MODELS OF MULTIPLE RISK - PSYCHIATRIC-DISORDER AND POOR SCHOOL PERFORMANCE

The objective of this paper is to explore the relationship between multiple risk indicators and two outcomes; one or more psychiatric disorders and poor school performance in a community sample of children. The value of linear and logistic models is explored using combinations of risk indicators that have a significant association with the outcomes of interest. Multiple risk indicators appear to have a cumulative effect on both the simple prevalence of outcome and the odds ratio. The nature of the cumulative effect, however, depends on the specific risk indicators, the type of outcome examined and whether the underlying analytic model is linear or logistic. Logistic models of risk that use odds ratio as the outcome have several advantages over linear models in terms of understanding etiologic mechanisms and identifying potential treatment and prevention strategies.