1007 — Measurement Strategies for Evidence-Based Psychotherapy for Posttraumatic Stress Disorder Delivery: Trends and Associations with Patient-Reported Outc

Lead/Presenter: Brian Shiner,  National Center for PTSD
All Authors: Shiner B (National Center for PTSD), Leonard Westgate C (White River Junction VA Medical Center), Maguen S (San Francisco VA Medical Center) Gui J (Geisel School of Medicine at Dartmouth) Watts BV (National Center for Patient Safety) Hoyt J (White River Junction VA Medical Center) Cornelius SL (White River Junction VA Medical Center) Schnurr PP (National Center for PTSD)

Objectives:
Evaluation of the implementation and clinical effectiveness of evidence-based psychotherapy (EBP) for PTSD in the VA has been limited by lack of patient-level data on treatment receipt. Our objectives were to measure the longitudinal implementation of EBPs for PTSD to a national cohort of Veterans and to determine whether more stringent treatment adequacy standards are associated with superior outcomes.

Methods:
We quantified EBP receipt during one year of treatment using both administrative data and natural language processing (NLP) of psychotherapy notes for 731,520 Veterans who initiated care for PTSD in the VA between 2004 and 2013. We added progressively stringent requirements (i.e. any sessions, 8 or more sessions, 8 or more sessions with the same therapist, 8 or more sessions with the same therapist within 14 weeks) to each strategy. We then compared symptomatic outcomes over the initial 8 sessions of treatment for patients meeting increasingly stringent adequacy requirements, among those with PTSD checklist (PCL) measurements proximal to the first and eighth treatment. We used inverse propensity of treatment weighting to balance differences in baseline characteristics that could plausibly be associated with the outcome of decrease in PCL score among patients meeting increasingly stringent treatment standards.

Results:
Receipt of any EBP as measured using NLP of psychotherapy notes among patients initiating VA care for PTSD increased from 0.7% in 2004-2005 to 14.1% in 2012-2013. When using administrative data only, outcomes were superior when patients met the most stringent standard for treatment adequacy. When using NLP data, a less stringent standard of 8 EBP sessions with the same therapist was required. In 2012-2013, 4.5% of patients met this standard during their initial year of treatment.

Implications:
Use of EBP for PTSD in an all-era cohort of VA users initiating care for PTSD increased dramatically from 2004 to 2013. Measures of treatment adequacy that rely on administrative data only require more stringent coding rules than measures that incorporate NLP of psychotherapy notes.

Impacts:
Use of EBP for PTSD has grown substantially in the VA. However, continued efforts to ensure patients receive an adequate course of treatment are indicated.