This review assessed the effectiveness of teaching problem-solving skills to caregivers of stroke patients. The authors found weak evidence for a positive effect of problem-solving interventions versus usual care. Several limitations, including poorly reported inclusion criteria, the possibility of language and publication bias and the lack of consideration of the validity assessment, render the authors' conclusions unreliable.

To review the effectiveness of teaching problem-solving skills to caregivers of stroke patients.

CINAHL, MEDLINE, PsycLIT, and the Cochrane Library were searched form 1970 to 2004; the search terms used were listed. The bibliographies of identified studies were checked. The search was limited to articles reported in the English language.

Study designs of evaluations included in the review

No inclusion criteria for eligible study designs were stated.

Specific interventions included in the review

Studies that examined the problem-solving skills of family caregivers of stroke patients, or evaluated educational interventions for family caregivers' problem-solving skills, were eligible for inclusion. Specific problem-solving interventions included positive problem orientation, confronting problems, analysing problems, undertaking solutions and goal setting. The interventions were delivered through one or more of the following: individual home visits, telephone contacts, group support and family counselling. Staff delivering the interventions included nurses, social workers and mental health workers. The interventions were mostly applied early after the stroke. The duration of the intervention ranged from 2 to 12 months. Where specified, the control intervention was usual care.

Participants included in the review

Studies that included caregivers of stroke patients were eligible for inclusion; studies of patients only were not eligible. The included studies were based on caregivers alone, caregiver plus patient, or caregiver, patient and friends. One of the included studies, in which the intervention was given to both the patients and caregivers, only reported outcomes for the patients. The stroke patients in the included studies were mainly men aged 61 to 94 years. Where reported, the majority of caregivers were female and the mean age ranged from 48 to 64 years.

Outcomes assessed in the review

No inclusion criteria for the outcomes of interest were stated. The studies in the review assessed a range of outcomes.

For the caregiver, these included stroke knowledge, family functioning, confidence in knowledge, stress, burden, depression, problem-solving abilities, perceived health, preparedness, satisfaction with services, social support, assertiveness, coping strategies and use of social services.

For the patient, these included adjustment, social activities, perceived health, functional ability, psychological condition, behavioural condition, social support, depression and sense of control.

Follow-up was for up to 12 months after the stroke.

How were decisions on the relevance of primary studies made?

One author identified relevant articles, but the authors did not state how the papers were selected for the review.

One author performed the validity assessment, using the Consolidated Standards of Reporting Trials (CONSORT) statement checklist for assessing the reporting of randomised controlled trials (RCTs) and a checklist of quality items for qualitative research designs. Other authors reviewed the decisions and any disagreements were resolved by discussion.

One author extracted the data, but the authors did not state how the data were extracted for the review. Data on the outcome measures, timing of data collection, and results reported by the included studies were extracted.

How were the studies combined?

The RCTs were combined in a narrative and the other studies were described separately.

How were differences between studies investigated?

No between-study differences were investigated.

Eleven studies were included in the review; six RCTs (n>1,010; it was unclear how many were caregivers as some RCTs only reported the number of caregivers and patients combined, and one RCT did not report the sample size for caregivers), one correlational study (40 caregivers), two qualitative studies (30 patients, 15 caregivers, 5 specialist nurses), one case study (number of participants not reported) and a descriptive study (no participants).

Caregivers' problem-solving abilities: three RCTs found that teaching problem-solving skills improved family problem-solving, positive problem orientation and knowledge about stroke care at 6 to 12 months post-discharge in comparison with usual care.

Caregivers' health: two RCTs found that teaching problem-solving skills improved caregivers' rates of depression, preparedness, vitality and coping with the patients' emotions. However, the intervention did not improve rates of physical health, stress or burden (two RCTs).

Patients' health: two RCTs found that teaching problem-solving skills did not improve the functional recovery of stroke patients.

Two RCTs found that teaching problem-solving by telephone contact was more beneficial than home visits. One RCT found no difference in outcomes when comparing group support with individual home visits.

Findings from the other study designs were also briefly described in the review.

There is weak evidence for a positive effect of problem-solving interventions for caregivers of stroke patients, including a short-term reduction of depression.

The review question and inclusion criteria were poorly described. The search strategy was limited to articles published in English, and the authors acknowledged that the possibility of omitting relevant papers published in other languages, or unpublished articles, represents a potential source of bias. The interventions included in the review were so varied that the usefulness of a synthesis of the results is questionable. The checklist used to assess the quality of the RCTs was a checklist for assessing the quality of the reporting of RCTs, rather than their conduct. Some study designs were neither RCTs nor qualitative studies, therefore, neither of the checklists used were relevant for these studies. The results of the validity assessment were neither reported nor considered in the narrative synthesis, so the reliability of the reported results are unknown. Given these limitations, the authors' conclusions are unlikely to be reliable.

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

Research: The authors stated that sufficiently powered, well-controlled studies using standardised outcome measures are required to further investigate interventions that may increase problem-solving abilities and reduce depression. There is also a need to study the link between the theoretical underpinning of interventions and the outcomes measured. Studies should assess the process of intervention delivery, identify individuals most likely to benefit, assess different populations, and assess long-term outcomes.

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.