The review concluded that a brief primary care intervention for long-term users of benzodiazepine was an effective strategy to decrease or stop medication without adverse consequences. The authors conclusions mostly reflect the evidence presented but they did not review adverse consequences. The small amount of evidence and its variable quality means the conclusions should be interpreted with some caution.

To evaluate the effectiveness of minimal interventions to reduce the long-term use of benzodiazepine in adults in primary care.

Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE and EMBASE were searched to August 2010 for relevant articles. Search terms were reported. There were no language restrictions. Reference lists of identified articles were scanned.

Randomised controlled trials (RCTs) of minimal interventions in a primary care setting for withdrawal from long-term use of benzodiazepine in adults (18 years or older) were eligible for inclusion. Minimal interventions were defined as a letter, self-help information or short consultation with a general practitioner regarding long-term use, adverse effects and advice on withdrawal. Long-term use was defined as three months or more. The primary outcome of interest was withdrawal from use of benzodiazepine. A secondary outcome was general health status.

All the included studies were conducted in UK general practice. Follow-up was reported as six months. Benzodiazepine dosage was expressed as diazepam equivalents of either 5mg or 10mg. Interventions included consultation with general practitioner and self-help booklet, single letter and/or information sheets. Some studies had more than one intervention group. Comparison groups were allowed to continue with their usual dose of benzodiazepine. Mean age was 60 years. More than 60% of the participants were women. Benzodiazepine usage was measured by prescription records and general health status was measured by general health questionnaire (GHQ) or short-form 36 (SF-36).

Two reviewers independently selected studies for inclusion. Disagreements were resolved through discussion with a third reviewer.

Quality was assessed with criteria of randomisation, allocation concealment and use of intention-to-treat analysis.

Two reviewers independently assessed study quality. Disagreements were resolved through discussion with a third reviewer.

Data were extracted and used to calculate risk ratios (RRs) and 95% confidence intervals (CIs) for outcomes. Study authors were contacted for additional data where necessary.

Two reviewers independently extracted data. Disagreements were resolved through discussion with a third reviewer.

Pooled risk ratios and 95% CIs were calculated using a random-effects model. The number needed to treat was calculated. Statistical heterogeneity was assessed using the Ι² statistic.

Three RCTs (615 participants) were included in the review. All three RCTs were reported as using adequate randomisation although one study used date of birth as randomisation. Baseline similarity was clearly reported in one RCT. One RCT reported that groups were treated equally. All the studies were unblinded but used objective prescription records. Losses to follow-up ranged from 3.7% to 5.8% for benzodiazepine reduction and from 15% to 30% for general health status.

There was a significant decrease in benzodiazepine consumption for minimal intervention groups compared to usual care (RR 2.04, 95% CI 1.48 to 2.83; five comparisons). There was a significant increase in cessation for the intervention groups (RR 2.31, 95% CI 1.29 to 4.17; five comparisons). There was no evidence of statistical heterogeneity for either of these analyses (Ι²=0). The number needed to treat was 12.

There was a significant proportional reduction in consumption of benzodiazepine from baseline to six months (two RCTs) for intervention groups compared to control groups.

There were significant improvements in general health status for the minimal intervention groups compared to control groups (two RCTs).

A brief intervention in the form of either a letter or a single consultation by general practitioners for long-term users of benzodiazepine was an effective and efficient strategy to decrease or stop their medication without causing adverse consequences.

The review question and inclusion criteria were reported clearly. Several relevant sources were searched without language restrictions, which reduced potential for language bias. No efforts were made to locate unpublished studies so some data may have been missed. Study quality was assessed and results were reported. Appropriate methods we used to reduce risks of reviewer error and bias. There was no evidence of statistical heterogeneity and combining the data in a meta-analysis appeared appropriate. The quality of the included studies was generally poor and the authors reported that a fourth study was excluded due to poor quality.

The authors conclusions mostly reflect the evidence presented but they did not review adverse consequences. Due to the small amount of evidence of variable quality these conclusions should be interpreted with a degree of caution.

Practice: The authors stated that only a modest proportion of patients would reduce or cease benzodiazepine use but the minimal effort required suggested a high benefit-to-effort ratio.

Research: The authors stated that further research was needed to find methods to enhance the impact of minimal interventions to reduce long-term use of benzodiazepines in primary care.

One author was supported by a fellowship from the National Health and Medical Research Council, Australia.

Erratum in: British Journal of General Practice 2013; 63(606): 12.

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.