Rapid, Adaptive Computer-Based Screener for Posttraumatic Stress Disorder Validated Among Veterans

A computerized adaptive diagnostic (CAD) screener that used 5 or 6 items effectively and rapidly assessed post-traumatic stress disorder (PTSD) among veterans. These findings were published in JAMA Network Open.

The CAD and computerized adaptive test (CAT) screeners for PTSD were developed using a bank of 211 PTSD-associated items. The tools were optimized using a bifactorial model and simulating responses. The tools were optimized by recruiting veterans (n=713) at a mountain state metropolitan Veterans Affairs Medical Center from 2017 to 2019 and validated with a subsample (n=304) of individuals.

The full and subsample had a mean age of 52.8±15.0 and 47.1±12.6 years, 82.3% and 80.9% were men, 78.6% and 72.7% were White, the mean duration of active-duty service was 8.2±7.5 and 7.2±6.4 years, and the mean number of deployments to combat zones were 1.0±1.5 and 5.5±1.8, respectively.

Compared with responses to 203 items about PTSD, a subset of 10 items was strongly correlated (r, 0.95). This 10-item CAT-PTSD assessment took a median time of 59 (interquartile range [IQR], 32-117) seconds to complete.

A sensitivity analysis was used to determine PTSD severity cutoffs. No PTSD vs mild PTSD corresponded with a CAT-PTSD score of 35 (sensitivity, 95%; specificity, 51%), 47 for mild vs moderate (sensitivity, 79%; specificity, 78%), and 60 for moderate vs severe (sensitivity, 50%; specificity, 93%).

With the subsample, the CAT-PTSD tool had an area under the curve (AUC) of 0.85 (95% CI, 0.79-0.89), which did not differ from the full 203-item PTSD questionnaire (AUC, 0.84; 95% CI, 0.79-0.89) and outperformed the PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (AUC, 0.75; 95% CI, 0.68-0.82).

For the CAD-PTSD tool, a 6-item interview of an individual without PTSD effectively assessed the individual as having a probability of PTSD of 0.01 and a 5-item interview of an individual with PTSD with a probability of 0.81.

The CAD-PTSD tool was found to be accurate (AUC, 0.91; 95% CI, 0.87-0.95), taking only 35 (IQR, 19-70) seconds to complete.

These tools may be limited by not allowing for the monitoring or evaluation of PTSD-specific symptoms.

The study authors concluded that the CAD-PTSD and CAT-PTSD tools may allow for rapid PTSD screening among veterans, potentially reducing the burden for both patient and clinician in the routine veteran health care setting.

Disclosure: Some study authors declared affiliations with biotech, pharmaceutical, device, and/or test companies. Please see the original reference for a full list of authors’ disclosures.

Reference

Brenner LA, Betthauser LM, Penzenik M, et al. Development and validation of computerized adaptive assessment tools for the measurement of posttraumatic stress disorder among US military veterans. JAMA Netw Open. 2021;4(7):e2115707. doi:10.1001/jamanetworkopen.2021.15707

This article originally appeared on Psychiatry Advisor