Learning from failed health reform in Uganda

Short abstract

Evaluation of health care in developing countries from a Western perspective is masking the failures of market based reforms

Health reforms based on market principles have been introduced widely in both developed and developing countries over the past 20 years. In developing countries, international donors have insisted on health reform as a precondition of providing external aid. The reform packages that have been introduced have been strikingly similar across countries as wide apart as Uganda, Bolivia, and Russia. Uganda embarked on market based health reforms in 1994. These reforms have not only failed to improve health services and the health of the population but have arguably been the key factor behind their deterioration. What can we learn from Uganda's experience?

Health sector reforms in Uganda

Uganda introduced health sector reforms, defined by the World Bank as market reforms, in 1994.¹ The reforms were based on four cardinal market principles²:

• Individuals, charities, and private organisations should be made responsible for health care

• Public funding of health care should be restricted to health promotion and prevention of disease

• Central government's role should be restricted to policy formulation and technical guidance, with delivery of services left to the private sector and local authorities

• The private sector and non-governmental organisations should be supported to become the key providers of health and social services.¹

In accordance with these principles, the central Ministry of Health abdicated service provision to local authorities and individuals. The idea was that people should be enabled to take responsibility for their own health through economic growth, which would lead to better household incomes and allow people to buy health care from privately owned health facilities.

But although Uganda's decentralisation has been praised, it has introduced major obstacles to service provision. This is because most of the funds distributed to districts have been earmarked by donors and the Ministry of Health for a specific use. The district health authorities have had to use the funds as directed and not been allowed to deploy them to meet local needs. As a result specific local priorities are often not addressed. Decentralisation has actually widened disparities in the nature and quality of health services. For example, the availability of emergency obstetric services now varies from 4% to 42%.³ This is because the richer districts and those with powerful local politicians who have been able to persuade nongovernmental organisations to work in their district have done better. Immunisation has improved fairly uniformly in all districts, but this is unique.⁴

Other organisational reforms that were introduced included granting hospitals greater autonomy and encouraging the formation of public-private partnerships. The Ministry of Health was restructured as part of civil service reform and non-governmental organisations given a mandate to become more involved in provision and delivery of health services.

User fees

User fees were introduced as one of the key methods to finance the health reforms. The first formal attempt to introduce user fees failed in 1990, which led to fees being charged illegally on an ad hoc basis by health workers. In 1993, user fees were universally introduced as a condition for getting a World Bank loan.

User fees were expected to generate resources, promote efficient use of these resources, and improve the quality and equity of health services. Unfortunately, this did not materialise. The funds generated were typically less than 5% of total expenditure for most hospitals and health districts and they had little or no effect on the quality or efficiency of services. Furthermore, their introduction was associated with a dramatic drop in take up of health services.⁵ User fees were abolished in 2001, largely in response to the findings of the World Bank's first Participatory Poverty Assessment Report (1999), which sparked an outcry about lack of access to health care and deteriorating standards of care.⁶ A definitive study that proved fees had not achieved their aim was published in 2004.⁷

The government set an unrealistically low ceiling on health and social welfare expenditure. For example, a realistic annual budget for a large teaching hospital in 1996 was 30bn shillings (£10m, $17m, €13m) but the ceiling was set at 12.26bn in 1996, 8.87bn in 1997, and 13.28bn in 1999. Given the rise in inflation over this time, expenditure on health services (particularly on hospitals) was effectively held constant⁸ and yet the population was growing at a rate of 3% a year.⁸

Figure 1.

Introducing “user fees” in Uganda adversely affected uptake of health services

Credit: SEAN SPRAGUE/PANOS

Effect of the reforms

Official reports of the effect of health sector reform in Uganda, typically written by donor funded expatriate staff, have tended to paint a rosy picture. But the reality has often been and remains shockingly different.

Although macroeconomic targets for an inflation rate of 5.7% and economic growth of 7% a year have been attained during 1987-2004, Uganda has failed to achieve a functioning health system. Performance of health services fell after the introduction of the reforms (table 1) and key health statistics from 1990-1995 to 2000-2004 have got worse (table 2). Important health and development targets such as the millennium development goals that cannot be achieved within the social spending ceilings have now been abandoned.⁸

Table 1.

Health sector performance in Uganda during reforms⁴

Aspect of performance	Expected outcome	Actual outcome
Access	Increased physical access; increased economic access	Increased physical access through private sector. Reduced economic access.
Equity	Increased equity	Reduced equity
Quality	Increased technical quality, reduced consumer based quality	Same or reduced overall quality of health services
Efficiency	Increased output/outcome for a given input	No evidence for better efficiency
Sustainability	Increasing proportion of internal financing of public services	Over 50% dependence on donors. More dependence on external support expected
Health status	Improved nutrition and reduced mortality	Persistently high mortality and childhood malnutrition

Table 2.

