Relative efficiency of equal versus unequal cluster sizes in cluster randomized trials with a small number of clusters
To calculate sample sizes in cluster randomized trials (CRTs), the cluster sizes are usually assumed to be identical across all clusters for simplicity. However, equal cluster sizes are not guaranteed in practice, especially when the number of clusters is limited. Therefore, it is important to under...
Gespeichert in:
Veröffentlicht in: | Journal of biopharmaceutical statistics 2021-03, Vol.31 (2), p.191-206 |
---|---|
Hauptverfasser: | , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 206 |
---|---|
container_issue | 2 |
container_start_page | 191 |
container_title | Journal of biopharmaceutical statistics |
container_volume | 31 |
creator | Liu, Jingxia Xiong, Chengjie Liu, Lei Wang, Guoqiao Jingqin, Luo Gao, Feng Chen, Ling Li, Yan |
description | To calculate sample sizes in cluster randomized trials (CRTs), the cluster sizes are usually assumed to be identical across all clusters for simplicity. However, equal cluster sizes are not guaranteed in practice, especially when the number of clusters is limited. Therefore, it is important to understand the relative efficiency (RE) of equal versus unequal cluster sizes when designing CRTs with a limited number of clusters. In this paper, we are interested in the RE of two bias-corrected sandwich estimators of the treatment effect in the Generalized Estimating Equation (GEE) models for CRTs with a small number of clusters. Specifically, we derive the RE of two bias-corrected sandwich estimators for binary, continuous, or count data in CRTs under the assumption of an exchangeable working correlation structure. We consider different scenarios of cluster size distributions and investigate RE performance through simulation studies. We conclude that the number of clusters could be increased by as much as 42% to compensate for efficiency loss due to unequal cluster sizes. Finally, we propose an algorithm of increasing the number of clusters when the coefficient of variation of cluster sizes is known and unknown. |
doi_str_mv | 10.1080/10543406.2020.1814795 |
format | Article |
fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_proquest_miscellaneous_2446667770</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3087511581</sourcerecordid><originalsourceid>FETCH-LOGICAL-c496t-d539abf4f640fdd25d278be8687be1f5bf6ee6b8190efa1bf90e33d5f9a32e973</originalsourceid><addsrcrecordid>eNp9kU1rFTEUhgdRbK3-BCXgxs3UfCezEUvxCwqC6DpkZk5sSiZpk8kt119vLvf2oi5cneTNc96cw9t1Lwk-J1jjtwQLzjiW5xTTJmnC1SAedadEUNwLRcjjdm5Mv4NOumel3GBMhNL8aXfC6KCwoPS023yDYFe_AQTO-clDnLYoOQR31Qa0gVxqQTXur1OoZYWMiv8FBfl4FLKNc1qaOqM1exsKuvfrNbKoLDYEFOsyNqrZHhrK8-6Jaxi8ONSz7sfHD98vP_dXXz99uby46ic-yLWfBRvs6LiTHLt5pmKmSo-gpVYjECdGJwHkqMmAwVkyulYZm4UbLKMwKHbWvdv73tZxgXmCuGYbzG32i81bk6w3f79Ef21-po3RinHOWDN4czDI6a5CWc3iywQh2AipFkM5l1IqpXBDX_-D3qSaY1vPMKyVIERo0iixp6acSsngjsMQbHbJmodkzS5Zc0i29b36c5Nj10OUDXi_B3x0KS_2PuUwm9VuQ8quBTT5Nsf___gNRs22Pw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3087511581</pqid></control><display><type>article</type><title>Relative efficiency of equal versus unequal cluster sizes in cluster randomized trials with a small number of clusters</title><source>MEDLINE</source><source>EBSCOhost Business Source Complete</source><creator>Liu, Jingxia ; Xiong, Chengjie ; Liu, Lei ; Wang, Guoqiao ; Jingqin, Luo ; Gao, Feng ; Chen, Ling ; Li, Yan</creator><creatorcontrib>Liu, Jingxia ; Xiong, Chengjie ; Liu, Lei ; Wang, Guoqiao ; Jingqin, Luo ; Gao, Feng ; Chen, Ling ; Li, Yan</creatorcontrib><description>To calculate sample sizes in cluster randomized trials (CRTs), the cluster sizes are usually assumed to be identical across all clusters for simplicity. However, equal cluster sizes are not guaranteed in practice, especially when the number of clusters is limited. Therefore, it is important to understand the relative efficiency (RE) of equal versus unequal cluster sizes when designing CRTs with a limited number of clusters. In this paper, we are interested in the RE of two bias-corrected sandwich estimators of the treatment effect in the Generalized Estimating Equation (GEE) models for CRTs with a small number of clusters. Specifically, we derive the RE of two bias-corrected sandwich estimators for binary, continuous, or count data in CRTs under the assumption of an exchangeable working correlation structure. We consider different scenarios of cluster size distributions and investigate RE performance through simulation studies. We conclude that the number of clusters could be increased by as much as 42% to compensate for efficiency loss due to unequal cluster sizes. Finally, we propose an algorithm of increasing the number of clusters when the coefficient of variation of cluster sizes is known and unknown.