Cochrane reviews used more rigorous methods than non-Cochrane reviews: survey of systematic reviews in physiotherapy

J Clin Epidemiol. 2009 Oct;62(10):1021-30. doi: 10.1016/j.jclinepi.2008.09.018. Epub 2009 Mar 17.


Objective: To describe the quality and methods of systematic reviews of physiotherapy interventions, compare Cochrane and non-Cochrane reviews, and establish the interrater reliability of the Overview Quality Assessment Questionnaire (OQAQ) quality assessment tool.

Study design and setting: A survey of 200 published systematic reviews was done. Two independent raters assessed the search strategy, assessment of trial quality, outcomes, pooling, conclusions, and overall quality (OQAQ). The study was carried out in the University research center.

Results: In these reviews, the five most common databases searched were MEDLINE, EMBASE, Cochrane Library, CINAHL, and Cochrane Review Group Registers. The Cochrane allocation concealment system and Jadad Scale were most frequently used to assess trial quality. Cochrane reviews searched more databases and were more likely to assess trial quality, report dichotomous outcomes for individual trials, and conduct a meta-analysis than non-Cochrane reviews. Non-Cochrane reviews were more likely to conclude that there was a beneficial effect of treatment. Cochrane reviews were of higher quality than non-Cochrane reviews. There has been an increase in the quality of systematic reviews over time. The OQAQ has fair to good interrater reliability.

Conclusion: The quality of systematic reviews in physiotherapy is improving, and the use of Cochrane Collaboration procedures appears to improve the methods and quality.

Publication types

  • Research Support, Non-U.S. Gov't
  • Review

MeSH terms

  • Databases, Bibliographic
  • Evidence-Based Practice / methods*
  • Evidence-Based Practice / standards
  • Humans
  • Information Storage and Retrieval / standards
  • Meta-Analysis as Topic
  • Physical Therapy Specialty / standards*
  • Publication Bias
  • Quality Control
  • Research Design / standards
  • Review Literature as Topic*