Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Measuring inconsistency in meta-analyses

Measuring inconsistency in meta-analyses Education and debate JulianPTHiggins, Simon G Thompson, Jonathan J Deeks, Douglas G Altman Cochrane Reviews have recently started including the quantity I to help readers assess the consistency of the results of studies in meta-analyses. What does this new quantity mean, and why is assessment of heterogeneity so important to clinical practice? MRC Biostatistics Systematic reviews and meta-analyses can provide intervals not overlapping. But the test of heterogeneity Unit, Institute of convincing and reliable evidence relevant to many yieldsaPvalue of 0.09, conventionally interpreted as Public Health, aspects of medicine and health care. Their value is being non-significant. Because the test is poor at Cambridge CB2 2SR especially clear when the results of the studies they detecting true heterogeneity, a non-significant result JulianPTHiggins include show clinically important effects of similar cannot be taken as evidence of homogeneity. Using a statistician magnitude. However, the conclusions are less clear cut-off of 10% for significance ameliorates this prob- Simon G when the included studies have differing results. In an Thompson lem but increases the risk of drawing a false positive director attempt to establish whether studies are consistent, 10 conclusion (type I error). reports of meta-analyses commonly present a statisti- Cancer Research Conversely, http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMJ British Medical Journal

Measuring inconsistency in meta-analyses

Measuring inconsistency in meta-analyses

BMJ , Volume 327 (7414) – Sep 6, 2003

Abstract

Education and debate JulianPTHiggins, Simon G Thompson, Jonathan J Deeks, Douglas G Altman Cochrane Reviews have recently started including the quantity I to help readers assess the consistency of the results of studies in meta-analyses. What does this new quantity mean, and why is assessment of heterogeneity so important to clinical practice? MRC Biostatistics Systematic reviews and meta-analyses can provide intervals not overlapping. But the test of heterogeneity Unit, Institute of convincing and reliable evidence relevant to many yieldsaPvalue of 0.09, conventionally interpreted as Public Health, aspects of medicine and health care. Their value is being non-significant. Because the test is poor at Cambridge CB2 2SR especially clear when the results of the studies they detecting true heterogeneity, a non-significant result JulianPTHiggins include show clinically important effects of similar cannot be taken as evidence of homogeneity. Using a statistician magnitude. However, the conclusions are less clear cut-off of 10% for significance ameliorates this prob- Simon G when the included studies have differing results. In an Thompson lem but increases the risk of drawing a false positive director attempt to establish whether studies are consistent, 10 conclusion (type I error). reports of meta-analyses commonly present a statisti- Cancer Research Conversely,

Loading next page...
 
/lp/british-medical-journal/measuring-inconsistency-in-meta-analyses-QrvGq03yU2

References (30)

Publisher
British Medical Journal
Copyright
© 2003 BMJ Publishing Group Ltd.
ISSN
0959-8138
eISSN
1468-5833
DOI
10.1136/bmj.327.7414.557
Publisher site
See Article on Publisher Site

Abstract

Education and debate JulianPTHiggins, Simon G Thompson, Jonathan J Deeks, Douglas G Altman Cochrane Reviews have recently started including the quantity I to help readers assess the consistency of the results of studies in meta-analyses. What does this new quantity mean, and why is assessment of heterogeneity so important to clinical practice? MRC Biostatistics Systematic reviews and meta-analyses can provide intervals not overlapping. But the test of heterogeneity Unit, Institute of convincing and reliable evidence relevant to many yieldsaPvalue of 0.09, conventionally interpreted as Public Health, aspects of medicine and health care. Their value is being non-significant. Because the test is poor at Cambridge CB2 2SR especially clear when the results of the studies they detecting true heterogeneity, a non-significant result JulianPTHiggins include show clinically important effects of similar cannot be taken as evidence of homogeneity. Using a statistician magnitude. However, the conclusions are less clear cut-off of 10% for significance ameliorates this prob- Simon G when the included studies have differing results. In an Thompson lem but increases the risk of drawing a false positive director attempt to establish whether studies are consistent, 10 conclusion (type I error). reports of meta-analyses commonly present a statisti- Cancer Research Conversely,

Journal

BMJBritish Medical Journal

Published: Sep 6, 2003

There are no references for this article.