The selection of summary statistics to use in a meta-analysis is very important for the interpretation and application of its results. This paper introduces some basic concepts of summary statistics in meta-analysis. The selection of a summary statistic for a meta-analysis depends on the following factors: design of the studies being combined, type of data, consistency among the included studies, mathematical properties and ease of interpretation. For continuous data, the weighted mean difference (WMD) is recommended when all trials use the same scale to report their outcomes, while standardized mean difference (SMD) is more appropriate when trials use different scales to report their outcomes, or the means of their outcomes differ greatly. For dichotomous data, rate ratio or relative risk (RR) is bly recommended to be the summary statistics for meta-analyses of randomized trials. The use of odds ratio (OR) as the summary statistic is similar to that of RR, if the event being studied in both the intervention (exposure) and the control group is rare. There is no single measurement that is uniformly best for all meta-analyses.
Citation:
WEN Jin,LI Youping. The Selection of a Summary Statistic for Use in Meta-analysis. Chinese Journal of Evidence-Based Medicine, 2007, 07(8): 606-613. doi:
Copy
Copyright © the editorial department of Chinese Journal of Evidence-Based Medicine of West China Medical Publisher. All rights reserved
1. |
Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions 4.2.6 [updated September 2006]. In: The Cochrane Library, Issue 4, 2006. Chichester, UK: John Wiley & Sons, Ltd.
|
2. |
Deeks JJ. Issues in the selection of a summary statistic for meta-analysis of clinical trials with binary outcomes. Statistics in Medicine, 2002, 21: 1575–1600.
|
3. |
Cook TD. Advanced statistics: up with odds ratios! A case for odds ratios when outcomes are common. Academic Emergency Medicine, 2002, 9: 1430-1434.
|
4. |
Sackett DL, Deeks JJ, Altman DG. Down with odds ratios! Evidence-Based Medicine, 1996, 1: 164-167.
|
5. |
Deeks J. When can odds ratios mislead? British Medical Journal 1998, 317: 1155.
|
6. |
Senn S. Odds ratios revisited. Evidence-Based Medicine 1998, 3: 71.
|
7. |
Olkin I. Odds ratios revisited. Evidence-Based Medicine 1998, 3: 71.
|
- 1. Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions 4.2.6 [updated September 2006]. In: The Cochrane Library, Issue 4, 2006. Chichester, UK: John Wiley & Sons, Ltd.
- 2. Deeks JJ. Issues in the selection of a summary statistic for meta-analysis of clinical trials with binary outcomes. Statistics in Medicine, 2002, 21: 1575–1600.
- 3. Cook TD. Advanced statistics: up with odds ratios! A case for odds ratios when outcomes are common. Academic Emergency Medicine, 2002, 9: 1430-1434.
- 4. Sackett DL, Deeks JJ, Altman DG. Down with odds ratios! Evidence-Based Medicine, 1996, 1: 164-167.
- 5. Deeks J. When can odds ratios mislead? British Medical Journal 1998, 317: 1155.
- 6. Senn S. Odds ratios revisited. Evidence-Based Medicine 1998, 3: 71.
- 7. Olkin I. Odds ratios revisited. Evidence-Based Medicine 1998, 3: 71.