposed CHIFs, which combine participatory and professional management, but unlike the SMCHFs are directly linked to CHFs, and also thanks to the agreement with the Kyela district public hospital, the Ipinda public health centre and a few public clinics. The CHIF model is likely to enhance the CHFs and make them more attractive. CHIFs are eligible for the matching fund and a MoU was signed in 2010 with the Kyela district.

17The organisation chart for the CHIF is shown below (Figure 1).

18The company Biolands had heard about the Mbozi SMHISs and was interested in subsidising the enrolment of farmers from the Kyela area in a mutual health organisation. But it did not want health coverage to be publicly managed, and was therefore satisfied with the CHIF model. Biolands agreed to pay a lump-sum amount for five years to cover over half (60%) of the district farmers’ contributions, whether or not these farmers were its suppliers, and collected contributions from its suppliers, thus reducing the work of the CHIFs. To avoid discrimination, the district agreed to subsidise the contributions of non-cocoa-producing members. Table 2 below presents the subsidisation process. Enrolment is free for children under the age of 5 years

Table 2: financing of contributions in the Kyela district

Item

Cocoa producers

Non-cocoa producers

Contribution (in TSh per person and per year) for individuals over the age of 5 years

5,000

Subsidy

3,000

from Biolands

3,000

from the district

Beneficiary fee

2,000 (around 1 euro)

19Coverage includes outpatient services and hospitalisation at the Matema Lutheran hospital (private), the Kyela district hospital, the Ipinda health centre (public), public clinics and the clinic of a mining company. The subsidisation of contributions facilitated the rapid start-up of the CHIF network, which was launched at the end of 2010. In June 2011, there were 10,569 beneficiaries with 40 CHIFs.

Coverage for people affected by HIV/AIDS

20The Kyela district also provided an opportunity to address a problem that had not been discussed in Mbozi: coverage for people living with HIV/AIDS. The prevalence of HIV/AIDS is high in the two districts: in 2008 it was 15.8% in Kyela and 16.7% in Mbozi among blood donors, versus 7% at the national level. This reality was a limiting factor for the development of CHIFs, since the cost of risk was much higher and the enrolment of patients or people living with HIV could threaten the balance of mutual organisations. The problem was not so much testing, routine follow-up examinations and the administration of antiretrovirals (ARVs), which were covered by a national programme, as it was the financing of care for HIV patients with opportunistic diseases. These were not, in practice, covered by the national programme and represented high costs for patients. Furthermore, some exhaustive examinations were also excluded from the national programme and were therefore charged to hospital establishments.

21An opportunity presented itself to the network when the Elton John Foundation, which contacted the CIDR, expressed a desire to intervene in the region. It proposed to develop an insurance fund to cover HIV/AIDS patients that were members of mutual organisations when they were found to be seropositive or infected with the disease. Thus, the financial risk related to the non-selection of HIV/AIDS patients could be controlled by the network of mutual organisations, as the people in question would be identified by care establishments involved in the control of HIV/AIDS and the insurance fund, financially separate from the mutual organisations, would cover their costs. This system therefore made mutual organisations attractive to people living with HIV/AIDS, and also allowed for the protocol to include NGOs specialising in the control of the disease such as Mango Tree, which was already a partner of the national programme. At the same time, Mango Tree paid the contributions of 1,330 beneficiaries, agents in the control of HIV/AIDS, and then those of 301 affected people. What was set up in the Kyela district should now be replicated in Mbozi, particularly with other HIV/AIDS associations, and perhaps in other high-prevalence areas.

Discussion

The experiment’s success factors

22The partnership was developed thanks to several general and local circumstances.

23The national health financing policy was not designed to be comprehensive and the government set priorities that gave rise to multiple health coverage schemes. It focused on affirming the principle of healthcare payment, which justified the promotion of health insurance. It also privileged the launch of the NHIF and procedures for agreements with establishments, overlooking health protection in the informal sector. The CHF concept was then adopted to develop voluntary health protection, but the pilot experiment was not conclusive and the Ministry of Health did not allocate sufficient resources for the national deployment of this type of organisation. When the coordination of coverage arose as an issue for national policy, the NHIF was appointed to propose solutions tailored to the informal sector, but its field approach did not allow headquarters to take rapid action. Locally, the NHIF’s regional management, in charge of managing the practical aspects of the matching fund, had an opportunity to assess the results achieved by the network of mutual organisations supported by the CIDR and encouraged the partnership. This was especially true since it was familiar with the MCMH, which welcomed the NHIF’s local beneficiaries and with which it had signed a specific pricing agreement.

24Several other factors released the tension between the network of mutual organisations and the Mbozi district and led to negotiations between the MCMH, the SMHIS network and the district. On the one hand, some funders from the Tanzanian State for health, which also financed the NGO, were in favour of changing the CHF model and considered that the SMHISs were a valuable experiment in this respect. They therefore wanted to reduce tension between public and private stakeholders, so as to not have to choose between two antagonistic models. On the other hand, from a more general perspective, the public/private partnership concept was promoted by funders and the government and began being applied in the health sector. Thus, the Mbeya regional health directorate did not have the same original reservations as the district regarding private stakeholders. Along the same lines, the chief doctor in the Mbeya region put together a committee on the subject, whose chairmanship was entrusted to the medical director of the MCMH.

