This essay discusses the advantages and disadvantages of charging for healthcare in Tanzania in two main ways: at the point of service (user fees) and in advance (pre-payment schemes). I will extrapolate the findings of Ansah et al's study (1) to Tanzania's rural situation and conclude that, alongside continued attempts to reduce patient costs and improve healthcare quality, a combination of user fees and pre-payment insurance schemes are currently appropriate as healthcare financing models.
Tanzania remains one of the poorest countries in the world. Poverty remains predominantly a rural phenomenon. The vulnerability of poor is increased by rising food and healthcare costs and the prevalence of disease, especially the rapid spread of HIV/AIDS. (1)
Given that Tanzania has so many poor people, inequitable healthcare access, escalating healthcare costs, how can Tanzania ensure a health system that is high quality, equitable, accessible, affordable and yet sustainable? If public and donor funding is not sufficiently sustainable, are user fees the best way to bridge this funding gap? Currently, healthcare is funded in several ways: public financing; donor funding; user fees and various pre-payment insurance schemes, e.g.the NHIF (National Health Insurance Fund), and the CHF (Community Health Fund).
In a context of scarce resources, such as in Tanzania, it is essential that decisions about changing healthcare funding models be based on as high quality evidence as possible (2). In Tanzania, stakeholder consultation prior to introducing user fees in 1993, showed a surprisingly large amount of people were willing to pay for good quality healthcare services, including three quarters of the poorest 40% of the population (3) As a result, it was decided that user fees would be introduced to increase revenue, demand, availability and quality of health services and decentralize healthcare budgets to community level (4).
Evidence unfortunately suggests that user fees have not led to the anticipated improvements in equity, access or quality of healthcare (4). The main problem is they contribute to catastrophic out-of-pocket spending and impoverishment (5). It is important to remember that there are other costs to patients other than user charges (6).
It may be that indirect costs are actually more important than user fees as a financial barrier (7).
So, user fees are inequitable and only partially effective in nature, having a disproportionately greater negative effect on utilization of health services on the poor rather than the rich (3).They therefore act as a barrier to access deterring the poorest and most vulnerable members, that is, those with greatest need for health services (7).
However, using access to healthcare as aprimary measure of the success of user fees is inadequate as it representsonly one of six types of outcome measures that could be used. The other measures are those of efficiency, equity, acceptability, effectiveness and relevance (8) and so we must balance the problems of user fees in a wider context.
Exemptions are available to protect the vulnerable against catastrophic expenditure. These apply to maternal and child health services (including immunizations), endemic diseases, cancer, AIDS and the over 60's (9). It has been shown that these exemptions are poorly implemented and often fail to protect those who need them (9).
So, it is clear that if we want to protect people from financial ruin from health costs, another system is needed to shift the financial burden away from individual patients to the community. So, risk-sharing was introduced in the form of pre-payment schemes, such as the CHF. Members all pay the same premium in advance (9). It was hoped that the CHF would improve healthcare inequities, such as access to services, reduce catastrophic spending and also reduce the burden of illness (10).
Whilst Uganda recently decided to abandon user fees which led to increased utilisation (7), in Tanzania the opposite effect was found where enrolment on the CHF led to reduced utilisation (10). The CHF was found to be more expensive than the user fee scheme. This shows that problems with user charges extend to any health insurance system where patients are charged for healthcare. So, it is of great interest to know whether a prospective study to remove health charges for pre-payment schemes would show different effects to studies analysing the retrospective impact of user fee removal.
A recent study by Ansah et al is the first prospective study designed to test this, as well as provide evidence about the heath benefits of free health insurance (2). This study is therefore timely and useful. It is a higher quality RCT and gives a different dimension to the evidence base by including health outcomes (the actual health impact of the popuation). Until now, it has been assumed that increased formal utilization of healthcare leads to improved health (7), but no studies have proven this.
The purpose of the study was to see whether removing direct costs of pre-payment schemes (free access) would improve members' health as well as increase service utilisation. The population under study (children under 5) were randomised into a pre-payment scheme allowing free primary care, or into a control group whose families paid user fees for health care (normal practice). Formal healthcare utilisation and two health outcomes -malaria-related anemia and mortality- were assessed.
The study found that removing user charges altered health-seeking behaviour moderately towards greater formal healthcare utilisation. However, enrolment encouraged greater utilisation of formal primary care only among the richest, with no effect on other-suggesting that pre-payment schemes may actually increase inequities (2). The study also showed that pre-payment schemes are not pro-poor, because the worst-off are rarely enrolled (2). As discussed earlier, user fees are only one part of the expenses incurred by the sick. So, abolishing user fees alone, or introducing health insurance schemes are not enough to relieve financial burden of using formal healthcare.
There were less clear effects on health outcomes in children. The population measures and analysis were very narrow, only health-related impacts were considered, and among all possible health benefits, only the potential gains in malaria-related outcomes were considered; and among malaria-related outcomes, the analysis was restricted solely to one indicator: the prevalence of severe and moderate anaemia. The study's statistical power was thus limited by choosing a relatively rare health outcome measure. Statistical power alone is not appropriate as a stand-alone item for interpreting community interventions such as this, as a small effect may have more significant effect on outcomes than realised. The intervention may have had more impacts for the wider community either on other heath outcomes, equity or financial and resource implications that weren't picked-up by the study design.
There were some biases in the assessment. Although the scheme benefits all members of participating households, the study only took into account a sub-population of beneficiaries (children); as a result, we can't generalize any conclusions from the population being studied to the wider demographics.
