Effects of social and cultural processes on health1

Future advances in the planning and evaluation of health service systems are highly dependent upon the development of a methodology that permits the empirical validation of hypotheses concerning the effects of social and cultural processes on health. In this paper such a methodology has been proposed. A structural model has been developed and analyzed, using the causal modeling technique of path analysis, in which social, economic, and demographic characteristics of New Mexico counties have been related to the infant mortality rate as an index of health. Path coefficients have been estimated from census data and vital statistics. Dynamic programming has been utilized to predict the combined direct and indirect effects on the health of New Mexico's population that would result from changes in the composition of the population.Analysis of the model suggests that these structural characteristics are causally interrelated, with economic factors a major determinant of infant mortality. While urbanization and migration also significantly affect infant mortality rates, education surprisingly is not important causally when compared to other variables, nor is proportion nonwhite when other social and economic factors are taken into account. In contrast, the proportion of Spanish-Americans residing in a county has a rather unexpected effect in that infant mortality rates decline as the proportion of this ethnic group rises. This finding suggests that it is important in health planning to consider cultural factors that may be unique to particular ethnic groups.In general the model building techniques and the algorithms presented here provide a valuable means of generating and testing hypotheses regarding the effects of social and cultural processes on health.