The Clinician Implications of Informant Variance, with Special Attention to Mania

Abstract

The Problem Multiple informants in the psychiatric assessment of children with behavior and emotional problems are a blessing and a curse. They are a blessing because one is able presumably to ascertain a much more comprehensive picture of the child’s functioning. They are a curse because clinicians and researchers are forced to reconcile information that is often conflicting. Do discrepancies tell us something about the informant (the person interviewed/completing a rating scale), something about the child, or both (De Los Reyes & Kazdin, 2005)? Informants generally consist of the parent and/or child and/or child’s teacher. Concerns about the reliability of the child’s self-reported behavior and emotional symptoms have led mental health clinicians, psychiatric epidemiologists and geneticists to rely on assessments collected from parents and teachers to measure the issues in question. Numerous studies have documented low to moderate associations between parent and teacher reports of children’s psychological symptoms, with Pearson correlations varying from approximately 0.2 to 0.4 (Achenbach, McConaughy, & Howell, 1987). It appears that prevalence (e.g., Boyle et al., 1996), genetic influence (e.g., Thapar and Rice, 2006) and diagnosis (e.g., Carlson & Blader, 2011; Owens & Hoza, 2003) varies depending on the informant. Discrepancies between informants may occur for a variety of reasons.