Academic journal article Bulletin of the World Health Organization

Data on Birth Weight in Developing Countries: Can Surveys Help?

Academic journal article Bulletin of the World Health Organization

Data on Birth Weight in Developing Countries: Can Surveys Help?

Article excerpt


Data on birth weight are important for several reasons: national and regional estimates of the incidence of low birth weight are internationally recognized indicators of the well-being of neonates and women of reproductive age; such estimates provide specific information on the health of pregnant women and the course of fetal development; and birth weight has been shown to be a leading determinant of the chances for survival of a newborn infant (1). Recently, the incidence of low birth weight was selected as one of the indicators for monitoring the health goals established by the World Summit for Children, i.e., to reduce the incidence of low birth weight (defined as <2500g) to no more than 10%.

The main source of data on birth weight in developing countries is statistics from health facilities. Most developing countries do not produce annual estimates of the incidence of low birth weight from these data. In many countries, the majority of babies are not born in health facilities; for example, over three-quarters of babies born recently in Indonesia, Morocco, Niger, Pakistan, and Yemen were delivered at home (2-6). Estimates limited to babies born in health facilities would therefore be subject to selection bias, as there are good reasons to assume that these babies are markedly different from the overall population. It is likely that the incidence of low birth weight would be underestimated and that the findings on the risk of low birth weight would also be misleading.

Retrospective data collected through surveys are an alternative source of information on birth weight. DaVanzo et al. showed that retrospective data from the Malaysia Family Life Survey (7) could be used to investigate the correlates of low birth weight, but that the inferences derived from reported birth weights are biased. In their analysis of the Peru Demographic and Health Survey (8), Moreno & Goldman showed that retrospective data could be used to estimate the incidence of low birth weight and concluded that estimates of birth weight based on reported weights were substantially underestimated. These analyses motivated the Demographic and Health Surveys programme (DHS) to include questions on both recalled birth weight and relative size at birth in national surveys. In this article, we use these data to assess whether crosssectional surveys can be used to improve national estimates of mean birth weight and the incidence of low birth weight. In addition, we examine whether data on size at birth can be used in analyses at the individual level, such as studies of determinants of infant mortality, child anthropometry, and effects of maternity care.

Materials and methods

DHS, which is funded by the United States Agency for International Development (USAID), assists developing countries to organize representative national surveys, which provide information for policy and programme decision-making and scientific research. Survey information includes a full birth history, which is used to estimate fertility and child mortality and various maternal and child health indicators. The survey respondents are women of childbearing age.

Since 1990, DHS surveys have included the following questions for all children born in the 5 years preceding the survey:

-- When (NAME) was born was he/she very large, larger than average, average, smaller than average, or very small?

-- Was (NAME) weighed at birth?

-- (IF YES) How much did (NAME) weigh?

Data from 15 surveys that included these questions were analysed. In 12 of 15 surveys, the question on relative size at birth was phrased exactly as above. In Cameroon and Yemen, the categories "very large" and "larger than average" were combined. In Paraguay, only three categories were used: small, average, and large. In all the surveys a question was asked about the relative size at birth of all children, including those whose mothers reported a numerical birth weight. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.