Ansah's study shows that lowering financial barriers could promote utilisation of health services. However, a wider array of indicators is needed to assess the complex effects of removing charging for healthcare. For example, there may be changes in revenue, resources and service quality. We also can't assume a direct relationship betwewen health outcomes and changes in utilisation. Health utilization is not just a result of access, but care quality and perception of services (12). Given these methodological issues, it isn't possible to draw definitive conclusions, such as that there are no health benefits to offering patients free healthcare.
I think the study findings are relevant and generalisable to Tanzania's situation as it is set in Ghana, a LMIC with similar health and economic issues. It mirrors the situation with the CHF which has not yet been proven to be pro-poor or offer a clear financial advantage over user fees (11).
In summary, this study adds to the current evidence on the limits of local health insurance systems in Africa, where the penetration rate, after more than 15 years of promotion by their organisations, remains very low (5%)(2) as in Tanzania. Other financial barriers may be more important than user fees. Healthcare inequity begins with household factors so the problem should be addressed from other levels such as, local transport funding, food policy, work opportunities, drug availability, information, gender and health-seeking and other socio-economic barriers.
However, clearly user charges increase vulnerability. There are equity arguments that they should be removed for everyone, however perversely, this may have negative consequences, such as overuse by the rich. With the elimination of fees for all, the incentive for less poor households to join risk pooling schemes would be reduced. Fees and subsidies have been found to be an important means of discouraging or encouraging preferential consumption of different health services (12)
Although policy and practice decisions are usually based on more than one study's findings, on basis of this study there is not enough evidence to suggest that pre-payment schemes provide a radically better solution than user fees, or vice versa. Ideally, we need to know the medium and long term effects on healthcare quality, utilization and sustainability of diferent fee structures. Removing user fees categorically before alternative mechanisms are in place will not necessarily solve the problem.
User fees have a current role in Tanzania for increasing resources, financial sustainability, maintaining community control of healthcare and improving staff motivation. They also increase personal responsibility, reduce healthcare inefficiency eg frivolous visits, over prescription and overuse of pharmaceuticals, and decrease hospital stays (12). However, protecting families against catastrophic spending should be a primary priority. In my opinion, users should be able to choose between the CHF as a prepayment scheme and user fees, and we should work towards making the CHF much more affordable (based on ability to pay) and preferable an option to user fees:
- Lowering direct costs such as premium fees for the CHF to make enrolment more attractive
- Address other modifiable indirect costs and access barriers (transport, sick pay, food availability)
- A wider tax-based revenue system to support exemptions and reduced insurance premiums
- Extend population coverage through the CHF
- Ensure exemptions are implemented (10) as other evidence suggests that if implemented can improve access for poor (6)
- Decide the level of cost sharing by the patients so more equitably distributed
- Increase the health budget
- Creatively encourage increased donor assistance
Some, or all of these steps, are important in helping Tanzania achieve its aim of universal health coverage, as well as an equitable and financially viable health system. Finally, it is imperative that healthcare quality is maximised as it is only high-quality formal healthcare that Tanzanian's demand, need and will strive to pay for.
- Tanzania Government (2009) Poverty Eradication http://www.tanzania.go.tz/poverty.html
- Ridde V, Haddad S. (2009). Abolishing User Fees in Africa. PLoS Med 6(1): e1000008. doi:10.1371/journal.pmed.1000008 http://www.plosmedicine.org/article/info:doi/10.1371/journal.pmed.10000085)
- Bonu S, Rani M, Bishai D, (2005) Using Willingness to Pay to investigate regressiveness of user fees in health facilities in Tanzania http://heapol.oxfordjournals.org/cgi/content/abstract/18/4/370
- Gilson, L (1996) Lessons from User Fees in Africa, EDI Health Policy: World Bank http://www.worldbank.org/hsr/library/sa/shaw.pdf#page=49
- Lagarde M, Palmer, N. (2008).The impact of user fees on health service utilization in low- and middle-income countries: how strong is the evidence? Bulletin of the World Health Organization;86:839-848.
- Jacobs B, Price N and Oeun S. (2007). Do exemptions from user fees mean free access to health services? A case study from a rural Cambodian hospital. Tropical Medicine & International Health; 12(11): 1391 - 1401
- Ansah EK, Narh-Bana S, Asiamah S, Dzordzordzi V, Biantey K, Dickson K, Gyapong JO, Koram KA, Greenwood BM, Mills A, Whitty CJ. (2009) Effect of removing direct payment for health care on utilisation and health outcomes in Ghanaian children: a randomised controlled trial. PLoS Med; 6(1):e1000007.
- Maxwell R (1984) Quality assessment in health. BMJ;288:1470-2 http://www.bmj.com/content/vol288/issue6428/
- Mushi, D (2007). Financing public health care: Insurance, user fees, or taxes? Welfare comparisons in Tanzania. SARPN http://www.sarpn.org.za/documents/d0002891/index.php
- Mtei G, Mulligan J, Ally M, Palmer N, Mills A. (2007) An assessment of the Health Financing System in Tanzania: Shield Project http://web.uct.ac.za/depts/heu//SHIELD/reports/Tanzania1.pdf
- Kamuzora P; Gibson L (2007) Factors influencing implementation of the CHF in Tanzania. Health Policy and Planning. doi:10.1093/heapol/czm001 http://heapol.oxfordjournals.org/cgi/content/full/czm001v1
- Schleimann F, Timmermans D, Makame M, Mahon J, Schmidt-Ehry B, McLaughlin J. (2005) Joint Statement on User Fees for Health in Tanzania: Health Advisors of the Agencies. http://www.districthealthservice.com/cms/upload/sectordevelopmentlink_11_7811.pdf