Susan V Eisen1, Mark R Schultz1, Pengsheng Ni1, Stephen M Haley1, Eric G Smith1, Avron Spiro1, Princess E Osei-Bonsu1, Sam Nordberg1, Alan M Jette1. 1. When this work was done, Dr. Eisen was with the Center for Healthcare Organization and Implementation Research at the Edith Nourse Rogers Memorial Veterans Hospital, Bedford, Massachusetts, where Dr. Schultz and Dr. Smith are affiliated. Dr. Eisen was also formerly with the Department of Health Policy and Management, Boston University School of Public Health, Boston (e-mail: seisen@bu.edu ). Dr. Smith is also with the Department of Psychiatry, University of Massachusetts Medical School, Worcester. Dr. Ni and Dr. Jette are with the Health and Disability Research Institute, Boston University School of Public Health, Boston, where the late Dr. Haley was affiliated. Dr. Spiro is with the Massachusetts Veterans Epidemiology Research and Information Center, Jamaica Plain Campus, Department of Veterans Affairs (VA) Boston Health Care System, and with the Department of Psychiatry, Boston University School of Medicine, Boston. Dr. Osei-Bonsu is with the Center for Chronic Disease Outcomes Research, Minneapolis VA Health Care System, Minneapolis, Minnesota. Dr. Nordberg is with Atrius Health, Harvard Vanguard Medical Associates, Boston.
Abstract
OBJECTIVE: The primary purpose was to develop, field test, and validate a computerized-adaptive test (CAT) for posttraumatic stress disorder (PTSD) to enhance PTSD assessment and decrease the burden of symptom monitoring. METHODS: Data sources included self-report and interviewer-administered diagnostic interviews. The sample included 1,288 veterans. In phase 1, 89 items from a previously developed PTSD item pool were administered to a national sample of 1,085 veterans. A multidimensional graded-response item response theory model was used to calibrate items for incorporation into a CAT for PTSD (P-CAT). In phase 2, in a separate sample of 203 veterans, the P-CAT was validated against three other self-report measures (PTSD Checklist, Civilian Version; Mississippi Scale for Combat-Related PTSD; and Primary Care PTSD Screen) and the PTSD module of the Structured Clinical Interview for DSM-IV. RESULTS: A bifactor model with one general PTSD factor and four subfactors consistent with DSM-5 (reexperiencing, avoidance, negative mood-cognitions, and arousal), yielded good fit. The P-CAT discriminated veterans with PTSD from those with other mental health conditions and those with no mental health conditions (Cohen's d effect sizes >.90). The P-CAT also discriminated those with and without a PTSD diagnosis and those who screened positive versus negative for PTSD. Concurrent validity was supported by high correlations (r=.85-.89) with the validation measures. CONCLUSIONS: The P-CAT appears to be a promising tool for efficient and accurate assessment of PTSD symptomatology. Further testing is needed to evaluate its responsiveness to change. With increasing availability of computers and other technologies, CAT may be a viable and efficient assessment method.
OBJECTIVE: The primary purpose was to develop, field test, and validate a computerized-adaptive test (CAT) for posttraumatic stress disorder (PTSD) to enhance PTSD assessment and decrease the burden of symptom monitoring. METHODS: Data sources included self-report and interviewer-administered diagnostic interviews. The sample included 1,288 veterans. In phase 1, 89 items from a previously developed PTSD item pool were administered to a national sample of 1,085 veterans. A multidimensional graded-response item response theory model was used to calibrate items for incorporation into a CAT for PTSD (P-CAT). In phase 2, in a separate sample of 203 veterans, the P-CAT was validated against three other self-report measures (PTSD Checklist, Civilian Version; Mississippi Scale for Combat-Related PTSD; and Primary Care PTSD Screen) and the PTSD module of the Structured Clinical Interview for DSM-IV. RESULTS: A bifactor model with one general PTSD factor and four subfactors consistent with DSM-5 (reexperiencing, avoidance, negative mood-cognitions, and arousal), yielded good fit. The P-CAT discriminated veterans with PTSD from those with other mental health conditions and those with no mental health conditions (Cohen's d effect sizes >.90). The P-CAT also discriminated those with and without a PTSD diagnosis and those who screened positive versus negative for PTSD. Concurrent validity was supported by high correlations (r=.85-.89) with the validation measures. CONCLUSIONS: The P-CAT appears to be a promising tool for efficient and accurate assessment of PTSD symptomatology. Further testing is needed to evaluate its responsiveness to change. With increasing availability of computers and other technologies, CAT may be a viable and efficient assessment method.
Authors: Lisa A Brenner; Lisa M Betthauser; Molly Penzenik; Anne Germain; Jin Jun Li; Ishanu Chattopadhyay; Ellen Frank; David J Kupfer; Robert D Gibbons Journal: JAMA Netw Open Date: 2021-07-01