2017 HSR&D/QUERI National Conference
1017 — Who Gets Evidence-Based Treatment? Factors Associated with the Receipt of Evidence-Based Psychotherapy for PTSD in VA
Lead/Presenter: Rebecca Sripada, COIN - Ann Arbor
All Authors: Sripada RK (Ann Arbor)
Pfeiffer PN (Ann Arbor)
Ganoczy D (Ann Arbor)
Bohnert KM (Ann Arbor)
The VA has mandated that evidence-based psychotherapies (EBPs) be offered to patients with PTSD, but only a small minority of the psychotherapy delivered to VA patients with PTSD is an EBP. It is unknown what factors are associated with receiving an EBP versus a non-EBP psychotherapy. The factors associated with receipt of different types of EBP are also unknown.
A nested case-control design categorized patients who received an EBP in FY2015 as cases (N = 5,583). Controls were patients at the same facility who received a non-EBP. Adjusted conditional logistic regression models examined differences in sociodemographic, mental health, and service utilization factors between cases and controls. To examine factors associated with receipt of different types of EBPs, generalized estimating equations were used to compare patients who received Cognitive Processing Therapy (CPT) to those who received Prolonged Exposure Therapy (PE), adjusting for facility.
Cases (recipients of any EBP) were more likely to have comorbid depression (OR = 1.14, CI = 1.06-1.22) and comorbid anxiety disorders (OR = 1.37, CI = 1.28-1.46) and had significantly fewer mental health visits in the year prior to the EBP (OR = 0.99, CI = 0.99-0.99). Cases were less likely to be older (OR = 0.80, CI = 0.74-0.87), service-connected for PTSD (OR = 0.63, CI = 0.59-0.67), or to be prescribed a benzodiazepine (OR = 0.82, CI = 0.75-0.89) or antipsychotic (OR = 0.90, CI = 0.82-0.99). Cases also had a lower average percent service connection for non-PTSD conditions (OR = 0.99, CI = 0.99-0.995). When recipients of CPT (n = 3,701) were compared to recipients of PE (n = 1,727), patients who received CPT were more likely to be older than 55 (OR = 1.29, CI = 1.04-1.60) and have a bipolar or psychosis diagnosis (OR = 1.26, CI = 1.01-1.57), and were less likely to be male (OR = 0.75, CI = 0.61-0.92), or have other anxiety disorders (OR = 0.82, CI = 0.69-0.97).
Recipients of EBPs differ from those who do not receive EBPs on several sociodemographic characteristics. Several indicators of treatment need (including overall mental health service use, service connection for PTSD and non-PTSD conditions, and the prescription of medications that are not first-line for PTSD) are less prevalent among cases, suggesting that these high-need veterans are less likely to receive EBPs.
Additional work is needed to determine ways to improve access to EBPs for high-need veterans with PTSD.