Experiences with user fees for health
A quick guide through the key issues
Have user fees achieved their goals?
The following sections review the experiences with user fees in achieving each of the seven goals of user fees identified in Charting the path to the World Bank's "no blanket policy" on user fees. In general, they appear to have raised less revenue than expected; have acted as a disincentive for both poor and non-poor people to use health services; and have not led to the degree of community participation envisaged under the Bamako Initiative. In addition, exemption mechanisms for the poor or very poor have in most cases failed, and there is mixed evidence on whether they have improved efficiency or quality.
Evidence does point, however, to some positive impacts. These include increased availability of drugs and medical supplies, which in turn have a positive impact on people's willingness to pay for health care. This has led to increased use of health services in some settings. In many cases, user fees are the main way of maintaining some liquidity (readily usable cash) in the lowest tiers of the health system, thus acting as essential life blood to these systems.
- Goal 1: mobilising revenues
- Goal 2: efficiency
- Goal 3: equity
- Goal 4: decentralisation and sustainability
- Goal 5: private sector development
- Goal 6: quality
- Goal 7: accountability and community participation
This page also examines Uganda's experience with abolishing user fees.
Goal 1. User fees mobilise revenues
In some African countries, the desire to increase revenues has been the main reason for adopting user fees. The share of revenues coming from user fees has been small, well below the 10 or 20 per cent anticipated in World Bank documents (see Lessons from cost recovery in health). However, Health financing in WHO member states shows that total out-of-pocket expenditure as a proportion of total health spending in Africa has increased substantially since the 1980s. There is also some evidence to suggest that revenues rise over time after being introduced, due to improved collection, more strictly imposed exemptions, increasing prices, and increased utilisation as populations become used to paying for social services.
User fee revenues have been low – but sometimes vital
Even if the share of user fee revenue seems low, it is often argued that this extra revenue source at the local level enables purchase of items that would otherwise not be afforded, such as a continuous stock of drugs, incentive payments for staff or additional staff. These have proven to be essential in keeping social services functioning and in improving the quality of care. User fees, health staff incentives, and service utilization in Kabarole District, Uganda argues that the proceeds from user fees enabled better incentives to be provided to health staff in remote areas, improving the quality of services. The effects of user fees in the Volta region of Ghana finds that user fees were important for non-salary operating costs. However, Mobilizing resources for health emphasises that these costs typically account for less than 30 per cent of total health sector costs.
Revenue figures also need to take into account the costs of collecting user fees. These include staff time and material costs for receipts, accounting, managing money, and banking. Lessons from cost recovery in health argues that little recognition has been given to these costs, or to the capacity needed for collecting fees and managing the proceeds in the interests of the poor. There is very little evidence on these costs. Some reports suggest that they are large relative to the amounts collected, although the true cost is difficult to determine when staff have multiple tasks.
Formal user fees could potentially have a beneficial effect on access if they led to a reduction in unofficial fees, which could be high or unpredictable. In practice it is difficult to determine whether this has happened, because few research studies have reported on the extent and nature of unofficial charges. See The cost of free healthcare: under-the-counter payments in Bulgaria and Free does not mean affordable: maternity patient expenditure in a public hospital in Bangladesh.
Goal 2. User fees promote efficiency
Using the right level of care
User fees could improve technical efficiency by setting higher prices for higher levels of care. This would mean that people do not use higher level facilities when lower levels can be used instead, and the incentives are in place for a referral system to work. See User financing of basic social services.
In practice there is often no such mechanism in place to penalise patients for using higher levels when lower levels would be adequate, and referral systems fail to encourage patients to seek care at the primary facility first. Sometimes this is because they perceive the quality of care to be too low in the primary care facilities. Cost recovery in Ghana finds that the use of hospitals in the Akim Region in 1991 was much higher than that of primary level facilities, and attributes this to a lack of improvement in the latter.
