Telephone Case Monitoring (TCM) for Veterans With Post-Traumatic Stress Disorder (PTSD)

Background: Poor compliance with aftercare may contribute to high rates of relapse and rehospitalization among Veterans who received residential treatment for posttraumatic stress disorder (PTSD). Telephone case monitoring has been shown to improve treatment adherence and reduced relapse among patients with chronic medical and substance use disorders, but has not been tested in PTSD patients.

Objectives: This multisite randomized controlled trial tested whether augmenting usual aftercare with telephone monitoring improved resulted in 1) improved clinical outcomes (less violence, substance use, and PTSD symptoms; 2) longer time to rehospitalization; 3) better compliance with aftercare in the year after discharge from residential treatment for PTSD.

Methods: This trial recruited 837 subjects from 6 PTSD residential treatment programs at 5 VA medical centers, 94.7% of the 884 projected. Patients who completed at least 14 days of residential PTSD treatment and discharged to VA outpatient care were eligible to participate. Subjects were randomized to usual aftercare care (n = 425) or usual aftercare plus biweekly telephone case monitoring calls during the first three months after discharge (n = 412). Telephone case monitors assessed current problems, encouraged treatment adherence, provided problem-solving support, and alerted providers to emergent care needs. Patient self-report measures of psychiatric symptoms, substance use, and violence were obtained at intake to residential treatment and 4 months (69% completion rate) and 12 months (64% completion rate) after discharge. Retention was lower than the investigators' planned 70% to 75% rate due to difficulty locating some patients who moved (even their collateral informants did not know where they were) and 45 participants asking to discontinue due to lack of time (n = 10), general dissatisfaction with VA (n = 6), distress during phone calls (n = 5), dissatisfaction with compensation (n = 1), or no specified reason (n = 24). Treatment utilization data was obtained from the VA National Patient Care Database.

Intent-to-treat analyses used mixed modeling to compare clinical outcomes in the telephone monitoring and usual care groups and 4 and 12 months after discharge. Survival analysis was used to compare conditions on time to rehospitalization. Having a slightly smaller-than-intended sample size resulted in modest reductions in statistical power, e.g., power to detect the expected d = .25 effect on PTSD outcomes was reduced from about 90% to 82%, and power to detect the anticipated W = .105 difference in rehospitalization rates was reduced from 88% to 85%. Secondary analyses assessed whether differences in outcomes between the telephone case monitoring and usual care groups were mediated by attending more outpatient visits and completing more medication refills. Exploratory analyses examined whether the effect of telephone support on the clinical outcome measures, number of treatment visits, and medication refills was moderated by number of outpatient mental health visits in the prior year, distance from clinic, treatment expectancies, therapeutic alliance, or co-occurring substance use problems.

Status: Enrollment, intervention, data collection, and primary analyses are completed. Primary results have been published in Psychiatric Services (Rosen, Tiet, Harris et al., 2013) and two secondary papers have been published in the Journal of Traumatic Stress (Belsher, Tiet, Garvert, \& Rosen, 2012; Rosen, Adler, \& Tiet, 2013).

A CDMRP-funded study extending this approach to PTSD outpatients at the Durham, Puget Sound and Palo Alto VA medical centers has recently been completed. Initial results of that second trial suggest that telephone care management improved treatment attendance but had weak effects on outcomes.