2019 HSR&D/QUERI National Conference

1126 — "She still doesn't know me." Describing the divergent experiences of Veterans who complete versus discontinue trauma-focused therapy for PTSD

Lead/Presenter: Shannon Kehle-Forbes,  COIN - Minneapolis
All Authors: Kehle-Forbes SM (Center for Care Delivery & Outcomes Research, Minneapolis), Ackland P (Center for Care Delivery & Outcomes Research, Minneapolis), Chard K (Cincinnati VA Healthcare System) Foa EB (University of Pennsylvania) Lyon A (Center for Care Delivery & Outcomes Research, Minneapolis) Meis L (Center for Care Delivery & Outcomes Research, Minneapolis) Orazem R (Center for Care Delivery & Outcomes Research, Minneapolis) Polusny M (Center for Care Delivery & Outcomes Research, Minneapolis) Schnurr P (National Center for PTSD) Spoont M (Center for Care Delivery & Outcomes Research, Minneapolis) Valenstein-Mah (Center for Care Delivery & Outcomes Research, Minneapolis) Zickmund S (Informatics, Decision-Enhancement and Analytic Sciences Center, Salt Lake City)

About one-third of Veterans who initiate prolonged exposure (PE) and cognitive processing therapy (CPT) do not complete. Quantitative studies of dropout have failed to identify consistent predictors, limiting the development of engagement interventions. The study objective was to understand reasons for premature dropout and factors that facilitate completion among Veterans who recently initiated a course of trauma-focused therapy.

We conducted semi-structured interviews with a national sample of Veterans who completed (n = 60) and prematurely discontinued (n = 68) PE and CPT. Our sample was stratified by service era (33% Vietnam), gender (34% women), and therapy modality and purposely sampled for race/ethnicity (25% African American; 14% Hispanic / Latino) and time of dropout. Interviews explored treatment response, experience with treatment elements, working alliance, beliefs about PTSD and treatment, social influences, and logistic barriers. We used a mixed deductive and inductive coding approach in which top-level codes were derived from the project's conceptual model and sub-codes were inductively derived within each top-level code. 20% of transcripts were double-coded. A modified constant comparative method was used to evaluate common and unique themes in each code between treatment dropouts and completers.

Commonalities shared by completers and dropouts included: non-specific support from therapists and loved ones, major and minor life stressors during treatment, perception of symptom exacerbation, and experience with the active treatment components. Completers were more likely to report flexible, patient-centered treatment delivery, therapy-specific support by providers, strategic use of other MH providers, and the perception that symptom exacerbation was part of the treatment process. Dropouts reported providers who were overly concerned with treatment protocol, feeling overwhelmed by completing life demands, anticipating a negative impact of treatment and/or symptom exacerbation on functioning, and interpreting perceived symptom exacerbation as proof that the treatment would not be effective.

There are several potentially-modifiable factors associated with PE/CPT dropout, including strength of the therapeutic relationship, expectations regarding symptom trajectory, and concerns regarding the impact of treatment on functioning.

Reducing rates of PE/CPT dropout will positively impact Veterans' health and well-being, lower the cost of treating PTSD, and decrease Veterans' long-term demand for PTSD services.