| Description |
Evaluating the effectiveness of psychotherapy services, such as through client ratings of mental health symptoms, is a vital component of maintaining quality of care. However, the performance of psychotherapy outcome measures is not fully understood. Specifically, there are gaps in knowledge regarding the convergent validity of two widely disseminated measures, the Counseling Center Assessment of Psychological Symptoms (CCAPS) and the Outcome Questionnaire (OQ). The purpose of this study is to test the comparability of the OQ and the CCAPS as psychotherapy outcome measures. The first step to address this research question is to test the convergent validity of OQ Total scores and CCAPS Distress Index scores. Further analyses examine the relationship between these two general distress scores and the subscales of both instruments, which informs the question of whether the two instruments are providing similar or unique information. Clients at one college counseling center completed both the CCAPS and the OQ at every counseling session. The primary method of analysis was multivariate multilevel modeling, in which occasions were nested within clients. More specifically, the Bayesian mixed effects models fit provides point estimates and highest posterior density (HPD) intervals from the simulated parameters. In examining the correlation of the client-level random effects for the OQ Total score and CCAPS Distress Index, the mode of the posterior distribution of the correlated random effects was r = 0.967, HPD[.962, .971], suggesting that the two measures are highly correlated. Unfortunately, when we included session number as part of the model, the multivariate multilevel model did not appear to converge appropriately. Analyses comparing various subscales on both instruments revealed high correlations frequently, though some smaller correlations did exist where they might be expected, thus demonstrating divergent validity. The CCAPS Distress Index and the OQ Total score provide very similar information. Further, the CCAPS subscales do provide some additive information beyond the general measure of distress. Thus, college counseling centers can consider other practical and psychometric factors in deciding which instrument to use, based on their center, clientele, and institution. |
