This review concluded that the limited available evidence suggests that exercise training is beneficial to patients with Parkinson's disease, especially those in the early stages of the disease. Given the small numbers of patients and lack of a 'no exercise' control group in most studies, the authors' conclusions may be over-optimistic.

To evaluate the effects of physical exercise on mortality, strength, balance, mobility and activities of daily living in patients with Parkinson's disease (PD).

The authors searched the Cochrane Database of Systematic Reviews, CINAHL, PubMed and MEDLINE. The search terms included 'Parkinson's disease', 'exercise' and 'physical exercise'. Search dates were not reported. Reference lists were screened. Only studies published in the English language were included.

Study designs of evaluations included in the review

No inclusion criteria for study designs were specified .

Specific interventions included in the review

Studies of any form of physical exercise were eligible for the review. Studies involving drug treatment in combination with exercise were excluded. The interventions in the included studies included balance and resistance training, body weight-supported treadmill training, individual physical exercise, pole striding, physiotherapy and standardised sport activities. The duration of the intervention, where reported, ranged from 1 month to 10 weeks. In most studies the comparator was another form of physical exercise or physiotherapy. Only one study had a no intervention control group. One included study appeared to be a cohort study that compared groups with different levels of self-reported exercise, but it was unclear whether a specific exercise intervention was evaluated.

Participants included in the review

Eligible studies had to include only patients with PD. Most of the studies included patients at stages 2 or 3 on the Hoehn-Yahr scale (mild to middle stages of PD).

Outcomes assessed in the review

Studies had to assess patients before and after the intervention but, otherwise, no inclusion criteria for the outcomes were specified. The included studies measured a range of outcomes, the Unified PD Rating Scale being the most frequently used scale. One study assessed mortality. Some studies assessed patient-assessed outcomes.

How were decisions on the relevance of primary studies made?

The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.

The quality of the description of the interventions was used to assess study quality. Some other aspects of validity were mentioned in the text and tables. The authors did not state how the validity assessment was performed.

The authors did not state how the data were extracted for the review, or how many reviewers performed the data extraction.

How were the studies combined?

The studies were combined in a narrative.

How were differences between studies investigated?

Differences between the studies were discussed in the text.

Seven studies (n=578) were included: three randomised controlled trials (n=85), one crossover trial (n=33), one cohort study (n=438), one before-and-after study (n=6) and one open uncontrolled study (n=16).

Methodological flaws included problems with selection of the patients, non-standardised treatments, use of non-standardised measures to assess the outcomes and small sample sizes.

All of the included studies reported improvements in one or more outcomes following the intervention. The cohort study reported that non-exercising patients had a higher risk of mortality compared with the exercising group (hazard ratio 1.83).

The evidence, although limited, suggests that exercise training is beneficial to patients with PD, especially those in the early stages of the disease.

The review addressed a clear research question, although no inclusion criteria were specified for study designs and outcomes. The search was rather limited in terms of the databases and search terms used and unpublished studies were excluded, so it is possible that relevant studies could have been missed. Only studies published in English were included, so the review may be at risk of language bias. The risk of errors and bias during the study selection and data extraction processes cannot be assessed because the methods used were not reported. Validity was not formally assessed, but the authors discussed some relevant issues in their synthesis.

Some relevant details of the included studies were presented. A narrative synthesis was appropriate in view of the heterogeneity of the interventions and outcomes in the review. The largest study in the review examined the relationship between exercise and risk of mortality, but it was not clear whether a specific exercise intervention was evaluated. Given the small numbers of patients and varying quality of the remaining evidence, and the lack of 'no exercise' control groups in most studies, the authors' conclusions may be over-optimistic.

Practice: The authors stated that doctors should encourage patients with PD to join a physical exercise programme.

Research: The authors stated that future studies should aim to develop standardised exercise programmes for patients with PD and appropriate methods for evaluating outcomes. They also stated that studies should include longer term follow-up (over 1 year).

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.