Our objective of the proposed project is to study the effects of socioeconomic and demographic variables on health behaviors at the community, household, and individual levels and the impact of these behaviors on morbidity, growth, and mortality in the first two years of life. The motivating principle behind the research is that a proper understanding of child health requires the integration of social science and biological models. We designed a study and collected data specifically for this purpose. The first phase of the analysis is being completed this year under NIH grant. This phase included development of the data bases and specification and estimation of the parameters of a child health production model. The major scientific objectives of the initial grant have been achieved using information from birth through age 12 months. Important as these achievements are, constructing the health production model can be regarded as only a first step (albeit a critical one) in investigating how child health is produced. The proposed continuation grant involves consolidating, deepening, and extending that which has been achieved to date. Specifically, we propose that the work be: (1) consolidated by estimating the child health production model for the full 24-month period; (2) deepened to give a full picture of how child health is produced by incorporating descriptions of key variables (such as feeding practices) which are more complex and realistic than the (appropriately) simple measures were used in the first phase, interactions between key determinants (such as feeding practices and environmental conditions, and dynamic relations that fully represent the progression of stages in the child's health; and (3) extended by comparing the results of the """"""""full"""""""" model from this project with conventional, """"""""partial"""""""" approaches to examining key intermediate-outcome relationships. This will allow an assessment of the empirical importance of a variety of methodological innovations (including those for the proper statistical treatment of endogeneity, unobserved heterogeneity, and sample selectivity) and conclusions on the implications of these methods for interpretating existing literature on the determinants of child health and for the design and analysis of studies of child health in the future.
Showing the most recent 10 out of 28 publications