Other factors include that higher level facilities may seek to maximise their income by encouraging patients to use their services, irrespective of efficiency concerns. Constraints such as geographical access or inappropriate advice by health workers also prevent referral systems from working properly.
There is very little support for the idea that charges could reduce "frivolous use"
Reducing frivolous use
It is also argued that charges can reduce "frivolous" or unnecessary use of the health system. But this argument has very little empirical foundation. Reducing frivolous use requires the patient to be informed enough to distinguish a serious from a non-serious condition. Frivolous use is unlikely to occur on a large scale in developing countries due to barriers such as geographical access to care. Instead, user fees are likely to reduce preventive measures and to reduce the speed with which the patient seeks care. They may therefore lead to delays in treatment and more severe cases, and also recourse to other treatment options, such as self-treatment, which are less effective.
User fees may even worsen efficiency by encouraging providers to provide more care than is needed, if their income is determined by the amount of care they provide. For example, if providers refer patients to their own private pharmacies to buy medication, then there are clearly incentives to over-prescribe. This phenomenon, termed "supplier-induced demand," has received much attention in the context of developed countries, although documented examples from developing countries are rare.
Goal 3. User fees foster equity
If user fees can be applied to the non-poor, and the proceeds channelled to services used by the poor, then they could potentially play a redistributive role. Fees could be applied to secondary and tertiary facilities, which are used more by richer segments of the population, and used to subsidise local (especially rural), lower level services. Exemptions for the poor could also help to make fees foster equity.
Central government public health subsidies do not necessarily favour the poor. Public spending on health care in Africa: do the poor benefit? and Benefit incidence analysis show that middle and higher income groups benefit more than poorer populations. Replacing some public subsidies with revenue from user fees, with effective exemptions for the poor, could therefore have a redistributive effect.
In practice, however, user fees seem to have worsened rather than improved equity. Some research has shown that patients, even poor patients, are willing to pay for improvements in quality. But "before and after" studies have shown that user fees deter service use, especially by poor and vulnerable groups, and the higher the charges, the greater the decline.
In practice, user fees seem to have worsened rather than improved equity
Before and after studies
The equity impacts of community financing activities in three African countries concludes that the impact of fees are highly dependent on the setting and fee policies in place. For instance, the impact was positive in Benin, partially positive in Kenya, and most negative in Zambia. Similarly a World Bank Africa Region bulletin claims that attendance increased in four countries (Benin, Niger, Liberia, Zaire), decreased in one (Ghana), with mixed results in three (Guinea, Nigeria, Senegal). Strategies for pricing publicly provided health services reviews several studies showing that an increase in user fees led to a drop in demand, with a higher impact for children and the elderly as well as for poorer rural households. Equity implications of health sector user fees in Tanzania argues that user fees were regressive in Tanzania and contributed to the exclusion of poor and vulnerable groups.
However, before and after studies, especially over short time periods, do not allow for quality improvements. The starting level of quality and improvements over time, particularly with regard to the availability of drugs, may have important implications for utilisation rates. When quality improves at the same time, user charges should increase the likelihood of poor people using local health centres rather than having to travel to more distant hospitals.
But the introduction of fees has rarely led to additional resources being allocated to basic services or targeted assistance for the poor. Nor are user fee revenues generally shifted from higher to lower level facilities. Revenue usually either stays at the point of collection, or reverts to the central Ministry of Health.
Exemption mechanisms do not seem to have worked well
Exemptions and waivers
The main mechanism adopted for protecting the poor from the financial burden of health service tariffs has been the user fee waiver or exemptions (see 22 policy questions about health care financing in Africa). But exemption mechanisms and sliding fee scales do not seem to have worked well. This is due to corrupt health staff, the poor working of administrative mechanisms, and difficulties in identifying the poor. Assessment of the free health care provision system in Northern Ethiopia found no relationship between income and receiving free health care in rural areas. A case study on health in Uganda reports that record keeping was inadequate to support credible or efficient quality control and auditing; staff corruption was also a problem. In Ghana in 1995, less than one in 1000 patients were granted an exemption.
