Print version ISSN 0042-9686
Bull World Health Organ vol.83 n.8 Genebra Aug. 2005
Countries need better information to receive development aid
Some donors are now only disbursing funds to countries that provide reliable data on how the money is spent and the outcome. This has led to a need for more reliable health information, but many countries are ill-equipped to provide this.
In some developing countries health agencies and donors do not know how many people are dying or what the causes of death are. But with increased donor demands for transparency and performance measurement, countries are having to provide reliable health information and efforts are under way to help them.
Countries that do not generate reliable health information or evidence lack the basis for making sound policy-decisions. "So you have a weak (health) sector with weak information. Therefore management decisions are often not rational; you don't provide enough information to get more resources, so it is vicious circle," said Bernhard Schwartlander, director of strategic information and evaluation at the Global Fund to Fight AIDS, Tuberculosis and Malaria.
Schwartlander said that more than 90% of low-income countries do not have a well-functioning health information system, which is vital for generating basic health data.
A health information system is essentially a collection of health statistics from various sources, used to derive information about health status, health care, provision and use of services, and the impact on a population's health. Until the 1990s, health information related mainly to clinical data and disease control and surveillance. Since then it has expanded to include population health data to determine major public health problems.
The first aim of a health information system is to identify health issues, said Chris Scott, a development economist at the London School of Economics. It can also be used to inform health programme design and policy choice; it can help to forecast future health scenarios as in the case of HIV/AIDS in Africa and to monitor and evaluate the implementation of health policies, Scott said.
For many low-income countries the problem is that even a rudimentary health information system costs too much. "As a result, decision-makers are unable to identify problems and needs and track progress, evaluate the impact of interventions and make evidence-based decisions on health policy, programme design, and resource allocation", according to a report by WHO's Health Metrics Network.
This network is a new global partnership, hosted by WHO, that aims to tackle the health information bottleneck in low- income countries and help those countries build or consolidate their fledgling health information systems.
"Health Metrics Network will work with countries to assess the current status, identify gaps, and develop a national plan for improving the health information system," said interim executive secretary Carla AbouZahr.
The idea is that efforts to strengthen health information systems within countries would be financed by resources in the countries themselves, for example through multilateral development partners such as the World Bank, bilateral donors and global health initiatives such as the Global Fund and the Global Alliance for Vaccines and Immunization (GAVI).
AbouZahr said that the Health Metrics Network hoped to sign agreements by October with a number of low- and middle-income countries to develop health information systems.
Helping countries build and sustain health informations systems is a challenge.
Scott said one way to address this would be for WHO to promote health information systems that are appropriate to the infrastructure, technological capacity and budgets of health ministries. In the poorest countries, it might be helpful to design and implement a simple pencil-and-paper-based health information system which contains data understood by and, therefore, more likely to be used by policy-makers at different levels, he added.
"Launching directly a nationwide, computer-based health information system which is vulnerable to power failures, software crashes and inadequate IT skills of health staff is a high-risk strategy," Scott said.
There may be a case in large federal states for experimenting with different types of health information systems in different parts of the country to see which one functions best in which set of circumstances, Scott said.
The Global Fund's Schwartlander agreed: "If solutions are brought in from outside, they tend not be sustainable … we are keen for countries to build their own capacity ... I think sustainability also means that countries have to invest and have to see health information systems as something essential".
Although Health Metrics Network was aimed at "oiling the wheels" of health information systems development, AbouZahr said it was developing a framework that countries and partners could use to guide the development of such systems.
The framework development would be informed by experiences among partners and with countries, in particular Ghana, Mexico and Thailand which are at different stages of health information systems development that are working with Health Metrics Network. The framework includes a health information systems quality assessment tool based on the work done by the International Monetary Fund (IMF) and the World Bank to improve the quality of financial data.
Health Metrics Network's partners include the World Bank and the Global Fund, which has allocated 57% of its grants to building country health information systems.
Schwartlander said the Global Fund was also working with WHO to develop an assessment tool that would identify the strengths and weaknesses of a health information system. The self assessment checklist breaks down the major components that need to be included in a decent health information system, he said.
"We are about to finalize the first draft and field test it with WHO", Schwartlander said.
Susan Stott, a manager from the World Bank's Operations Policy and Country Services Division, agreed that low-income countries need to develop health information systems particularly vital registration (or sample registration) to measure results.
"We are highly committed to improving our focus on 'results' and increasingly recognize that we're not going to be able to be more effective if our countries cannot [focus on results], and that country willingness and capacity to manage for results is fundamental to the entire development enterprise," Stott said.
Experts like Scott and Stott agree that for health expenditure to have the greatest impact on reducing mortality and disability, information is required on the diseases that have the greatest impact and on how health spending is allocated.
Scott cited the successful use of a health information system in rural districts in the United Republic of Tanzania that lacked disease burden and expenditure mapping information as an example.
In the mid-1990s in the United Republic of Tanzania most people died at home rather than in clinics or hospitals, so were excluded from the official morbidity data, while district health budgets had more than 1000 expenditure items that made it difficult to identify spending patterns.
As a result there was often a mismatch between the burden of disease and the allocation of health expenditure, Scott said. In one district, malaria accounted for 30% of years of life lost but received only 5% of health spending in 1996.
As part of an innovative pilot scheme in two districts Morogoro and Rufiji the Tanzania Essential Health Interventions Project (TEHIP) did a sample survey which asked whether anyone had died or fallen sick recently in the household, and if so, with what symptoms.
The results were used to construct a burden of disease profile for the local population and a profile of local health spending.
Comparison of the two profiles revealed the extent of the misallocation of health resources and the pattern of health spending was altered to provide a closer match with the disease burden.
"The results of this change were remarkable," Scott said. In a four-year period 199798 to 200102 the under-five mortality rate in Morogoro fell by 43%, while in Rufiji it fell by 46% between 19992000 and 200203, said Scott.
Meanwhile the director of Uganda's Bureau of Statistics J.B. Male-Mukasa told a similar success story. Uganda's health information system was developed in the 1960s, stagnated in the 1970s and 1980s, and was revitalized in the 1990s.
Uganda's Central Ministry of Health was set up to monitor epidemics of major communicable infectious diseases and had recent success in managing Ebola fever and HIV/AIDS.
"Today Uganda's health information system is strong and vibrant", said Male-Mukasa who is also a member of WHO's Health Metrics Network board.
However, one of the problems Uganda and other countries with established health information systems face is the ability to respond to new information needs, he said.
Information gathering widened from data on diseases to general information on resources for health personnel, infrastructure and financial resources for health, he added.
Information has also been collected on the main causes of disease from several national household surveys conducted by Uganda Bureau of Statistics (UBOS) since the 1990s. All of these have been used by the network to track major causes of morbidity including HIV, tuberculosis and malaria.
A key lesson was that Uganda improved its health information system by working with other national agencies, Male-Mukasa said.