Is universal health coverage via a national health insurance scheme financially feasible in Zanzibar?
The Government of Zanzibar is in the process of planning a health insurance scheme expected to contribute towards the aim of universal health coverage (UHC). The scheme is expected to be implemented either as part of or in collaboration with the Zanzibar Social Security Fund (ZSSF). The Zanzibar Social Security Fund Act, 1998, specifies that ZSSF is to pay medical benefits to its members and the health insurance is one way of fulfilling this legal requirement. Current coverage of ZSSF includes formal sector employees only – both public and private, however, for UHC the access to health services and financial protection of the entire population must be considered. A study to assess the financial feasibility of national health insurance (NHI) in Zanzibar was undertaken using the SimIns (health insurance simulation software) tool in July 2012. This article reports on that assessment. There is strong indication that health insurance in Zanzibar is financially feasible in the medium term leaving policy-makers with some room for designing the technical aspects of a health insurance within the financial parameters, i.e. population and costs, leaving the other dimension on health services access to be considered separately.
Le gouvernement de Zanzibar est en train de planifier la mise en place d’un régime d'assurance-santé qui devrait contribuer à l’atteinte de l'objectif de la couverture sanitaire universelle (CSU). Le système devrait être mis en œuvre soit dans le cadre de la Caisse de sécurité sociale de Zanzibar (ZSSF) soit en collaboration avec celle-ci. La Loi sur la Caisse de sécurité sociale de Zanzibar de 1998, précise que la ZSSF doit verser des prestations médicales à ses membres et l'assurance santé est un moyen de satisfaire à cette exigence légale. La couverture actuelle de la ZSSF concerne seulement les employés du secteur formel – tant public que privé ; toutefois, pour ce qui est de la CSU, il convient d’envisager que l'ensemble de la population ait accès aux services de santé et à la protection financière. Une étude visant à évaluer la faisabilité financière de l'assurance santé nationale (NHI) à Zanzibar a été réalisée en utilisant l’outil SimIns (logiciel de simulation de l'assurance santé) en Juillet 2012. Cet article rend compte de cette évaluation. Tout semble indiquer que l'assurance santé à Zanzibar est financièrement viable à moyen terme, en laissant aux décideurs une certaine marge pour la conception des aspects techniques d'une assurance santé dans le cadre des paramètres financiers, à savoir la population et les coûts, en faisant en sorte que l'autre dimension sur l'accès aux services de santé soit examiné séparément.
O governo de Zanzibar está em processo de planeamento de um esquema de seguros de saúde que se espera que contribua para a meta da cobertura sanitária universal (CSU). Espera-se que este esquema seja implementado parcialmente ou em colaboração com o Zanzibar Social Security Fund (ZSSF). A lei de 1998 deste Fundo especifica que o ZSSF deverá pagar serviços médicos aos seus membros e o seguro de saúde é uma das formas de cumprir este requisito legal. A actual cobertura do ZSSF inclui apenas funcionários do sector formal - tanto público como privado. Contudo, para a CSU, deverá ser considerado o acesso de toda a população a serviços de saúde e protecção financeira. Foi efectuado um estudo para avaliar a viabilidade financeira dos seguros nacionais de saúde (SNS) em Zanzibar, usando o instrumento SimIns (software de simulação de seguros de saúde), em Julho de 2012. Este artigo relata essa avaliação. Há fortes indícios de que os seguros de saúde em Zanzibar são financeiramente viáveis, a médio prazo, o que deixa uma certa margem aos decisores políticos para conceberem os aspectos técnicos de um seguro de saúde, dentro de certos parâmetros financeiros, ou seja, população e custos, deixando para ser considerada separadamente a outra vertente do acesso aos serviços de saúde.
Zanzibar is a state within the United Republic of Tanzania. The Ministry of Health (MoH) in Zanzibar collaborated with the President’s Office of Public Service and Good Governance (PoPSGG), the President’s Office of Finance, Economy and Development Planning (PoFEDP), the Zanzibar Social Security Fund (ZSSF), the Zanzibar Insurance Corporation (ZIC) and the Office of Chief Government Statistician (OCGS) from June 2011 to September 2012 to undertake a feasibility study on the introduction of a health insurance scheme as an option to finance health care in Zanzibar. The financial feasibility assessment consisted of calculating and projecting revenues and expenditures of the scheme from 2013 to 2021. Quantitative data from government and other sources and qualitative data from discussions with health financing stakeholders were gathered.
The team selected to lead the implementation of the health insurance scheme in Zanzibar was led jointly by the directors of the MoH and the PoPSGG and included representatives from each collaborating institution. It also included a consultant recruited to conduct a training workshop on SimIns – a health insurance simulation that allows the financial forecasting and evaluation of the financial feasibility of health insurance.
The purpose of the study was to provide a solid foundation upon which policy-makers could make an informed and evidence-based decision on the establishment of a health insurance scheme in Zanzibar as an element of health financing reform for UHC in Zanzibar. Moreover, the findings of this study can be used to design features of a potential health insurance scheme in Zanzibar.
