P-211 - Ethical concerns about psychiatric genetic studies: item functioning and bias of the questionnaire about genetic risk

Abstract

High level of perceived concerns surrounding unethical use of genetic data could hinder participation in psychiatric genetic studies. We investigated the quality of Questionnaire about Genetic Research (QGR) for measuring the extent of perceived concern in a population with Bipolar disorder, and examined the robustness of the response items to differential bias by ethnic status. Six questions measured the degree of concerns about negative consequences from psychiatric genetic studies in the QGR. Each question has 5 response categories corresponding to increasing degree of concern, ranging from “not at all” concerned to “very concerned” We coded responses from 0 to 4, with larger scores indicating greater concerns. We used polytomous item response models (IRM) with latent constructs for analysis of complete responses by respondents. Most items of the QGR provided reliable measure of the extent of concerns. The instrument also validated the findings of greater concern about genetic studies among American Blacks compared to Whites. Item bias was observed for concerns about racism, procreation and privacy, but adjusting for these latter 3 items in a sensitivity analysis did not substantially change the association between Blacks and level of concern. For the most part, the QGR is a reliable measure of the extent of perceived concern, based on response patterns to its six items. However, cautious interpretations of some items are warranted when the ethnicity of the study sample is diverse, due to vulnerability to response bias.