The Role of Permanent Income and Family Structure in the Determination of Child Health in the Ontario Child Health Study

A statistical association between poor child health and low family income has been well established by Lipman, Offord and Boyle (1994, 1995) and Dooley and Lipman (1995) in a series of papers using data from the Ontario Child Health Study (OCHS). The incidence of psychiatric disorders and poor school performance is higher among children of poor families than among children of non-poor families. The OCHS data generally show that children in one parent families have more problems than do children in two parent families, but this finding is less robust than that for low income. Most previous, however, has been done with the initial 1983 wave of the OCHS and little been done with the data from both the 1983 and 1987 surveys. Our objective in the proposed paper is to investigate these relationships using both waves of the OCHS data. We are particularly interested in the relationship between child health and “permanent low income” as measured by the income data from both waves. We will investigate the possibility that there is a lagged relationship between family status and child health. Cross tabular and multivariate methods will be used to analyse the relationship between socioeconomic status and child health status. We will use ordinal and cardinal measures of child health derived from the Health Utilities Index Mark 2 (Feeny et al., 1992 and Torrence et al., 1992). This HUI system affords one means of assessing the overall impact of socioeconomic status on child health status and health-related quality of life. The longitudinal results indicate that lone motherhood (both current and long-term status) was negatively associated with all outcome measures except cognition. Most notable was the fact that current low income exhibited a much weaker relationship with our health outcome measures than did longer term low income. In most cases the marginal effect on child health of long term poverty was equal to or greater than the marginal effect of ever having lived in a lone-mother family. We believe this demonstrates the necessity of obtaining measures of permanent income rather than current income levels when investigating child health status.