The financial sustainability assessment of the proposed scheme made use of macroeconomic, health and demographic data. The key data required for the macroeconomic input were GDP and its growth rate, a measure of inflation, interest rates, national account data as well as public finances, i.e. general government revenues and expenditures by different categories. Most of these data were obtained from the Office of the Chief Government Statistician Zanzibar (OCGSZ), using its March 2012 statistical report.1
Other documents consulted were the OCGSZ household budget survey 2009/10, for the household consumption data and the World Economic Outlook Database of the International Monetary Fund for the US dollar inflation estimates as of April 2012.
The health care data needed for the study centred on variables that can explain what drives the cost of covering health care services by a health insurance scheme. Information on public and private expenditures was also used, although the estimate for total private expenditure on health had to be estimated in the absence of concrete data. For the preliminary scenarios the group used the approximate ratio of public to private expenditure on health from the Tanzanian mainland2 and applied it to the total government expenditure on health as taken from the public expenditure review reports, thereby making the working assumption that the ratio of total private expenditure on health to total public expenditure on health in Zanzibar is the same as that on the Tanzanian mainland.
Health care services were categorized by types of facilities and by in- and outpatient services. For each category utilization rates were calculated using full head counts of patients from every facility, as provided by the MoH Health Management Information System (HMIS).3
Data estimating the average cost per health service came from two principal sources: the 2007 review of the essential health care package (EHCP review)4 for primary-level care (PHCUs); and the MOH medium-term expenditure framework for secondary and tertiary health care. It is important to note that it was decided not to include the cost of personnel in the calculation of average cost per health service, as the current policy of having basic salaries paid directly by the government is expected to continue and thus does not affect the health insurance bottom line.
The EHCP review provided costs for PHCUs by input type. To estimate the part of the cost that is funded from MoH budgets, the team assumed that maintenance costs and 40% of drugs and supplies would represent the government-funded share of the cost.
For secondary and tertiary health care services, data from the Medium-Term Expenditure Framework (MTEF)5 was used to estimate average costs. For this, the MTEF expenditure was added by health care facility type (which did not include human resources either) and cost-sharing revenue, before receiving the limitation of budget figures from the MoF. To estimate the government-funded share of this cost, the sum was compared with MTEF following the budgetary figures provided by MoF. Non-food inflation data from OCGSZ were used to standardize figures using 2011 as the base year.
As the MTEF data do not distinguish between outpatient (OP) and inpatient (IP) services, a working assumption of the two as IP/OP cost ratio was applied to the total sums, to be able to artificially split the cost into these health care categories. Thus, the preliminary scenarios assume that admissions are, on average, twice as costly as outpatient cases. The same assumption was applied to cost-sharing data, i.e. an admitted patient is assumed to pay twice that of an outpatient in user fees.
Most of the population data were obtained from the OCGSZ economic survey, whereas the workforce data predominantly came from ZSSF/POFEDD/POPSG annual reports and database. The data sets were analysed using the SimIns tool, which helps analyse the basic financial mechanisms of health insurance. Its principal use is in financial projections for social health insurance.
Four SimIns scenarios for health insurance in Zanzibar have emerged on the basis of the data inputs described in the previous section, each varying in terms of population coverage and payment into health insurance.
Scenario 1 models a simple policy of mandatory enrolment of all working in the formal sector of the economy into the health insurance, while excluding the rest of the population. This scenario translates into a population coverage level of 16.4%. The contribution rate entered into the model corresponds to 3% of gross wages. As a result, the first scenario has only a modest impact on the overall structure of health expenditure in Zanzibar, which currently registers only around 5% of total health expenditure. In other words, the insurance as such would have little impact on the way health services are financed for the population as a whole, but would improve access to services for the formal sector.
In terms of financial feasibility, the projection shows a probable large surplus for the health insurance under the configurations of Scenario 1. As detailed in Table 1 the insurance is estimated to have a surplus worth 66% of total revenue in the first year, i.e. only a third of all revenues would be spent as expenditure on health care services for the insured and as administrative costs. This ratio of revenues to expenditures would decline during the projection period to about 50%, i.e. the health insurance would spend one Tanzanian shilling (TZS) for every two TZS in revenue.
Scenario 2 (and the following scenarios) builds on Scenario 1 by adding a gradual expansion of population coverage to include the informal sector of the economy in addition to the formal sector. In scenarios 2 to 4, the health insurance covers the entire formal sector in the first year of operation and is then modelled to extend coverage to the rest of the population gradually, until full coverage is reached in 2021. The informal sector is assumed to pay a flat contribution of TZS 60 000 per year per household, with an increase of TZS 20 000 every three years. In this scenario, it is assumed that 13.2% of the total population would be exempt from contributions, a percentage that corresponds to the estimate of the poor in the population, who are assumed to be part of the informal sector. Figure 1 shows at a glance the resulting population coverage by groups over the projection period. This coverage projection forms the basis of scenarios 2 to 4.
