Systematic reviews by hcj


									                                     Systematic reviews
                               Advice to authors on presentation

This document gives advice to those submitting reviews. Clinical Rehabilitation will generally
only publish systematic reviews.

A systematic review is one that has a specified method underlying the identification and
selection of the papers from the specified data sources that are being used. A systematic review
may or may not include a meta-analysis, which is the use of a statistical method to combine
data, usually summary data from several separate studies. A meta-analysis always depends
upon an initial systematic review, and so the methodology advised for meta-analytic studies is
also advised for other systematic reviews.

Authors are strongly recommended to read the QUOROM statement checklist for improving
the quality of reports of meta-analyses of randomised controlled studies (see and its following page,

The abstract should use the following headings:

Objective            State what the goal of the review was; what questions are being asked.
Data sources         Define the data-bases searched, and any secondary sources of data
Review methods       Summarise search and selection strategies, including patient population,
                     study type and design selected, how studies were selected, any quality
                     criteria used
Results              Give some results – actual data (e.g. number of studies and patients
                     included, percent showing some specified outcome)
Conclusion           This should relate back to your objective; the question should be answered
                     (or the fact that it cannot be answered made clear)

The paper itself should follow the usual structure (Introduction, Methods, Results, Discussion),
and will often have tables to present data. Some points are worth making.

It is helpful to include, as an appendix, a copy of the computer data-base search strategy for one

The discussion should not only discuss the limitations of the studies included, but should also
discuss the weaknesses and limitations of the review itself.

If a meta-analysis is undertaken, then presentation of the results using figures such as produced
by the Cochrane Collaboration’s Review Manager (RevMan) are useful
(see )

Derick Wade
July 7th 2006

To top