Review
Estimates of world-wide distribution of child deaths from acute respiratory infections

https://doi.org/10.1016/S1473-3099(01)00170-0Get rights and content

Summary

Acute respiratory infections (ARI) are among the leading causes of childhood mortality. Estimates of the number of children worldwide who die from ARI are needed in setting priorities for health care. To establish a relation between deaths due to ARI and all-cause deaths in children under 5 years we show that the proportion of deaths directly attributable to ARI declines from 23% to 18% and then 15% (95% confidence limits range from ±2% to ±3%) as under-5 mortality declines from 50 to 20 and then to 10/1000 per year. Much of the variability in estimates of ARI in children is shown to be inherent in the use of verbal autopsies. This analysis suggests that throughout the world 1·9 million (95% CI 1·6–2·2 million) children died from ARI in 2000, 70% of them in Africa and southeast Asia.

Section snippets

Age range

Mortality declines substantially with age over the first 5 years of life and the age-dependent pattern of mortality varies substantially for different diseases.6 Perinatal and neonatal mortality are conventionally judged to extend to the age of 7 days and 1 month, respectively; infant mortality extends up to the age of 1 year, and childhood mortality extends up to the age of 5 years. We are concerned with under-5 mortality expressed as 5m0, the number of children that die each year as a

Results

In figure 1 the percentage of deaths attributable to ARI is plotted against the total mortality in children for the data assembled by Garenne et al.5 The line, which is a log-linear fit to Preston's data,6 is consistent with the data of Garenne et al5 but only explains 4% of the variance. The data assembled for this study are given in table 1 and plotted in figure 2 with a log-linear curve fitted to the data and weighted as described in the panel. The fitted curve then explains 61% of the

Discussion

The analysis presented here confirms the association put forward by Garenne et al5 on the basis of their analysis of Preston's data6 in which the proportion of deaths attributable to ARI diminshes as general mortality diminshes. As under-5 mortality diminshes from 50 to 20, 10 and then 5/1000 children at risk per year, the proportion of deaths attributable to ARI decreases from 23% to 18%, 15%, and then 11% (with 95% confidence limits of about ±2% to ±3% at all levels). There is, however, a

Conclusion

ARI is still one of the most important causes of childhood mortality in poor countries. It is important to obtain reliable estimates of the number of children who die from ARI and to understand how and why this varies among communities.

We have been able to show that the proportion of deaths due to ARI declines as general mortality declines following a log-linear association of the form suggested by Garenne et al5 on the basis of their analysis of historical data collected by Preston.6 Several

Search strategy and selection criteria

Articles referring to mortality in children were identified from previous reviews, a searches of Medline and unpublished work, as well as an extensive, unpublished electronic and a paper-based search were done by the Department of Child Health of the World Health Organization. Keywords in the search of electronic databases were “pneumonia”, “acute respiratory infections”, “childhood mortality” and “deaths”, and “verbal autopsies”.

The variability can be reduced if care is taken only to

References (33)

  • RR Puffer et al.

    Patterns of mortality in childhood

    (1973)
  • C Victora

    Potential interventions to improve the health of mothers and children in brazil

    Paper commissioned by the World Bank

    (March, 2000)
  • K Coakley et al.

    The Asaro Valley surveillance unit of the Papua New Guinea Institute of Medical research: methodology, demography and mortality report

    (1993)
  • D Shamedo et al.

    The Butajira rural health project in Ethiopia: mortality pattern of the under fives

    J Trop Pediatr

    (1991)
  • AH Baqui et al.

    Causes of childhood deaths in Bangladesh: results of a nationwide verbal autopsy study

    Bull World Health Organ

    (1998)
  • JS Spika et al.

    Acute lower respiratory infections: a major cause of death in children in Bangladesh

    Ann Trop Paediatr

    (1989)
  • Cited by (0)

    View full text