With health care cost und utilization variables unchanged from Scenario 1, this scenario also runs an estimated surplus. However, with the addition of the informal sector, the ratio of revenue to expenditure declines to 33% by the end of the projection period. This still represents an enormous surplus for a health insurance scheme, as can be seen in Table 2.
Scenario 3 adds a further dimension to the projection by varying the cost (in real terms) of health care. Concerns can arise about the quality of care that the health care system is able to provide. Consequently, this scenario looks at the likely implications of using the introduction of health insurance to increase the amounts spent on care, continuously by 50% over and above inflation. The scenario thus assumes that measures to implement improvements in health service quality will be put in place concurrently with health insurance expansion, and that the health insurance disbursements will fund these measures.
The results as projected by SimIns are shown in Table 3. Even with these high cost assumptions, the projection remains in surplus for all but the last year of the projection period.
Finally, Scenario 4 adds a co-payment to the projection while keeping all other variables the same as in Scenario 3. The amount of co-payment is modelled on current cost sharing levels, by taking the proportion of cost sharing revenues from all revenues from the MTEF data. As can be seen from Table 4 the difference in expenditure by the health insurance is relatively modest.
Discussion and conclusions
The four scenarios presented in this article provide a snapshot of the possible financial evolution of health insurance for Zanzibar, on the basis of readily available data and clear assumptions. Looking beyond the specific figures shown, the projections provide several useful pointers about the financial feasibility of health insurance in Zanzibar, including the strong indication that health insurance in Zanzibar is financially feasible in the medium term. The Government now needs to explore the willingness and the ability/capacities of the providers and employers to participate in the scheme.
All of the scenarios project a financial surplus, and a very considerable one in some cases. Even Scenario 3, in which the highest levels of health expenditures are modelled, shows a surplus that lasts for eight years. This leaves policy-makers with some room for designing the technical aspects of a health insurance (scheme) that would remain within acceptable financial parameters.
A key question for policy-makers concerns how national health insurance (NHI) will impact on the three UHC axes – population, costs and health services. Whereas the first scenario models coverage of the formal sector only, the other scenarios look at the implications of reaching full population coverage thus meaning that NHI would be an important vehicle for attaining UHC. Regarding the cost axis of UHC, modifying co-payments may modify the scheme and its financial viability but this could have a negative effect on achieving the objective of population coverage of UHC. Besides the population and cost aspects, policy-makers will also have to include in their projections the health services package that will meet universal health coverage goals.
In terms of population coverage, it should not be assumed that a health insurance scheme must cover all. However, the health implications of insurance should be considered for both the population group that is covered and the group that is not. In other words, should a formal sector health insurance be pursued, it is equally important to ensure that the informal sector retain the same access and quality to health care as the formal sector. This could, for example, be funded via transfers from the health insurance surplus to the publicly accessible health care services, or through increases in public funding to the health system. The same would be true for an insurance system that aims to cover the entire population: the health needs of the population groups of the informal sector should not be neglected during the years before full coverage is reached.
The projections show that, in the medium term, the financial equilibrium of a health insurance scheme can be threatened by inflation. The health insurance revenue is based on income and a flat rate for the informal sector set by policy-makers. In all scenarios it is clear that without changes in inflation, sooner or later cost control measures would have to be considered for the health insurance and/or revenue increased via higher contributions or subsidies. The crucial variables here, those of wage growth, inflation and formalization of the working population, are outside the direct influence of the health sector.
The results of this study reflect the current thinking on health insurance in Zanzibar. The variables used in the projections include the issues of population coverage (who should be covered), the contribution levels (e.g. is 3% of wages acceptable and feasible) and others. It is important that all stakeholders be consulted in the process of establishing whether these variables reflect a consensus. Such consultations should also seek to establish how a health insurance policy fits into the wider health sector strategy of Zanzibar, which also includes delivery of health services.
In addition, it is recommended that assessment of the technical details of how a potential health insurance scheme would be set up and run be started. This would include issues of technical feasibility, organizational set up, infrastructure, human resources and health services expansion capacities that are available or need to be built. The NHI plans need to be seen in the light of UHC and not in isolation, and it should be recognized that different scenarios for NHI will need different actions in a number of other areas of intervention. v
- Government of Zanzibar. Zanzibar Health Sector Public Expenditure Review, Ministry of Health and Social Welfare, 2010.
- WHO. Global Health Expenditure Database, 2012; United Republic of Tanzania National Expenditure on Health (http://apps.who.int/nha/).
- Government of Zanzibar. Health Management Information Systems pivot tables, Ministry of Health, 6 July 2012.
- Bijlmakars L et al. Review of the essential health care package in Zanzibar, Partners in International Health, 2007.
- Tanzania National Bureau of Statistics. Tanzania Demographic and Health Survey (TDHS), 2010.
- Government of Zanzibar. Zanzibar Health Bulletin, Ministry of Health, 2010.
- Government of Zanzibar. Zanzibar Labour Force Survey, Office of the Chief Government Statistician, 2008.