Predictors of PTSD Symptom Reduction in a Secondary Analysis of a Randomized Controlled Clinical Trial

Objective: In a secondary analysis of a randomized, sham-controlled trial, we prospectively investigated baseline comorbidities, demographics, and intervention as predictors of clinically meaningful (≥15-point) CAPS-5 reduction in PTSD symptom reduction. Methods: This four-year (2018-2022), two-arm, parallel-group, prospective randomized placebo controlled clinical trial was conducted at the Long Beach VA Healthcare System among 71 treatment-seeking 18-55-year-old Veterans with chronic combat-related PTSD. Hierarchical and backward multivariable logistic regression models were conducted to compare the predictive capabilities of discriminating between 15-point reduction or more in CAPS-5 at follow-up. Results: Hierarchical multivariable logistic modeling found demographic variables alone provided a nearly acceptable prediction of 15-point reduction (c-statistic = 0.69) while clinical assessments alone provided an acceptable prediction (c-statistic = 0.75). Together, the baseline demographic and clinical variables indicated strong prediction (c-statistic = 0.92) and the addition of the group intervention variable increased the prediction (c-statistic = 0.94). In a backwards stepwise regression retaining variables with an alpha = 0.10 significance, females (adjusted odds ratio (AOR) = 14.7), and those receiving acupuncture (AOR = 4.17), indicating better physical health (AOR = 1.14) and less pain (AOR = 0.95), were statistically more likely to result in a 15-point CAPS-5 reduction at follow-up after controlling for other variables in the model. Conclusions: In this small sample, demographic and baseline clinical variables were independently predictive of symptom reduction and, together with the acupuncture intervention, presented a near perfect prediction of PTSD symptom reduction, though further validation is warranted. Patient characteristics that may indicate a more favorable response for PTSD symptom reduction include less baseline pain, better physical functioning, females, increasing age, and sociodemographic variables including higher income and not employed.