Targeting those who merit exemption or reduced fees is a major problem. Targeting by employment status (e.g. the unemployed) or special groups (civil servants, military personnel) has been more common than by geographical area or occupational group. Some of these groups that have been targeted are not the poorest, and so the targeting has actually been damaging to equity. Means testing was the most common method of targeting, and is argued to be the best way to identify a person who is poor (see 22 policy questions about health care financing in Africa). But criteria used are usually vague and income thresholds are rarely used because of the administrative costs.
More readings on waivers and exemptions...
Goal 4. User fees support decentralisation and sustainability
An appropriate pricing structure could encourage the use of district-based health facilities close to the home of the patient, rather than more distant tertiary (specialist consultancy) centres. If user fees are kept at the point of collection, this would increase resources and spending at lower levels of the system, improving their sustainability.
User fees have had some success in making local services sustainable
The evidence suggests that user fees have been considerably more successful in reaching this goal than in most of the others. More and better services have been provided at the primary care level. Key quality indicators such as availability of drugs have particularly benefited (see, for example, Assessment of Niger's national cost recovery policy implementation in the primary health care sector).
However, a possible side effect of user fees is that central government becomes less concerned with financing at decentralised levels. In general, governments tend to spend a large proportion of their budget at the central ministerial level and in national "flagship" hospitals and referral hospitals, except where there have been special drives (usually with donor backing) to decentralise spending as in Uganda, Ghana, and Tanzania.
Access to basic social services contends that user fees are often used to substitute for funding from central government. Results from Zambia suggested that provision of some key service shifted away from hospitals towards health centres after the introduction of user fees, although for overall outpatient services this trend was not so evident, and the most striking effect was a large drop in attendance at both types of facility. Priority service provision under decentralization reports that local revenue was not increased significantly by user fees for maternal and child health services, because many people were reluctant or unable to pay the fees, and the revenues raised went to the district level for reallocation elsewhere.
Goal 5. User fees foster private sector development
When public services are free of charge, the private sector cannot compete, except by providing better quality, more accessible (closer) or more convenient (e.g. longer opening times) services, for which the population are willing to pay. So user fees in the public sector could ease this constraint on private sector development. See Public spending on health care in Africa.
There is very limited research to answer specific questions about the impact of user fees on the private sector. "Shopping around" between different facilities is likely to be important mainly in urban settings, where there is more choice of health providers. For example, in Dar es Salaam, Tanzania, there are four private health facilities for every public one. But indigenous or traditional health systems, rather than the private modern sector, are often the main competitors to government facilities, especially in Africa and in rural areas.
Non-governmental organisation (NGO) and mission facilities are also important in sub-Saharan Africa. These tend to be located more in rural areas and tend to be popular due to their non-profit status and the generally positive attitudes of staff. In countries such as Tanzania and Rwanda they have often been given official status as government facilities. These facilities often have cost recovery policies, and as in the for-profit sector, they might find it harder to maintain these policies when public sector facilities are free. Thus the adoption of user fees in public facilities may make it easier for NGO facilities to survive.
However, it is not only price that determines health seeking behaviour. A substantial literature reports that patients will also compare characteristics such as payment schedule, proximity, previous treatment habits and experiences, and perceptions about likelihood of cure.
Goal 6. User fees improve quality of service
Additional revenues from user fees can be spent on different aspects of service delivery that improve the quality of care. More resources available at the decentralised levels of the government structure should also improve accountability, which in turn should improve quality. Unfortunately, the lack of data, as well as difficulties defining what constitutes quality, have made it difficult to monitor the quality impact.
Retention of collected funds at the local level where they were collected has, however, been shown to be important in many settings. Retained revenues can be spent on immediate quality improvements such as drug purchases, upkeep of buildings and equipment, and salary supplements for providers to enhance motivation. Without local retention of the revenues, the links between user fee collection, accountability and quality improvement are broken. See Cost-sharing in Kabarole District, Western Uganda.
