Elsevier

The Lancet

Volume 360, Issue 9329, 27 July 2002, Pages 284-289
The Lancet

Articles
Meeting international goals in child survival and HIV/AIDS

https://doi.org/10.1016/S0140-6736(02)09550-8Get rights and content

Summary

Background

After a decade of focused efforts to reduce under-5 mortality results have been mixed, with the HIV/AIDS epidemic often used to explain reversing trends. We investigated this assumption by estimating HIV-attributable under-5 mortality in 39 countries in sub-Saharan Africa and comparing it to total mortality in children younger than 5 years between 1990 and 1999.

Methods

Annual HIV-attributable under-5 mortality was estimated in a four-step process with previously obtained data from antenatal clinics in 39 countries based on HIV seroprevalence among pregnant women. This process produced estimates of: the number of children born to HIV-infected women; the number of children infected via mother-to-childtransmission of HIV; the number of HIV-infected children who died before age 5 years; and the number of HIV-infected children who died before age 5 years adjusted for competing causes of mortality.

Findings

An estimated 330 000 children younger than age 5 years died in sub-Saharan Africa in 1999 with HIV infection. When corrected for competing causes of mortality, HIV infection caused 7·7% of under-5 deaths in 1999, a significant rise from 2% in 1990. Five countries (Botswana, Namibia, Swaziland, Zambia, and Zimbabwe) have rates of HIV-attributable under-5 mortality of above 30 per 1000. An additional 16 countries have HIV-specific under-5 mortality rates between 10 and 25 per 1000. The remaining 18 countries have rates under 10 per 1000.

Interpretation

HIV/AIDS is an increasingly important cause of under-5 mortality in Africa. However, in many African countries, slowing or reversal of trends in child mortality cannot be explained by HIV alone. To meet international goals for reductions in child mortality, efforts must continue to focus on HIV/AIDS, but must also strengthen health systems that can deliver available interventions for the other major killer diseases of children in the developing world.

Published online April 30, 2002 http://image.thelancet.com/extras/01art9188web.pdf

Introduction

Improvements in child survival have been one of the major targets of development programmes during the past three decades, and child mortality rates have shown substantial and consistent declines in all regions of the world since 1960.1, 2 In recent years however, these positive trends have stagnated, and even reversed, in many countries in sub-Saharan Africa, whereas they have continued to improve in other regions.2

Most countries of the world have agreed to the Millennium Development Goal (MDG) of reducing child mortality by two-thirds by 2015. The UN prepared a report in 2001 on progress towards child-survival goals.3 The available information clearly shows that unless new resources and reinvigorated efforts are made by countries and their technical assistance partners, the MDG will not be met.

Results of several studies have shown that child mortality increases substantially in children who are infected with HIV via their mothers.4, 5, 6, 7, 8 In this report we address the question of whether the prevention of HIV/AIDS in children can make a significant contribution to the achievement of the MDG. We have focused on sub-Saharan Africa, where more than 85% of infections among children via mother-to-child transmission arise.9, 10

There have been previous estimates of the effect of HIV/AIDS on under-5 mortality. One of the early and most comprehensive estimation exercises was by Nicoll and colleagues.11 More recently, both the UN Population Division and the US Bureau of the Census have estimated infant and under-5 mortality in countries where HIV prevalence is high among adults.12, 13 These analyses have emphasised the effect that HIV/AIDS could have in countries with high adult prevalence.

A challenge to previous and current estimates of HIV prevalance and child survival is that for most countries in sub-Saharan Africa estimates of all-cause and causespecific under-5 mortality are based mainly on models and few data points. For a few countries there are good data on all-cause child mortality from demographic and household surveys, but these surveys do not capture cause-specific mortality. Furthermore, a recent effort to use these data as the basis for estimating the effect of HIV on child survival highlighted the fundamental limitations of this approach.2 One purpose of the work we have done is to start from an area where there is a large amount of data (HIV surveillance in pregnant women) and use scientific evidence on rates of transmission and survival of children with HIV/AIDS to produce a more robust estimate of HIV-associated under-5 mortality.

Previous estimates of HIV/AIDS and child survival have also been limited by scarcity of data on timing of mother-to-child transmission of HIV and on survival times for children infected with HIV. In this report, new data on timing of mother-to-child transmission and on survival times have been incorporated into the estimation models.

HIV prevalence rates among women visiting antenatal clinics were used as the starting point. Drawing on existing research, high and low estimates for key variables were developed, including rates of mother-to-child transmission of HIV and survival times for children infected with HIV. These variables were used to estimate the number of children who died with HIV infection. The estimates were corrected to account for competing causes of mortality, and ranges of uncertainty were developed. We aimed to measure the direct effect of HIV infection on child mortality. These may be conservative estimates, as we do not address the indirect effects that may occur due to the disruption of families and communities caused by HIV, including children who lose their parents to HIV/AIDS.14

These data are needed to assist Ministries of Health and their partners in making decisions about how to allocate resources to reduce child mortality and improve child health. This work can also serve as a stimulus to improve the amount and quality of data on HIV transmission, survival times, and prevalence, as well as cause-specific mortality among children and its determinants.

Section snippets

Methods

We estimated HIV-attributable under-5 mortality in a four-step process. First, we made estimates of the number of children born to HIV-infected women in countries in sub-Saharan Africa. Second, we estimated the number and timing of the children infected via mother-to-child transmission. Third, we used survival schedules to estimate the number of HIV-infected children who died before age 5 years. Finally, we adjusted the number of HIV-infected children who died before age 5 years for competing

Results

Data were available for a total of 39 countries in sub-Saharan Africa (table 2). Eritrea, Somalia and several small island states (eg, Cape Verde, Comoros, Mauritius, Sao Tome and Principe, Seychelles) were excluded from this analysis either because WHO all-cause mortality estimates in children younger than age 5 years were not available or because of an absence of data for prevalence of HIV. Regional totals have not been adjusted to include additional deaths from these countries because they

Discussion

We have presented estimates of the effect of HIV on under-5 mortality in sub-Saharan Africa from 1990 to 1999, with HIV seroprevalence among pregnant women as a starting point. Use of these data represents an improvement on estimates that use under-5 mortality as the starting point, in which there are few data points on which to build an estimation model. To account for uncertainties in the underlying data, we used four sets of assumptions to produce the estimates and ranges. We have assumed

References (34)

  • R Spira et al.

    Natural history of human immunodeficiency virus type 1 infection in children: a five year prospective study in Rwanda

    Pediatrics

    (1999)
  • Y Kamelan-Tano et al.

    Children with early HIV-1 infection in Abidjan, Cote d'Ivoire: immunologic and nutritional status, mortality rates, and causes of death

    (July 9–14, 2000)
  • R Bobat et al.

    Mortality in a cohort of children born to HIV-1 infected women from Durban, South Africa

    SAMJ

    (1999)
  • M Ota et al.

    Maternal HIV-1 and HIV-2 infection and child survival in The Gambia

    AIDS

    (2000)
  • T Taha et al.

    Morbidity among human immunodeficiency virus-1-infected and -uninfected African children

    Pediatrics

    (2000)
  • Report on the global HIV/AIDS epidemic, June 1998

    (1998)
  • Report on the global HIV/AIDS epidemic, June 2000

    (2000)
  • Cited by (0)

    View full text