This review concluded that bibliotherapy and cognitive-behavioural therapy-based websites and computer programs might be effective for general practitioners, in primary care, who have minimal psychotherapy training, for the management of adults with depression, but further research was needed. Given the limited synthesis and reporting, the author's conclusions should be interpreted with caution. The recommendation for further research was appropriate.

To assess the effectiveness of brief non-pharmacologic interventions by primary care physicians, with minimal training in psychotherapy, to manage depression in adult patients.

MEDLINE (from 1996), EMBASE (from 1980), and EBM Reviews (from 1999) were searched to January 2007, for articles published in English. Studies published before 1980 were excluded. Search terms were reported and reference lists were screened.

Randomised controlled trials (RCTs) of interventions that could be used in the primary care setting by physicians with little (less than three days) or no training in psychotherapy and without support from other health care professionals, to manage patients with a diagnosis of depression, were eligible for inclusion. Eligible trials were required to diagnose depression, using recognised criteria or a depression rating scale, and to measure clinically relevant outcomes (i.e. improvement of symptoms on a validated depression scale). Trials of patients with postpartum depression were excluded, as were those of patients in a geriatric setting or patients suffering depression due to a medical condition.

Included trials were mainly of women and their average age ranged from 37 to 50 years. Some patients were reported to be taking medication, but no further details were provided. Where reported, the participants were diagnosed with depression, major depression, or anxiety and a wide range of validated depression and mood disorder scales were used to define the inclusion criteria for most of the trials. The majority of RCTs compared cognitive-behavioural therapy (CBT) with a control. CBT could be accessed via books, computer programs, or interactive websites, with or without reminders or feedback from health care professionals or trial personnel. Controls were no formal control group, waiting list, traditional short-term individual therapy, usual care, or discussions. Interventions lasted from four to 32 weeks and outcomes were assessed using various measures.

The author did not state how many reviewers performed the trial selection.

The quality of the trials was assessed using criteria from the Evidence-Based Medicine Working Group and these included items on: randomisation, blinding, baseline comparisons, equal treatment of groups except for the intervention, follow-up, and intention-to-treat analyses.

The author did not state how many reviewers performed the validity assessment.

The author did not state how the data were extracted for the review.

Data were presented as a narrative synthesis and in tables.

Nine RCTs (n=4,575 patients plus 84 general practitioners, GPs) were included in the review. Samples sizes ranged from 80 to 2,794 patients. Six RCTs reported that groups were comparable at baseline, two indicated some differences, and one did not compare them at baseline. All trials treated groups equally and none of them included blinding to outcomes. Seven trials included intention-to-treat analyses and two accounted for all participants at the end of the trial (completion rates ranged between 20% and 91%). Where participants were followed up, the duration ranged from three to six months.

Six RCTs reported that a brief intervention with minimal therapist or physician contact resulted in significant positive outcomes for depression (p<0.05). The most effective interventions included bibliotherapy (books on CBT), websites on CBT, and computer programs on CBT. One trial showed that extended CBT significantly reduced depressive symptoms compared with brief CBT, problem-solving techniques, and behavioural strategies. The two remaining trials showed no effect of a CBT-based website, with no contact from trial personnel, and an educational package on CBT given to GPs. The greatest adherence to the interventions was seen in those with more structure, shorter intervention periods, and frequent contact or reminders from trial personnel.

It appears that this review looked at effects within trial groups rather than between groups, but this was not clear.

Bibliotherapy, CBT-based websites, and CBT-based computer programs might be effective tools for GPs in primary care and with minimal training in psychotherapy, for the management of adults with depression. Their effectiveness might be increased through patient contact with health care personnel, but further research was needed.

The review question was clear and it was supported by appropriate inclusion criteria. The literature search included three electronic databases, but it was limited to published articles in English, which suggests that relevant papers may have been missed. The author used published criteria to assess the quality of the included trials, but they did not include criteria such as the method of randomisation or concealment of treatment allocation. The author did not state how many reviewers performed the trial selection, data extraction, and validity assessment, which means that reviewer error and bias cannot be ruled out.

Given the differences between methods of recruitment, study inclusion and exclusion criteria, interventions, and measurement tools used, a narrative synthesis was appropriate. This synthesis was limited because of a lack of numerical or statistical data and a general lack of detail across trials. The analysis appeared to be based on within-group rather than between-groups comparisons. The author acknowledged some limitations of the included studies, such as their small samples and the generalisability of their populations. Further limitations included large drop-out rates in some trials, and short durations and follow-up periods.

The limitations in validity assessment and data synthesis, along with the lack of reporting of the review process, suggest that the author's conclusions should be interpreted with caution. The recommendation for further research was appropriate.

Practice: The author stated that health care providers must use their discretion to identify patients who might not benefit from these types of intervention, such as those in crisis, those with severe depression, or those unwilling or unable to think through their feelings and problems.

Research: The author stated that further research was needed to include a wider range of interventions (such as those recommended and mediated by GPs), using larger samples, and comparing the interventions with each other.

Not stated.

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.