The burden of disease among the global poor

Summary

Background

Global and regional estimates show that noncommunicable diseases in old age are rising in importance relative to other causes of ill health as populations age, and as progress continues against communicable diseases among infants and children. However, these estimates, which cover population groups at all income levels, do not accurately reflect conditions that prevail among the poor. We estimated the burden of disease among the 20% of the global population living in countries with the lowest per capita incomes, compared with the 20% of the world's people living in the richest countries.

Methods

Estimates for the global poorest and richest 20% were prepared for 1990 for deaths and disability-adjusted life years (DALYs), by a procedure used in a prominent recent study of the global disease burden. Projected mortality rates in the year 2020 were established for the world's poorest and richest 20% under various assumptions about the future rate of decline in communicable and noncommunicable diseases.

Findings

In 1990, communicable diseases caused 59% of death and disability among the world's poorest 20%. Among the world's richest 20%, on the other hand, noncommunicable diseases caused 85% of death and disability. A raised baseline rate of communicable disease decline between 1990 and 2020 would increase life-expectancy among the world's poorest 20% around ten times as much as it would the richest 20% (4·1 vs 0·4 years). However, the poorest 20% would gain only around a quarter to a third as much as the richest 20% from a similar increase in noncommunicable diseases (1·4 vs 5·3 years). As a result, a faster decline in communicable diseases would decease the poor-rich gap in 2020, but under an accelerated rate of overall decline in non-communicable diseases, the poor-rich gap would widen.

Interpretation

Our estimates are crude, but despite their limitations, they give a more accurate picture of changes in attributable mortality among the world's poor than do the global averages in current use.

Introduction

In recent years, international researchers have paid increasing attention to the demographic-epidemiological transition: the process by which falling fertility and mortality produce markedly rapid declines in communicable disease among the young, leading to ageing populations with a rising proportion of older members among whom chronic disorders predominate.1, 2, 3, 4, 5, 6 Recent studies show that, because of this transition, noncommunicable disorders among adults and the elderly already cause a majority of the world's death and disability, and that the relative importance of this type of disorder is likely to increase greatly over the coming decades.

The Global Burden of Disease study by Christopher Murray, Alan Lopez, and colleagues7, 8, 9, 10, 11 provided the first comprehensive estimates of prevailing disease patterns worldwide. Murray, Lopez, and colleagues began preparation of their estimates by assembling and correcting data available from vital registration systems, sample registration programmes, and small-scale population studies. For areas without sufficient data from sources like these, they based their data on the advice of informed observers; and used statistical techniques based on earlier work12, 13, 14 that had shown a systematic relation between a society's overall mortality level and the pattern of diseases prevailing within it. By use of such approaches, the researchers produced estimates of global disease patterns for the year 1990 and projected them to the year 2020. They also developed a new measure of health status, the disability-adjusted life year (DALY), which took disability and mortality into account; and they prepared DALY estimates for 1990 and for each subsequent decade until 2020.

That research showed that by 1990, non-communicable diseases had already overtaken communicable diseases as the leading cause of mortality worldwide (56% of all deaths, compared with 34% attributable to communicable diseases, and 10% caused by accidents and injuries). Non-communicable diseases were estimated to cause 73% of all deaths worldwide by 2020, compared with an estimated 15% of deaths from communicable diseases. The trend expressed in terms of DALYs was only slightly less marked.

Such findings, and the approach that produced them, are important and welcome advances. However, the estimates produced thus far do not apply necessarily to the poor, who have been historically the population group of greatest concern to the international health community. Estimates currently give societal averages that include data for rich as well as poor. Average figures would accurately represent the poor only if disease patterns were similar across income groups.

Such similarity is unlikely, since the pattern of disease found in a population group varies systematically according to its longevity. Inferences about health status among the global poor cannot be drawn from global average figures. The work undertaken thus far must therefore be extended if it is to produce estimates that are specific to the poor.

What follows is an initial step in this direction. Our work has been prepared for the World Bank, to help guide implementation of the Bank's recent health policy, which gives highest priority to improvements in the health, nutrition, and population status of the world's poor.¹⁵ The work features two types of estimate. The first concerns the health of the 20% of the world's population living in those countries with the lowest per capita incomes in 1990 (and, for comparison, that of the 20% of the world's population living in the richest countries). The second deals with the implications of disease-reduction strategies during the 1990–2020 period for the world's poorest and richest 20%. The estimates are necessarily crude: our objective is not perfection but a closer approximation of the health status among the world's poor than is available from global or regional averages.