Foa 2013.

Methods	Design: RCT
Participants	Setting: Treatment‐seeking individuals were recruited through advertisements and professional referrals and treated on an outpatient basis. Inclusion criteria: Current PTSD and alcohol dependence according to DSM‐IV; clinically significant trauma‐related symptoms, as indicated by a score of at least 15 on the PSS‐I; and heavy drinking in the past 30 days, defined as an average of more than 12 standard alcohol drinks per week with at least 1 day of 4 or more drinks determined by the TFBI. Exclusion criteria: Current substance dependence other than nicotine or cannabis; current psychotic disorder (e.g. schizophrenia, bipolar disorder); clinically significant suicidal or homicidal ideation; opiate use in the month prior to study entry; medical illnesses that could interfere with treatment (e.g. AIDS, active hepatitis); or pregnancy or nursing. Sample size: 657 individuals were assessed for eligibility; 165 were randomised, and all were included in the analyses. PTSD diagnosis: All participants met full diagnosis for PTSD as measured by the CAPS. SUD type and diagnosis: All participants met full diagnosis for alcohol dependence. Mean age: prolonged exposure + naltrexone 40.1 (95% CI 36.7 to 43.5); prolonged exposure + placebo 44.7 (95% CI 41.8 to 47.7); supportive counselling + naltrexone 44.9 (95% CI 41.8 to 47.9); supportive counselling + placebo 41.2 (95% CI 38.6 to 43.9) Gender: 108 (65.5%) male; 57 (34.5%) female Ethnicity: 63% African American; 30% white; 4.2% Hispanic; 0.6% other Country: USA
Interventions	Before randomisation to treatment, participants completed outpatient medical detoxification (at least 3 consecutive days of alcohol abstinence) with oxazepam as required to manage alcohol withdrawal symptoms. Group 1: Prolonged exposure + naltrexone + supportive counselling: n = 40. Prolonged exposure therapy consisted of 12 weekly 90‐minute sessions followed by 6 biweekly sessions and included repeated imaginal exposure (i.e. revisiting and recounting traumatic memories) and processing the memory (i.e. discussing thoughts and feelings related to revisiting the memory). The target dose of naltrexone was 100 mg/d, starting with 50 mg/d for a minimum of 3 days and titrating up within 1 week. Supportive counselling was available as described below. Group 2: Prolonged exposure + placebo + supportive counselling: n = 40. Participants received PE as described above. Supportive counselling was available as described below. Group 3: Supportive counselling + naltrexone: n = 42. Supportive counselling was based on the BRENDA model, which combines medication management with compliance enhancement techniques based on motivational interviewing. Supportive counselling sessions were administered by a study nurse and lasted 30 to 45 minutes. Input included dispensing medication, monitoring compliance, assessing and providing education about alcoholism, and offering support and advice concerning drinking. Visits were weekly during the first 3 months and biweekly during the remaining 3 months. Group 4: Supportive counselling + placebo: n = 43. Supportive counselling was as described above. Experimental intervention modality: Combined
Outcomes	PTSD: PSS‐I SUD: TLFB; PACS Treatment acceptability: Reported in terms of the mean number of sessions attended for PE. Other: ‐ Follow‐up: End of treatment and 6 months' post‐treatment
Notes	For the purpose of the review, we were interested in the comparison between prolonged exposure plus supportive counselling and supportive counselling
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants and personnel were not be blinded to allocation
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Evaluators were blind to treatment group assignment
Incomplete outcome data (attrition bias) All outcomes	Low risk	Treatment dropouts and withdrawals were clearly reported. An ITT approach was employed using hierarchical linear and non‐linear modelling, which took into account dropouts and missing data
Selective reporting (reporting bias)	Low risk	Key outcomes are as specified in the protocol registered with ClinicalTrials.gov
Other bias	Low risk	There was no significant difference between the treatment groups on any demographic or baseline diagnostic characteristics. We identified no other potential biases