Issues in the selection of a summary statistic for meta-analysis of clinical trials with binary outcomes - PubMed

Affiliations

• PMID: 12111921
• DOI: 10.1002/sim.1188

Comparative Study

Issues in the selection of a summary statistic for meta-analysis of clinical trials with binary outcomes

Jonathan J Deeks. Stat Med. 2002.

Abstract

Meta-analysis of binary data involves the computation of a weighted average of summary statistics calculated for each trial. The selection of the appropriate summary statistic is a subject of debate due to conflicts in the relative importance of mathematical properties and the ability to intuitively interpret results. This paper explores the process of identifying a summary statistic most likely to be consistent across trials when there is variation in control group event rates. Four summary statistics are considered: odds ratios (OR); risk differences (RD) and risk ratios of beneficial (RR(B)); and harmful outcomes (RR(H)). Each summary statistic corresponds to a different pattern of predicted absolute benefit of treatment with variation in baseline risk, the greatest difference in patterns of prediction being between RR(B) and RR(H). Selection of a summary statistic solely based on identification of the best-fitting model by comparing tests of heterogeneity is problematic, principally due to low numbers of trials. It is proposed that choice of a summary statistic should be guided by both empirical evidence and clinically informed debate as to which model is likely to be closest to the expected pattern of treatment benefit across baseline risks. Empirical investigations comparing the four summary statistics on a sample of 551 systematic reviews provide evidence that the RR and OR models are on average more consistent than RD, there being no difference on average between RR and OR. From a second sample of 114 meta-analyses evidence indicates that for interventions aimed at preventing an undesirable event, greatest absolute benefits are observed in trials with the highest baseline event rates, corresponding to the model of constant RR(H). The appropriate selection for a particular meta-analysis may depend on understanding reasons for variation in control group event rates; in some situations uncertainty about the choice of summary statistic will remain.

Similar articles

• Bayesian random effects meta-analysis of trials with binary outcomes: methods for the absolute risk difference and relative risk scales.

  Warn DE, Thompson SG, Spiegelhalter DJ. Warn DE, et al. Stat Med. 2002 Jun 15;21(11):1601-23. doi: 10.1002/sim.1189. Stat Med. 2002. PMID: 12111922

• Benefits and risks of adjunctive inhaled corticosteroids in chronic obstructive pulmonary disease: a meta-analysis.

  Sobieraj DM, White CM, Coleman CI. Sobieraj DM, et al. Clin Ther. 2008 Aug;30(8):1416-25. doi: 10.1016/j.clinthera.2008.08.004. Clin Ther. 2008. PMID: 18803985

• Estimating risk difference from relative association measures in meta-analysis can infrequently pose interpretational challenges.

  Murad MH, Montori VM, Walter SD, Guyatt GH. Murad MH, et al. J Clin Epidemiol. 2009 Aug;62(8):865-7. doi: 10.1016/j.jclinepi.2008.11.005. Epub 2009 Feb 20. J Clin Epidemiol. 2009. PMID: 19230610

• Meta-analysis: role of Helicobacter pylori eradication in the prevention of peptic ulcer in NSAID users.

  Vergara M, Catalán M, Gisbert JP, Calvet X. Vergara M, et al. Aliment Pharmacol Ther. 2005 Jun 15;21(12):1411-8. doi: 10.1111/j.1365-2036.2005.02444.x. Aliment Pharmacol Ther. 2005. PMID: 15948807 Review.

• [Meta-analysis of therapeutic and preventive clinical trials].

  Lugosi L, Molnár I. Lugosi L, et al. Orv Hetil. 2000 Dec 17;141(51):2779-84. Orv Hetil. 2000. PMID: 11196238 Review. Hungarian.

Cited by

• Use of colloids and crystalloids for perioperative clinical infusion management in cardiac surgery patients and postoperative outcomes: a meta-analysis.

  Chen SD, Ma YT, Wei HX, Ou XR, Liu JY, Tian YL, Zhang C, Xu YJ, Kong Y. Chen SD, et al. Perioper Med (Lond). 2024 Jul 24;13(1):83. doi: 10.1186/s13741-024-00445-0. Perioper Med (Lond). 2024. PMID: 39049111 Free PMC article.

• Antiviral treatment and other therapeutic interventions for herpes simplex virus epithelial keratitis.

  Wilhelmus KR. Wilhelmus KR. Cochrane Database Syst Rev. 2015 Jan 9;1(1):CD002898. doi: 10.1002/14651858.CD002898.pub5. Cochrane Database Syst Rev. 2015. PMID: 25879115 Free PMC article. Review.

• Assessing subgroup effects with binary data: can the use of different effect measures lead to different conclusions?

  White IR, Elbourne D. White IR, et al. BMC Med Res Methodol. 2005 Apr 29;5:15. doi: 10.1186/1471-2288-5-15. BMC Med Res Methodol. 2005. PMID: 15862124 Free PMC article.

• The choice of effect measure for binary outcomes: Introducing counterfactual outcome state transition parameters.

  Huitfeldt A, Goldstein A, Swanson SA. Huitfeldt A, et al. Epidemiol Methods. 2018 Dec;7(1):20160014. doi: 10.1515/em-2016-0014. Epub 2018 Jul 27. Epidemiol Methods. 2018. PMID: 30637184 Free PMC article.

• Short-term treatment of gastroesophageal reflux disease.

  van Pinxteren B, Numans ME, Lau J, de Wit NJ, Hungin AP, Bonis PA. van Pinxteren B, et al. J Gen Intern Med. 2003 Sep;18(9):755-63. doi: 10.1046/j.1525-1497.2003.20833.x. J Gen Intern Med. 2003. PMID: 12950485 Free PMC article. Review.

Publication types

MeSH terms

Substances

LinkOut - more resources

• Medical

  • MedlinePlus Health Information

• Miscellaneous

  • NCI CPTAC Assay Portal