E. Hertz et al., SOCIAL AND ENVIRONMENTAL-FACTORS AND LIFE EXPECTANCY, INFANT-MORTALITY, AND MATERNAL MORTALITY-RATES - RESULTS OF A CROSS-NATIONAL COMPARISON, Social science & medicine, 39(1), 1994, pp. 105-114
Using data from United Nations sources we conducted an international c
omparison study of infant and maternal mortality rates and life expect
ancy at birth. We examined these three dependent variables in relation
to a range of independent variables including dietary factors, medica
l resource availability, gross national product (GNP/capita), literacy
rates, growth in the labor force, and provision of sanitation facilit
ies and safe water. Based on exploratory stepwise regression models, w
e fitted a series of general linear models for each of the three depen
dent variables. For the models with the highest explanatory ability, t
he percent of households without sanitation facilities showed the stro
ngest association with all three dependent variables: life expectancy
at birth (R2 = 0.83, B = -0.088, P = 0.0007); infant mortality rate (R
2 = 0.87, B = + 0.611, P < 0.0001); and maternal mortality rate (R2 =
0.54, B = + 8.297, P = 0.002). Additional significant predictors of li
fe expectancy at birth and infant mortality rate included the quantity
of animal products consumed, the percent of households without safe w
ater, excess calories consumed as fat, and the total literacy level. M
aternal mortality rate was significantly associated with total energy
consumption and excess energy consumed as fat. Using residuals from th
e general linear models we chose three outlying countries: Costa Rica,
Sri Lanka and Egypt, on which to do case studies. These country case
studies are discussed briefly in regard to characteristics that could
account for their differing statistical relationships.