Show simple item record

dc.contributor.authorBakbergenuly, Ilyas
dc.contributor.authorHoaglin, David C
dc.contributor.authorKulinskaya, Elena
dc.date2022-08-11T08:10:36.000
dc.date.accessioned2022-08-23T17:14:01Z
dc.date.available2022-08-23T17:14:01Z
dc.date.issued2020-02-29
dc.date.submitted2020-03-23
dc.identifier.citation<p>Bakbergenuly I, Hoaglin DC, Kulinskaya E. Methods for estimating between-study variance and overall effect in meta-analysis of odds-ratios. Res Synth Methods. 2020 Feb 29. doi: 10.1002/jrsm.1404. Epub ahead of print. PMID: 32112619. <a href="https://doi.org/10.1002/jrsm.1404">Link to article on publisher's site</a></p>
dc.identifier.issn1759-2879 (Linking)
dc.identifier.doi10.1002/jrsm.1404
dc.identifier.pmid32112619
dc.identifier.urihttp://hdl.handle.net/20.500.14038/46865
dc.description.abstractIn random-effects meta-analysis the between-study variance (tau(2) ) has a key role in assessing heterogeneity of study-level estimates and combining them to estimate an overall effect. For odds ratios the most common methods suffer from bias in estimating tau(2) and the overall effect and produce confidence intervals with below-nominal coverage. An improved approximation to the moments of Cochran's Q statistic, suggested by Kulinskaya and Dollinger, yields new point and interval estimators (KD) of tau(2) and of the overall log-odds-ratio. Another, simpler approach (SSW) uses weights based only on study-level sample sizes to estimate the overall effect. In extensive simulations we compare our proposed estimators with established point and interval estimators for tau(2) and point and interval estimators for the overall log-odds-ratio (including the Hartung-Knapp-Sidik-Jonkman interval). Additional simulations included three estimators based on generalized linear mixed models and the Mantel-Haenszel fixed-effect estimator. Results of our simulations show that no single point estimator of tau(2) can be recommended exclusively, but Mandel-Paule and KD provide better choices for small and large K, respectively. The KD estimator provides reliable coverage of tau(2) . Inverse-variance-weighted estimators of the overall effect are substantially biased, as are the Mantel-Haenszel odds-ratio and the estimators from the generalized linear mixed models. The SSW estimator of the overall effect and a related confidence interval provide reliable point and interval estimation of the overall log-odds-ratio.
dc.language.isoen_US
dc.relation<p><a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&list_uids=32112619&dopt=Abstract">Link to Article in PubMed</a></p>
dc.relation.urlhttps://doi.org/10.1002/jrsm.1404
dc.rightsAccepted manuscript posted after 12 months as allowed by the publisher's self-archiving policy athttps://authorservices.wiley.com/author-resources/Journal-Authors/licensing/self-archiving.html.
dc.subjectbetween-study variance
dc.subjectbinary outcome
dc.subjectheterogeneity
dc.subjectmeta-analysis
dc.subjectrandom-effects model
dc.subjectBiostatistics
dc.subjectEpidemiology
dc.subjectHealth Services Research
dc.subjectStatistics and Probability
dc.titleMethods for estimating between-study variance and overall effect in meta-analysis of odds-ratios
dc.typeJournal Article
dc.source.journaltitleResearch synthesis methods
dc.identifier.legacyfulltexthttps://escholarship.umassmed.edu/cgi/viewcontent.cgi?article=2341&amp;context=qhs_pp&amp;unstamped=1
dc.identifier.legacycoverpagehttps://escholarship.umassmed.edu/qhs_pp/1338
dc.legacy.embargo2021-03-01T00:00:00-08:00
dc.identifier.contextkey16955926
refterms.dateFOA2022-08-23T17:14:01Z
html.description.abstract<p>In random-effects meta-analysis the between-study variance (tau(2) ) has a key role in assessing heterogeneity of study-level estimates and combining them to estimate an overall effect. For odds ratios the most common methods suffer from bias in estimating tau(2) and the overall effect and produce confidence intervals with below-nominal coverage. An improved approximation to the moments of Cochran's Q statistic, suggested by Kulinskaya and Dollinger, yields new point and interval estimators (KD) of tau(2) and of the overall log-odds-ratio. Another, simpler approach (SSW) uses weights based only on study-level sample sizes to estimate the overall effect. In extensive simulations we compare our proposed estimators with established point and interval estimators for tau(2) and point and interval estimators for the overall log-odds-ratio (including the Hartung-Knapp-Sidik-Jonkman interval). Additional simulations included three estimators based on generalized linear mixed models and the Mantel-Haenszel fixed-effect estimator. Results of our simulations show that no single point estimator of tau(2) can be recommended exclusively, but Mandel-Paule and KD provide better choices for small and large K, respectively. The KD estimator provides reliable coverage of tau(2) . Inverse-variance-weighted estimators of the overall effect are substantially biased, as are the Mantel-Haenszel odds-ratio and the estimators from the generalized linear mixed models. The SSW estimator of the overall effect and a related confidence interval provide reliable point and interval estimation of the overall log-odds-ratio.</p>
dc.identifier.submissionpathqhs_pp/1338
dc.contributor.departmentQuantitative Health Sciences


Files in this item

Thumbnail
Name:
jrsm.1404.pdf
Size:
1.220Mb
Format:
PDF

This item appears in the following Collection(s)

Show simple item record