DSM-5 Level 1 cross-cutting measure in an online sample: evaluating its latent dimensionality and utility detecting nonspecific psychological distress. (December 2020)

• Record Type: Journal Article
• Authors: Lace, John W.; Merz, Zachary C.
• Is Part Of: Psychiatry research. Volume 294(2020)
• Publication Date: 2020-12
• Subjects: Psychological distress -- Cross-cutting symptom measure -- Factor analysis -- Mental health; Psychiatry -- Periodicals; Psychiatry -- periodicals; Psychiatrie -- Périodiques; 616.89
• DOI: 10.1016/j.psychres.2020.113529
• Languages: English
• ISSNs: 0165-1781

Abstract

Highlights: Little work has investigated the factor structure and clinical utility of the Level 1 Cross-Cutting Symptom Measure within the DSM-5. Exploratory and confirmatory factor analyses supported a two-factor structure for CCSM items that captured dimensions of externalizing/serious mental illness and internalizing/affective psychopathology. Several variables derived from CCSM scores showed poor ability to identify participants who reported they were currently receiving mental health treatment, suggesting limited utility of the CCSM as an epidemiological screening measure in the present sample. Future work should critique and/or validate the factor analytic results and continue examining the CCSM's potential clinical utility in populations of interest. Abstract: Included in the most recent Diagnostic and Statistical Manual of Mental Disorders is the Level 1 Cross-Cutting Symptom Measure (CCSM), a self-report checklist with 23 items assessing 13 major psychiatric symptom clusters. To date, minimal literature has examined the factor structure of the CCSM and its utility identifying significant psychological distress, and existing studies pose notable limitations. Four hundred (400) American adults ( M age = 34.13; 50.2% female) participated online via Amazon Mechanical Turk as part of a larger study, which included the CCSM. The sample was randomly split to conduct exploratory and confirmatory factor analyses (EFAs and CFAs). EFAs revealed structures with one ( general psychopathology ) and two ( externalizing/serious mental illness and internalizing/affective ) factors accounting for 67.3% and 73.7% of the variance, respectively. CFAs indicated good fit for both models, though statistical comparison of the models via χ 2 difference test revealed the two-factor model provided significantly better fit. Areas under the receiver operating curve (AUCs) suggested that all CCSM variables of interest poorly differentiated those currently receiving mental health treatment from those who have never received mental health treatment (AUCs ranged from .57 to .68). Implications of these findings, various limitations, and recommendations for future lines of inquiry were discussed.