BIVARIATE LOGISTIC-REGRESSION ANALYSIS OF CHILDHOOD PSYCHOPATHOLOGY RATINGS USING MULTIPLE INFORMANTS

A central issue in studies of risk factors for childhood psychopatholo gy is utilization of the information obtained about the child's mental health status from multiple informants, In this paper, the authors pr opose a new approach to the analysis of risk factor data when the outc omes are binary ratings (presence/absence of symptoms), This new appro ach has several attractive features in this setting, The strategy take n is to perform a single analysis using multivariate modeling, in whic h simultaneous logistic regressions are conducted for the outcomes giv en by each of several informants, The advantages of this approach incl ude the following: 1) it retains the complete information about case s tatus for each informant; 2) it permits assessment of informant-risk f actor interactions as well as ''overall'' risk factor effects; 3) it p rovides measures of association between the multiple informants and ad justs for the association between responses in the analysis; and 4) mi ssing data on a subset of respondents can be incorporated in a straigh tforward way, permitting all subjects with at least one informant to b e used in the analysis, To illustrate the methods, the authors present findings on risk factors for measures of ''internalizing'' and ''Exte rnalizing'' behaviors from two surveys using parent and teacher rating s of 6- to 11-year-old children in Connecticut between 1986 and 1989.