Adaptation and Validation of the Psychological Consequences of Screening Questionnaire (PCQ) for Cognitive Screening in Primary Care

Background Context-specific measures with adequate external validity are needed to appropriately determine psychosocial effects related to screening for cognitive impairment. Methods Two-hundred adults aged >= 65 y recently completing routine, standardized cognitive screening as part of their Medicare annual wellness visit were administered an adapted version of the Psychological Consequences of Screening Questionnaire (PCQ), composed of negative (PCQ-Neg) and positive (PCQ-Pos) scales. Measure distribution, acceptability, internal consistency, factor structure, and external validity (construct, discriminative, criterion) were analyzed. Results Participants had a mean age of 73.3 y and were primarily female and socioeconomically advantaged. Most had a normal cognitive screening result (99.5%, n = 199). Overall PCQ scores were low (PCQ-Neg: x <overline> = 1.27, possible range 0-36; PCQ-Pos: x <overline> = 7.63, possible range 0-30). Both scales demonstrated floor effects. Acceptability was satisfactory, although the PCQ-Pos had slightly more item missingness. Both scales had Cronbach alphas >0.80 and a single-factor structure. Spearman correlations between the PCQ-Neg with general measures of psychological distress (Impacts of Events Scale-Revised, Perceived Stress Scale, Kessler Distress Scale) ranged from 0.26 to 0.37 (P's < 0.001); the correlation with the World Health Organization-Five Well-Being Index was -0.19 (P < 0.01). The PCQ-Neg discriminated between those with and without a self-reported subjective cognitive complaint ( x <overline> = 2.73 v. 0.89, P < 0.001) and was associated with medical visit satisfaction (r = -0.24, P < 0.001) on the Patient Satisfaction Questionnaire. The PCQ-Pos predicted self-reported willingness to engage in future screening ( x <overline> = 8.00 v. 3.00, P = 0.03). Conclusions The adapted PCQ-Neg is an overall valid measure of negative psychological consequences of cognitive screening; findings for the PCQ-Pos were more variable. Future studies should address measure performance among diverse samples and those with abnormal screening results.