Treating Depression and Adherence (CBT-AD) in Patients with HIV in Care: A Three-arm Randomized Controlled Trial

BACKGROUND: Depression, highly prevalent in HIV, is consistently associated with worse ART adherence. Integrating CBT for depression with adherence counseling using the “Life-Steps” approach (CBT-AD) has an emerging evidence base. The aim of the current study was to test the efficacy of CBT-AD. METHODS: We conducted a three-arm RCT (N=240 HIV-positive adults with depression), comparing CBT-AD to Life-Steps integrated with information and supportive psychotherapy (ISP-AD) (both 12 sessions), and to ETAU (1 session Life-Steps). Participants were recruited from three sites in New England area, two being hospital settings, and one being a community health center. Randomization was done via a 2:2:1 ratio, using random allocation software by the data manager, in pairs, stratified by three variables: site, whether or not the participant was prescribed antidepressant medications, and history of injection drug use. The primary outcome was adherence assessed via electronic pill caps (MEMs) with correction for “pocketed” doses. Secondary outcomes included depression, plasma HIV RNA and CD4. Follow-ups occurred at 4, 8 and 12 months. We used intent-to treat analyses with ANCOVA for independent-assessor pre-post assessments of depression and mixed effects modeling for longitudinal assessments. Clinical Trial Registration: NCT00951028, https://clinicaltrials.gov/ct2/show/NCT00951028), closed to new participants. FINDINGS: The period of recruitment was February 26, 2009 to June 21, 2012, with the 12-month follow-up period extending until April 29, 2013. There were no study-related adverse events. CBT-AD (n=94 randomized, 83 retained) had greater improvements in adherence (Est.=1·00, CI=0·34, 1·66, p=0·003) and depression (CES-D Est.=−0·41, CI=−0·66, −0·16, p=0·001; MADRS Est.=−4·69, CI=−8·09, −1·28, p=0·007; CGI Est.=−0·66, CI=−1·11,-0·21, p=0·005) than ETAU (49 randomized, 46 retained) at post-treatment (4-month). Over follow-ups, CBT-AD (84 retained) maintained higher adherence (Est.=8·93, CI=1·90, 15·97, p=0·013) and lower depression on the CES-D (Est=−3·56, CI=−6·08, −1·05, p=·005) and CGI (Est.=−0·39, CI=−0·77, −0·18, p=·04) than ETAU (86 retained); however, not for the MADRS. There were no significant differences between CBT-AD and ISP-AD (97 randomized, 87 retained) for the post-treatment or follow-up (86 retained) analyses. There were no intervention effects on HIV RNA or CD4, though a higher percentage (91·4%) than expected was virally suppressed at baseline. INTERPRETATION: Integrating evidenced-based treatment for depression with evidenced-based adherence counseling is helpful for individuals living with HIV/AIDS and depression. Future efforts should examine how to best disseminate of effective psychosocial depression treatments such as CBT-AD to individuals living with HIV/AIDS, as well as examine the cost-effectiveness of such approaches. FUNDING: National Institute of Mental Health (R01MH084757) and some author time from NIAID 5P30AI060354, and P30AI042853.