The review assessed psychological therapies involving family members to treat problem drinkers and concluded that such therapies can reduce alcohol-related outcomes. The authors' conclusions are likely to be reliable.

To evaluate the clinical and cost effectiveness of psychological therapies involving family and friends for people who misuse (or are dependent on) alcohol. This abstract is concerned only with clinical effectiveness.

The following databases were searched (with date ranges from 1966 to July 2006): MEDLINE, MEDLINE In-Process, Cochrane Database of Systematic Reviews, DARE, HTA, Cochrane Central Register of Controlled Trials, EMBASE, CINAHL, PsycINFO, Campbell Collaboration, ASSIA, IBSS, SCI-expanded, SSCI, National Research Register, metaRegister and ClinicalTrials.gov. There were no language restrictions. Reference lists of relevant papers, trial registries and the Internet were searched. Experts and relevant organisations were contacted.

Randomised controlled trials (RCTs) where more than half the participants were classed as problem drinkers, that evaluated any psychotherapeutic intervention to treat problem drinkers and which incorporated concerned significant others in some way were eligible for inclusion in the review. Studies had to assess alcohol-related outcomes or other generic outcomes (such as quality of life or hospitalisation).

Populations of the included studies were varied and mostly consisted of men or women with alcohol abuse problems. For most studies the focus of the therapy was on the problem drinker, but some focused on partners or concerned significant others. Intervention and comparator treatments were varied, but were often based on counselling or therapy sessions (typically family-based in intervention groups and non family-based in comparator groups). Abstinence rates, hospitalisation rates, relationship functioning and drinker treatment entry rates were the most common outcomes assessed.

Studies were first screened by one reviewer. A second reviewer checked a random sample. Disagreements were resolved by discussion. Full texts were selected for inclusion by one reviewer and checked by another; disagreements were resolved by discussion. Authors were contacted when there were insufficient details.

Study quality was assessed using a modified Jadad scale, which assessed randomisation, allocation concealment, blinding of outcome assessment and losses to follow up. One reviewer assessed study quality.

Numbers of outcomes and/or means (with standard deviations) in intervention and comparator groups were extracted. Odds ratios and/or standardised mean differences with 95% confidence intervals (CIs) were calculated. Comparators were categorised as other forms of family therapy, other forms of counselling or psychotherapy and other forms of care.

Data was extracted by one reviewer and checked by a second. Disagreements were resolved by discussion.

Meta-analyses examining either pooled odds ratios or standardised mean differences were performed using a fixed-effect model. A random-effects model was used where there was significant clinical heterogeneity. The studies were weighted, but the method of weighting was not stated. Statistical heterogeneity was assessed using the Χ² and I² statistic.

Thirty-four RCTs were included in the review (n=3,531). The sample sizes ranged from 12 to 742. Follow up periods ranged from four weeks to four years.

Study quality was varied, with a median score of 2 out of a possible 5. Losses to follow up were greater than 20% in nine RCTs.

Abstinence: Family therapy was no more effective than other forms of psychotherapy or counselling at increasing abstinence rates, odds ratio 1.17 (95% CI: 0.89 to 1.54, p=0.25; seven trials), but was more effective than other forms of care, odds ratio 8.59 (95% CI: 3.46 to 21.38, p<0.00001; three trials). Heterogeneity was only observed in the other forms of care analysis (I²=44.7%). Family therapy was also more effective than other forms of psychotherapy or counselling at increasing mean abstinent or percentage abstinent days, standardised mean difference 0.38 (95% CI: 0.21 to 0.56, p<0.0001, I²=22.6%; eight trials).

Hospitalisation rates: No statistically significant differences were found between any groups for hospitalisation rates.

Relationship functioning: For positive relationship functioning of drinkers, family therapy was more effective than other forms of psychotherapy or counselling, standardised mean difference 0.59 (95% CI: 0.34 to 0.83, p<0.00001; eight trials), with no statistical heterogeneity found. For positive relationship functioning of partners, family therapy was also more effective than other forms of psychotherapy or counselling, standardised mean difference 0.52 (95% CI: 0.07 to 0.97, p=0.02, I²=64.6%; six trials).

Drinker treatment entry rates: Family therapy directed at the partner or concerned significant other was found to be more effective than no counselling, waiting list and non-counselling interventions, odds ratio 5.65 (95% CI: 2.79 to 11.44, p<0.00001, I²=3.9%; five trials).

Further results were reported.

Therapy involving family and friends can reduce alcohol-related outcomes.

The review addressed a clear question and was supported by appropriate inclusion criteria. Attempts to identify all relevant studies in any language were undertaken by searching electronic databases and through other sources. Good methods were used to minimise the risks of reviewer error and bias for data extraction, but the processes used for study selection and quality assessment were less stringent and much less likely to reduce such risks. Study details were provided and study quality was adequately assessed, although individual quality scores were mostly not reported, making interpretation of results more difficult. Appropriate methods were used to pool results and assess statistical heterogeneity; the authors acknowledged the influence of clinical heterogeneity when pooling data. Despite the review having one or two flaws, the authors' overall conclusions reflect the evidence presented and are likely to be reliable.

Practice: The authors stated that implementing family therapy seemed likely to reduce the incidence of alcohol-related disease and in-patient hospital stays.

Research: The authors stated that it would be useful to investigate whether patient or family characteristics were indicators of success or failure of family therapy. Also, it would be useful to know if different forms of family therapy were equally effective.

West Midlands Health Technology Assessment

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.