Evidence-based assessment/Step 5: Gather collateral, cross-informant perspectives

=Prediction: Gather Collateral (Cross-Informant) Perspectives=

Overview
When working with children it is usually the parent who buys the ticket for the therapeutic journey – the parent initiates the referral, schedules the appointment, transports the child, and decides whether to continue with treatment. Authorities agree that routinely gathering data from multiple informants is important to understand the context of the child’s behavior. However, the agreement between perspectives is only moderate, with meta-analyses and cross-national studies finding correlations of .2 to .3 between parent, teacher, and youth report on the same measures  and .4 between adults and a collateral about internalizing or externalizing problems, rising to .68 for substance use. Clinically, the common scenario is unimpressive levels of agreement. If the average level of parent-reported concerns at a clinic had a T-score of 70 (two SDs above the mean, commonly considered “clinically elevated”), then the average level of teacher or youth reported concerns on the same scale would be T-scores of 54 to 56 – well within the normal range. Given the high internal consistency of each informant report, some interpret the data as indicating a high degree of situational specificity in behavior, though this is challenging to reconcile with models that formulate cases in terms of the presence or absence of a diagnosis. Others have interpreted findings as indicating that some informants are valid for one diagnosis, and not others. Clinicians often consider youth report more accurate for internalizing problems, and not accurate about attention problems, for example. Rater validity also may change by setting: parents may be better informants than teachers about sleep disturbance, for example, because teachers will not observe bedtime behavior or waking. More subtly, “teacher report” is not all the same context: Children spend most of the school day in the same room in elementary school, and start shifting between rooms and teachers for different subjects in middle and high school. Yet another factor contributing to the disagreement is that different informants may notice or care about different symptoms. Agreement also changes longitudinally with development.