Difficulty defining comparisons in Cochrane systematic reviews of rehabilitation interventions

Article type
Authors
Coupar F, Pollock A, Legg L, Langhorne P
Abstract
Background: Cochrane systematic review methods can successfully compare the effectiveness of one clearly defined intervention (e.g. a drug) with a placebo or control intervention. However trials of rehabilitation interventions often do not fit into the simple comparison model of 'intervention vs. control'. Registration of a title 'Simultaneous bilateral training for improving arm function after stroke' was originally rejected by the Cochrane Stroke Group. One of the key reasons for this rejection was that "there are too many comparisons, and the comparisons presented are confusing". The number and nature of the comparisons in relation to studies of effectiveness of rehabilitation interventions will have to be dealt with within Cochrane systematic reviews to allow for appropriate evaluation of complex interventions.

Objectives: To explore the comparisons carried out within trials included in a systematic review of a complex rehabilitation intervention. This will be done using an ongoing Cochrane systematic review of 'Simultaneous bilateral training for improving arm function after stroke' as an example.

Methods: Following the application of the inclusion/exclusion criteria, two independent reviewers will document and categorise the comparison groups as:

- Simultaneous bilateral training vs. no treatment;

- Simultaneous bilateral training vs. usual therapy;

- Simultaneous bilateral training vs. placebo intervention;

- Simultaneous bilateral training vs. other upper limb intervention;

- Simultaneous bilateral training plus usual therapy vs. usual therapy;

- Other comparison.


Results: Results will include total number of trials, and number of trials within each of the comparison categories. Descriptions of the study interventions will be documented and the clinical relevance, and methodological implications, of combining some of the comparison groups (e.g. into simultaneous bilateral training vs. control) will be discussed.

Conclusions: The implications for future Cochrane systematic reviews and randomized controlled trials of complex rehabilitation interventions will be highlighted.