This review evaluated the effectiveness of primary care psychosocial treatments for depression in older adults. The authors concluded that evidence for symptom reduction favours a fully interdisciplinary care approach. Methodological limitations in the review process, together with the inclusion of some inadequately reported and variable studies, mean that the reliability of the authors' conclusions is unclear.

To evaluate the effectiveness of primary care-based psychosocial treatments for depression in older adults.

MEDLINE and PsycINFO were searched for studies published in English from January 1994 to April 2004; the search terms were reported. Reference lists were screened for additional relevant articles.

Randomised controlled trials (RCTs) of primary care-based psychosocial treatments for depression, compared with a control or usual care, in patients aged 55 years and older were eligible for inclusion. Studies of patients with bipolar or other psychiatric disorders were excluded. The included patents had a mean age of 72.7 years, and there was substantial variation in the proportion of males (14 to 100%), Caucasians (60 to 95%) and severity of depression. Amongst the included interventions were various systems of care, direct interventions or psychotherapy, telephone care, and psychoeducation delivered by nurses, social workers, psychologists, counsellors and physicians. Many patients were also in receipt of antidepressant medication. Usual care comprised the physician as the primary provider, offering treatment as appropriate. Eligible outcome measures were self-report or clinician-ratings of symptom reduction. Various depression rating scales and symptom checklists were used.

One reviewer initially selected potentially relevant studies, following which two reviewers decided on final inclusion in the review. Any disagreements were resolved by discussion, or through a third party if necessary.

The quality of the studies was assessed using a nine-dimension published scale to produce a total quality score (maximum 45) derived from the average of the two raters' scores. Studies scoring 30 or above were included in the review. The measurement items were significance of the problem, clarity of problem definition, appropriateness of design measure, adequacy of control variables, sample selection procedure, validity and reliability of measures, validity of data analysis, appropriateness of interpretations, and generalisations and adequacy of research report. Drop-out rates were also reported.

Two independent reviewers assessed study quality. Any disagreements were resolved by discussion.

To calculate the effect size, the percentage symptom reduction was calculated for each included study using the change in mean score from baseline to each assessment point. Where mean values were not available, the reported or assumed p-values were used.

One reviewer extracted the data.

A narrative synthesis of the results was reported, owing to variations in interventions, providers and populations amongst studies. The included studies were grouped and differences were explored according to two intervention models: Geriatric Evaluation Management (using an interdisciplinary approach with emphasis on communication between physicians, nurse practitioners, social workers, psychologists and clinical pharmacists), and Integrated Health Care (using a multidisciplinary approach with more emphasis on independent working and minimum communication between team members).

Eight RCTs (3,864 patients) were included in the review.

Study quality scores ranged from 32 to 44 (mean 38). Sample sizes ranged from 96 to 1801. Follow-up ranged from 11 weeks to 24 months, and drop-out rates were lower for Geriatric Evaluation Management studies.

Four studies (954 patients) used the Geriatric Evaluation Management model. Two of these reported on short- and long-term follow-up from the same trial, revealing a medium to large effect size for symptom reduction at 12 months and a smaller one at 24 months: 0.59 and 0.20, respectively. One other study showed symptom reduction at 18 months, but the estimated effect size (0.20) was small. The fourth study showed no difference in effect between intervention and control groups.

Four studies (2,910 patients) used the Integrated Health Care model. Three of these reported symptom reduction in the intervention groups, although effect sizes were classed as medium in only one (0.47 at 3 months and 0.57 at 12 months). One study also reported decreased suicidal ideation (result not reported).

There is some evidence in favour of both Geriatric Evaluation Management and Integrated Health Care models for the psychosocial treatment of depression. The more robust evidence appears to be linked to interventions using an interdisciplinary approach.

This review addressed a clear research question and was supported by detailed inclusion criteria. The search strategy relied largely on two electronic databases, and the restriction to English language publications means that language and publication biases cannot be ruled out; the authors acknowledged these limitations. Although there were some attempts to minimise bias and error in the review process, not all stages were conducted with complete transparency. Detailed validity assessment criteria were reported, and the review included some methodological items of relevance to the evaluation of RCTs. Attempts were made to capture the wide variety of included study characteristics, but inadequate reporting within the primary studies presents limitations to the reliability and generalisability of the results. The method of synthesis was appropriate in view of the wide variation between the included studies. The extent to which the authors' conclusions are reliable is unclear given some of the methodological concerns highlighted.

Practice: The authors did not state any implications for practice.

Research: The authors stated that truly interdisciplinary psychosocial therapies provided by mental health providers as first-line treatments need more testing. Future research should also aim to provide clearer descriptive data to assist in the replicability of interventions, and concentrate on goals linked to potential service use.

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.