Ultimately, the significance of research on infant mortality resides in the fact that few, if any, human experiences are more tragic or emotionally devastating as the death of an infant or child. At the societal level, the loss of human potential, economic or otherwise, occasioned by infant death is dramatic. Thus, the infant mortality rate (IMR) has been employed worldwide as a key social indicator (World Bank 1998), e.g., as a ''critical test'' for identifying countries as ''superior health achievers,'' (Caldwell, 1986:173), as an inverse proxy measure of development (Pattnayak and Shai 1995), and as ''a synoptic indicator of the health and social condition of a population'' (Gortmaker and Wise 1997: 147).
The general substantive issues motivating the study of infant mortality are the same as those in other areas of research—the documentation and explanation of variation in the outcome of interest. Demographers have documented variation in infant mortality along many dimensions—temporal, spatial (neighborhoods, communities, nations, etc.), and between groups within societies. There is also a growing interest in the development of conceptual frameworks that can adequately inform studies based on multilevel models (those which include both individual and contextual variables).
Among the most prominent, specific substantive issues are those having to do with inequalities in risk according to socioeconomic status (SES), race/ethnicity, nativity, and other socially relevant categories. The very substantial research efforts directed toward identifying and modeling the effects of factors that produce variation in infant mortality risk along these dimensions mirror the concerns and goals enunciated in such recent major public health policy statements as Healthy People 2000 and Healthy People 2010.
* The author gratefully acknowledges the support for this analysis provided by the National Institute of Child Health and Human Development under Grant RO1 HD 41147.
It has proven advantageous for organizational and analytical purposes to group risk factors according to general substantive rubrics. Risk factor categories frequently employed include the demographic, socioeconomic, cultural, behavioral, and biomedical. A certain degree of theoretical utility accrues to such classifications, but it is limited by the seemingly inevitable overlap of categories. For example, there is strong consensus that the risk of infant death increases if the mother is a smoker. However, smoking can be classified as either a behavioral or a biomedical risk (or both), and some scholars view smoking behavior as representative of cultural differences (e.g., Scribner 1996).
Integral to demographic research on infant mortality are studies of three birth outcome dimensions: birth weight, gestational age, and maturity,1 not only as risk factors typically viewed as having the most powerful and proximate impact on infant mortality, but also as pregnancy outcomes of interest in their own right (Cramer 1987, 1995; Frisbie, Fovbes, and Pullum 1996; Kiely and Susser 1992; Kline, Stein, and Susser 1989).2 Among surviving infants, low birth weight, prematurity, and/or immaturity are associated with greater risk of congenital anomalies, neurodevelopmental and behavioral disorders, lower levels of educational achievement, and problems of family functioning (Hack, Klein, and Taylor 1995; McCormick 1985). Due to tremendous advances in neonatal technology and care in recent years, an increasing number and proportion of infants with highly adverse birth outcomes survive. Despite such progress, just as with infant mortality, the human costs of compromised birth outcomes for the child and its family are extremely high, as, of course, are the monetary costs of medical and social intervention designed to save such infants and maintain their health in subsequent months and years. Further, because infant mortality today is a relatively rare event in the U.S. and in many other societies, and since many of the variables believed to be crucial determinants are available only from surveys and/or small clinical data sets, it is often the case that inferences about the effects of risk factors must necessarily focus on birth outcomes which are very strongly correlated with infant mortality.
There is also an important literature on the consequences of infant mortality— consequences for demographic change in general and for reproductive behavior in particular (e.g., Palloni and Rafalimanana 1999; Preston 1978). This topic is not covered in this chapter because the focus of such research is typically on fertility differences and/or trends in population growth.
Was this article helpful?