25In practice, the NGO was the driving force behind the partnership, but its actions were greatly facilitated by funders from the sector (GTZ, USAID, Swiss Cooperation, French Cooperation). Without the support that they gave to the Ministry of Health, the Regional Directorate and the NHIF, local managers probably would not have agreed to the derogations that allowed the SMHISs to be positioned in the national policy. It is true that this position was acquired experimentally, but it was not certain that the matching fund would benefit organisations other than the CHFs and channel resources to a private hospital. Funders also played an intermediary role in bringing the Elton John Foundation into the project.

What are the next steps?

26The role played by the NGO, which was intended to transfer the scheme that it supported to Tanzanian leaders, and the experimental nature of the partnership raise questions regarding its sustainability and its possible extension.

27Strictly from the viewpoint of its sustainability, the partnership first and foremost depends on external contributions of financial resources from the matching fund and Biolands. If the network of mutual organisations no longer receives contributions from the public authorities or the company, its financial viability will become questionable and private establishments will be weakened. However, the sustainability of external resources should be considered differently for these two types of contributions.

28Since the matching fund’s subsidies are essential over the short-term to public health services and the CHFs, there is no reason why it should suddenly disappear. Its end should be programmed according to the leverage effect it produces, i.e. after period of time that is sufficient to allow mutual organisations to be adequately developed. The main risk is the premature exclusion of SMHISs or CHIFs from the fund’s benefits, which would depend upon a political decision related to the universal health coverage scheme. This would be conceivable only if the experiment is negatively assessed and if there are alternatives to the current coverage for the informal sector. As it is, the CHIF model appears promising and it will take some time for the national health coverage scheme to significantly evolve. Total reconsideration of the model (co-management and separation of buyers and payers through the development of an independent association) is the main political risk. The argument of the experiment’s critics is that it will lead to the privatisation of the CHF.

29Is the contribution from Biolands sustainable? It should first be noted that the company’s commitment is limited (five years) but significant in terms of time and its contribution goes beyond its own suppliers. All of the district’s farmers, whether or not they sell their cocoa beans to Biolands, benefit from subsidies for contributions, as long as they agree to enrol in a CHIF. It is likely that Biolands expects this strategy to increase the number of its suppliers. In principle, the partnership will have a windfall effect for Biolands, but it may prove counterproductive. Firstly, the results in commercial terms may be less promising than expected. The company could then decide to subsidise only its own suppliers, or even leave the partnership. It could even be forced to do so by the public authorities due to unfair competition, since other cocoa-selling companies may fear or observe a loss of suppliers. While the monopolistic position of Biolands was almost too obvious thanks to the subsidisation of mutual organisations, it could be called into question. The best antidote here would undoubtedly be to bring other companies into the partnership, but this would require mediation by public authorities that will not occur until the government truly attempts to bring into line the process that will lead to universal coverage.

30The development of the national scheme for universal coverage may either draw from this experiment or move in another direction. In terms of the general development of social protection in the country, it should be noted that the SMHISs and CHIFs are examples that combine micro-health insurance and free care, since children under the age of 5 do not pay the premium. In this respect, they are consistent with the general trend that can currently be observed in international discussions.

31The NHIF is officially in charge of extending health coverage to the entire population. It is to develop a selection of products and distribute it through organisations that are capable of intervening in households in the informal sector. It is clear that the CHFs would need to be transformed to play this role, and the CHIF model from this standpoint is a valuable solution that should be replicated elsewhere. The fact remains that new organisations, set up at district level, having real capacities and managed by the NHIF, could, according to the Rwandan model, take the place that is currently held in Mbozi and Kyela by the networks of participatory management organisations that the project promoted. It would certainly be logical if, along this line, the CHIFs became district agencies for the NHIF, especially since the SMHISs and CHIFs have capacities, particularly in terms of mobilisation, that the NHIF does not have.

32One can hope that the results achieved through the current partnership will be used to determine the best path for the country.

Top of page

Bibliography

GTZ, Health Financing Reform in Tanzania, joint presentation by MoHSW and GTZ, Tanzanian Kenyan Financing Exchange Meeting, 20 May 2011

Mc Intyre D., Garshong B., Mtei G., Meheus F., Thiede M., Akazaly J., Ally M., Aikins M., Mulligan J-A., Goudge J., Beyond fragmentation and towards universal coverage: insights from Ghana, South Africa and the United Republic of Tanzania, Bulletin of WHO, vol.86, n°11, November 2008

Ministry of Labour, Youth Development and Sports, The National Social Security Policy, January 2003

Mtei G., Mulligan J.-A., Community Health Funds in Tanzania: A literature review, Ifakara Health Institute, January 2007

Top of page

References

Electronic reference

Bruno Galland, Josselin Guillebert and Alain Letourmy, “A public/private partnership experiment in the area of social health protection in Tanzania”Field Actions Science Reports [Online], Special Issue 8 | 2013, Online since 06 November 2012, connection on 28 March 2024. URL: http://journals.openedition.org/factsreports/1826

Top of page

About the authors

Bruno Galland

CIDR, Director of the Health Systems and Social Protection Department

Josselin Guillebert

CIDR, Technical Advisor for the Programme to Promote Micro-Health Insurance Institutions in Tanzania

Alain Letourmy

CERMES/CNRS, Economist, letourmy@vjf.cnrs.fr

Top of page

Copyright

CC-BY-4.0

The text only may be used under licence CC BY 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search