Post-traumatic stress disorder (PTSD) is a debilitating mental health disorder that may develop after exposure to traumatic events. Substance use disorder (SUD) is a behavioural disorder in which the use of one or more substances is associated with heightened levels of distress, clinically significant impairment of functioning, or both. PTSD and SUD frequently occur together. The comorbidity is widely recognised as being difficult to treat and is associated with poorer treatment completion and poorer outcomes than for either condition alone. Several psychological therapies have been developed to treat the comorbidity, however there is no consensus about which therapies are most effective.

To determine the efficacy of psychological therapies aimed at treating traumatic stress symptoms, substance misuse symptoms, or both in people with comorbid PTSD and SUD in comparison with control conditions (usual care, waiting-list conditions, and no treatment) and other psychological therapies.

We searched the Cochrane Depression, Anxiety and Neurosis Group’s Specialised Register (CCDANCTR) all years to 11 March 2015. This register contains relevant randomised controlled trials from the Cochrane Library (all years), MEDLINE (1950 to date), EMBASE (1974 to date), and PsycINFO (1967 to date). We also searched the World Health Organization International Clinical Trials Registry Platform and ClinicalTrials.gov, contacted experts, searched bibliographies of included studies, and performed citation searches of identified articles.

Randomised controlled trials of individual or group psychological therapies delivered to individuals with PTSD and comorbid substance use, compared with waiting-list conditions, usual care, or minimal intervention or to other psychological therapies.

We used standard methodological procedures expected by Cochrane.

We included 14 studies with 1506 participants, of which 13 studies were included in the quantitative synthesis. Most studies involved adult populations. Studies were conducted in a variety of settings. We performed four comparisons investigating the effects of psychological therapies with a trauma-focused component and non-trauma-focused interventions against treatment as usual/minimal intervention and other active psychological therapies. Comparisons were stratified for individual- or group-based therapies. All active interventions were based on cognitive behavioural therapy. Our main findings were as follows.

Individual-based psychological therapies with a trauma-focused component plus adjunctive SUD intervention was more effective than treatment as usual (TAU)/minimal intervention for PTSD severity post-treatment (standardised mean difference (SMD) -0.41; 95% confidence interval (CI) -0.72 to -0.10; 4 studies; n = 405; very low-quality evidence) and at 3 to 4 and 5 to 7 months’ follow-up. There was no evidence of an effect for level of drug/alcohol use post-treatment (SMD -0.13; 95% CI -0.41 to 0.15; 3 studies; n = 388; very low-quality evidence), but there was a small effect in favour of individual psychological therapy at 5 to 7 months (SMD -0.28; 95% CI -0.48 to -0.07; 3 studies; n = 388) when compared against TAU. Fewer participants completed trauma-focused therapy than TAU (risk ratio (RR) 0.78; 95% CI 0.64 to 0.96; 3 studies; n = 316; low-quality evidence).

Individual-based psychological therapy with a trauma-focused component did not perform better than psychological therapy for SUD only for PTSD severity (mean difference (MD) -3.91; 95% CI -19.16 to 11.34; 1 study; n = 46; low-quality evidence) or drug/alcohol use (MD -1.27; 95% CI -5.76 to 3.22; 1 study; n = 46; low-quality evidence). Findings were based on one small study. No effects were observed for rates of therapy completion (RR 1.00; 95% CI 0.74 to 1.36; 1 study; n = 62; low-quality evidence).

Non-trauma-focused psychological therapies did not perform better than TAU/minimal intervention for PTSD severity when delivered on an individual (SMD -0.22; 95% CI -0.83 to 0.39; 1 study; n = 44; low-quality evidence) or group basis (SMD -0.02; 95% CI -0.19 to 0.16; 4 studies; n = 513; low-quality evidence). There were no data on the effects on drug/alcohol use for individual therapy. There was no evidence of an effect on the level of drug/alcohol use for group-based therapy (SMD -0.03; 95% CI -0.37 to 0.31; 4 studies; n = 414; very low-quality evidence). A post-hoc analysis for full dose of a widely established group therapy called Seeking Safety showed reduced drug/alcohol use post-treatment (SMD -0.67; 95% CI -1.14 to -0.19; 2 studies; n = 111), but not at subsequent follow-ups. Data on the number of participants completing therapy were not for individual-based therapy. No effects were observed for rates of therapy completion for group-based therapy (RR 1.13; 95% CI 0.88 to 1.45; 2 studies; n = 217; low-quality evidence).

Non-trauma-focused psychological therapy did not perform better than psychological therapy for SUD only for PTSD severity (SMD -0.26; 95% CI -1.29 to 0.77; 2 studies; n = 128; very low-quality evidence) or drug/alcohol use (SMD 0.22; 95% CI -0.13 to 0.57; 2 studies; n = 128; low-quality evidence). No effects were observed for rates of therapy completion (RR 0.91; 95% CI 0.68 to 1.20; 2 studies; n = 128; very low-quality evidence).

Several studies reported on adverse events. There were no differences between rates of such events in any comparison. We rated several studies as being at ‘high’ or ‘unclear’ risk of bias in multiple domains, including for detection bias and attrition bias.

We assessed the evidence in this review as mostly low to very low quality. Evidence showed that individual trauma-focused psychological therapy delivered alongside SUD therapy did better than TAU/minimal intervention in reducing PTSD severity post-treatment and at long-term follow-up, but only reduced SUD at long-term follow-up. All effects were small, and follow-up periods were generally quite short. There was evidence that fewer participants receiving trauma-focused therapy completed treatment. There was very little evidence to support use of non-trauma-focused individual- or group-based integrated therapies. Individuals with more severe and complex presentations (e.g. serious mental illness, individuals with cognitive impairment, and suicidal individuals) were excluded from most studies in this review, and so the findings from this review are not generalisable to such individuals. Some studies suffered from significant methodological problems and some were underpowered, limiting the conclusions that can be drawn. Further research is needed in this area.