Ugandan health status outcomes⁹

	1990-1995	2000-2004
Nutrition (% of children with wasting)	6.2	7-8
Access to safe water (%)	39.4	53.8
Access to proper sanitation (%)	34	51
Infant mortality (deaths/1000)	81 (97)*	88 (100)*
Neonatal mortality (deaths/1000)	27	33.2
Malaria morbidity (%)	25	37
Diarrhoea (%)	17.7	17.8
Maternal mortality ratio (deaths/100 000 births)	506	505
Child mortality (deaths/1000)	147	151
Life expectancy (years)	50	47
% of deliveries in health facilities	38%	38%
Fertility rate (No of children)	7.4	6.9

^*By indirect method.

Although access to health services is said to have increased as a result of the growth in private clinics, socioeconomic inequality within Uganda has grown and the poor, who make up the majority of the population, have become worse off.¹⁰,¹¹ Fixed funding of hospitals has not made it possible to increase staffing or provide the extra drugs and equipment needed to treat the growing population. Inequity in provisions of health services has increased in parallel with rising economic differentials; income inequality has worsened from a Gini coefficient (a measure of equity where 0 is perfect equity and 1 is perfect inequality) of 0.35 in 1992 to 0.45 in 2003.¹¹

Both consumer assessed and technical quality of health care have fallen to 30% or less.¹²,¹³ In participatory poverty assessments (studies commissioned by the World Bank to hear what poor people have to say about poverty) carried out in 1999, 2002, and 2004, most people expressed dissatisfaction with health care, asserting that it is getting worse.⁴ A recent survey of emergency obstetric facilities in Uganda, which typifies the entire health sector, indicates that technical efficiency (extent to which medical inputs and procedures conform to minimum standards) among health facilities ranged from 3.9% to 41%.³

Healthcare outputs are also more expensive than they used to be.⁵ Thus the reforms have not increased efficiency. Nor have they reduced reliance on external support. Currently, over 50% of Uganda's health sector spending comes from external aid. Re-establishment of a functioning healthcare system will require heavy investment in infrastructure, which means more dependence on outsiders not less. Thus, in all aspects of health sector performance, the picture is of failure.

Partisan interpretation

The World Bank and other institutions that advocate market style healthcare reforms typically emphasise the success of the organisational change and understate their adverse effects on health and social welfare. For example, increasing global life expectancy is cited as evidence of improved global health and welfare.¹⁴,¹⁵ Poverty, it is claimed, has been reducing, and inequity does not exist (if equity is the same as equal opportunity).¹⁵,¹⁶ When inequity is acknowledged, it is portrayed as necessary for economic growth and social mobility.¹⁷

In Uganda, both the justification for, and the effect of, market based health reform has been questionable. Poverty, which should include lack of access to basic social services, has become synonymous with income poverty, which is said to have reduced.¹¹ Yet access to basic health care and the quality of services have largely remained low or even worsened (table 1).

Ministry staff and expatriate technical advisers have cited the reduced prevalence of HIV and AIDS (from 30% in 1990s to 4.1% 2004)¹⁸ and the increased use of outpatient services as evidence for the effectiveness of health sector reforms. Yet the reduction in HIV and AIDS was due to a political strategy of openness and massive public education; it had little to do with health sector reforms. The doubling of outpatient attendance was due to the abolition of user fees not their inception.

Increasingly, donor governments in the West demand to see clear benefits from the external aid they provide. To meet this demand technical advisers now use short term surrogate output indicators such as the number of health related radio messages broadcast, the number of health workers trained, the level of the budget funded, and the availability of district health plans. This effectively hides the reality of massive failures in health and social welfare.

Some would like to blame the poor health and social welfare status of the majority of the population in Uganda on AIDS and the war. But arguably these cannot account for the evident failure of individual health reform strategies. Besides, the effect of AIDS on infant mortality is not significant.¹⁹ Furthermore, other countries that are also experiencing a protracted civil war, such as Sri Lanka, have maintained good social welfare through a non-market based health system.²⁰

What can be learnt from the market reform experience?

Several lessons can be learnt from the market reform experiment. Firstly, market economic principles are good for generating wealth but poor at improving health and social welfare. Secondly, to carry out genuine health sector reforms based on agreed health objectives, the health reforms must be disentangled from market based economic reforms. In particular, governments must guarantee social security and essential health services to all. This means macroeconomic variables should be manipulated to facilitate the attainment of social welfare targets without jeopardising economic growth. This is possible, as countries such as Costa Rica and Sri Lanka have shown.²⁰

Thirdly, the market philosophy, which is based on greed and the accumulation of profit, cannot be the proper basis for maximising welfare. Instead, the principles of solidarity and compassion should provide the basis for social welfare. The study of how to achieve maximum health within a given economy should be developed as a separate discipline from market based health economics, which is primarily concerned with minimisation of spending. Fourthly, aid and technical assistance to poor countries must be given and managed in a way that ensures a positive effect on health and social welfare. In the long run, poor countries such as Uganda should adopt a universal health and welfare framework, which they must increasingly finance from internal sources. Lastly, health and welfare must be assessed by using genuine indicators that comprehensively measure the reality of people's lives.

Summary points

Donor agencies increasingly require market based health reforms before giving grants to developing countries

In Uganda market reforms have not improved health care

Inequity in access to health care has increased and important health indicators have worsened

More emphasis needs to placed on social welfare