The Personality Inventory for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Brief Form (PID-5-BF) was developed with an assumption of invariance across sexual and gender minority (SGM) individuals. This assumption has yet to be tested empirically. Using multigroup confirmatory factor analysis, we examined measurement invariance in the PID-5-BF across the SGM status in clinical (N = 1,174; n = 254 SGM) and nonclinical (N = 1,456; n = 151 SGM) samples. Measurement invariance was supported for the PID-5-BF structure, item thresholds, and factor loadings, but not at the item intercept level. SGM individuals endorsed higher negative affectivity, antagonism, disinhibition, and psychoticism domains in both samples. In the clinical sample, adjusting for partial invariance decreased detachment and antagonism levels for SGM persons. In the nonclinical sample, adjusting for partial invariance reduced antagonism disparities in the SGM group, even rendering original group differences null. Our results support the use of the PID-5-BF in SGM populations but indicate that some measurement bias may drive observed disparities in maladaptive trait domains and, in turn, personality disorder diagnosis.
Keywords: LGBT health; alternative model for personality disorders; measurement invariance; personality disorder; personality inventory for DSM-5; sexual and gender minority.