Corruption, bureacracy, and poor planning can stop quality from improving
In many settings there have not been major observed improvements in quality. Reasons include corruption, bureaucracy, lack of planning for use of funds, insufficient funds, difficulty of obtaining the resources or services needed for major quality improvements, and inefficient spending. See User fees in government health units in Uganda. In a number of countries (Eritrea, Ethiopia, Namibia, Zimbabwe), revenues collected were sent back to the central treasury. In others, central government has withdrawn its funding because it assumed that the decentralised levels had become self-sufficient in covering recurrent expenditure.
It has been argued widely that when there are quality improvements, the negative effect of user charges on utilisation is often off-set, with some increases in utilisation found amongst even the poorest. Lessons from cost recovery in health shows that improved quality more than offset price effects of user fees, resulting in net increases in utilisation of health services in utilisation in health services in Cameroon, the Gambia, Niger, Sierra Leone, Sudan and Zaire. In Ghana the outpatient consultations held roughly constant in the years before and after introduction of user fees, with an improvement in drug supply (see Health for some? The effects of user fees in the Volta region of Ghana).
Goal 7. User fees encourage community participation in management
Whether a good is free or whether it is charged for has major implications for the way the consumer views it. If a good such as health care is free, but not good quality, the consumer may not feel motivated or enabled to complain and demand better services. One major contention of the Bamako Initiative is that the community takes more interest in public health services when they contribute financially to these services. Therefore cost sharing would lead to community involvement, which improves quality of care and makes health services more responsive to community needs.
Community involvement has been recorded in many ways, for example through community representation on health boards, election of representatives by the community, and patient feedback mechanisms. See the World Bank's background paper on the Bamako Initiative in West Africa. But because these mechanisms were often linked to the implementation of the Bamako Initiative, it is difficult to say whether they developed because of user fees or because community participation was being promoted generally. It has, however, been reported that health management committees met less frequently or not at all, and that central government was taking a larger role in the management of health units, after the discontinuation of cost sharing in Uganda.
Uganda’s experiences with abolishing user fees in health
Health user fees were abolished in Uganda in 2000. A World Bank report compares selected indicators before and after the abolition of user fees in Uganda. The results show improved access and increased use of services, less work days lost due to sickness, and a tendency for wealthy households to opt out of public services. The figure below shows the trend in monthly outpatient rates in Kisoro District, a trend which was reflected across virtually the whole of Uganda.
Nationwide, the number of new cases treated by health centres increased by 19 per cent for children under five and 31 per cent for the rest of the population. Referrals increased by 26 per cent. Monthly attendance figures are considerably higher now, although they vary according to drug availability: when the drugs run out attendance rates drop. However, these outcomes are significantly better than those predicted by the World Bank, which predicted an increase in utilisation of around 2 per cent.
However, the impacts on illness reporting, use of preventive services, and quality appear to have been limited. Drug stocks have run out more often, due partly to increased demand for services and the inability of the supply system to react to this increase. There were also some delays in fund disbursement from central level. A WHO study reports that health workers felt morale had declined after fees were abolished, and many health unit management committees stopped meeting regularly.
Despite these shortfalls, user fee abolition has been considered a success in Uganda. The country avoided falling back into the situation of fund-starved, centrally administered health services, which led to the introduction of user fees in the first place. This success has been attributed mainly to political decisions to increase the health budget and replace the lost user fee revenues with additional funding. Great efforts were undertaken to broaden the tax base, and donor budget support also increased.
A second factor credited for Uganda's success is its efforts to decentralise budgets during the 1990s. Management Committees, which had become active in the co-management of community level health centres through their role in administering user fees, remained in place but were allocated tax-funded government budgets instead.
See also: Free government health services: are they the best way to reach the poor.
More readings on health user fees in Uganda...