</description><identifier>ISSN: 1054-3406</identifier><identifier>EISSN: 1520-5711</identifier><identifier>DOI: 10.1080/10543406.2020.1814795</identifier><identifier>PMID: 32970522</identifier><language>eng</language><publisher>England: Taylor & Francis</publisher><subject>Bias ; Bias-corrected sandwich estimator ; Cluster Analysis ; cluster randomized trial (CRT) ; Computer Simulation ; Efficiency ; generalized estimating equation (GEE) ; Humans ; intracluster correlation coefficient (ICC) ; Randomized Controlled Trials as Topic ; relative efficiency (RE) ; Sample Size</subject><ispartof>Journal of biopharmaceutical statistics, 2021-03, Vol.31 (2), p.191-206</ispartof><rights>2020 Taylor & Francis Group, LLC 2020</rights><rights>2020 Taylor & Francis Group, LLC</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c496t-d539abf4f640fdd25d278be8687be1f5bf6ee6b8190efa1bf90e33d5f9a32e973</citedby><cites>FETCH-LOGICAL-c496t-d539abf4f640fdd25d278be8687be1f5bf6ee6b8190efa1bf90e33d5f9a32e973</cites><orcidid>0000-0002-9434-7907 ; 0000-0002-1425-1623 ; 0000-0003-0238-1831 ; 0000-0003-1340-6694</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>230,314,780,784,885,27924,27925</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/32970522$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Liu, Jingxia</creatorcontrib><creatorcontrib>Xiong, Chengjie</creatorcontrib><creatorcontrib>Liu, Lei</creatorcontrib><creatorcontrib>Wang, Guoqiao</creatorcontrib><creatorcontrib>Jingqin, Luo</creatorcontrib><creatorcontrib>Gao, Feng</creatorcontrib><creatorcontrib>Chen, Ling</creatorcontrib><creatorcontrib>Li, Yan</creatorcontrib><title>Relative efficiency of equal versus unequal cluster sizes in cluster randomized trials with a small number of clusters</title><title>Journal of biopharmaceutical statistics</title><addtitle>J Biopharm Stat</addtitle><description>To calculate sample sizes in cluster randomized trials (CRTs), the cluster sizes are usually assumed to be identical across all clusters for simplicity. However, equal cluster sizes are not guaranteed in practice, especially when the number of clusters is limited. Therefore, it is important to understand the relative efficiency (RE) of equal versus unequal cluster sizes when designing CRTs with a limited number of clusters. In this paper, we are interested in the RE of two bias-corrected sandwich estimators of the treatment effect in the Generalized Estimating Equation (GEE) models for CRTs with a small number of clusters. Specifically, we derive the RE of two bias-corrected sandwich estimators for binary, continuous, or count data in CRTs under the assumption of an exchangeable working correlation structure. We consider different scenarios of cluster size distributions and investigate RE performance through simulation studies. We conclude that the number of clusters could be increased by as much as 42% to compensate for efficiency loss due to unequal cluster sizes. Finally, we propose an algorithm of increasing the number of clusters when the coefficient of variation of cluster sizes is known and unknown.</description><subject>Bias</subject><subject>Bias-corrected sandwich estimator</subject><subject>Cluster Analysis</subject><subject>cluster randomized trial (CRT)</subject><subject>Computer Simulation</subject><subject>Efficiency</subject><subject>generalized estimating equation (GEE)</subject><subject>Humans</subject><subject>intracluster correlation coefficient (ICC)</subject><subject>Randomized Controlled Trials as Topic</subject><subject>relative efficiency (RE)</subject><subject>Sample Size</subject><issn>1054-3406</issn><issn>1520-5711</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp9kU1rFTEUhgdRbK3-BCXgxs3UfCezEUvxCwqC6DpkZk5sSiZpk8kt119vLvf2oi5cneTNc96cw9t1Lwk-J1jjtwQLzjiW5xTTJmnC1SAedadEUNwLRcjjdm5Mv4NOumel3GBMhNL8aXfC6KCwoPS023yDYFe_AQTO-clDnLYoOQR31Qa0gVxqQTXur1OoZYWMiv8FBfl4FLKNc1qaOqM1exsKuvfrNbKoLDYEFOsyNqrZHhrK8-6Jaxi8ONSz7sfHD98vP_dXXz99uby46ic-yLWfBRvs6LiTHLt5pmKmSo-gpVYjECdGJwHkqMmAwVkyulYZm4UbLKMwKHbWvdv73tZxgXmCuGYbzG32i81bk6w3f79Ef21-po3RinHOWDN4czDI6a5CWc3iywQh2AipFkM5l1IqpXBDX_-D3qSaY1vPMKyVIERo0iixp6acSsngjsMQbHbJmodkzS5Zc0i29b36c5Nj10OUDXi_B3x0KS_2PuUwm9VuQ8quBTT5Nsf___gNRs22Pw</recordid><startdate>20210304</startdate><enddate>20210304</enddate><creator>Liu, Jingxia</creator><creator>Xiong, Chengjie</creator><creator>Liu, Lei</creator><creator>Wang, Guoqiao</creator><creator>Jingqin, Luo</creator><creator>Gao, Feng</creator><creator>Chen, Ling</creator><creator>Li, Yan</creator><general>Taylor & Francis</general><general>Taylor & Francis Ltd</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0002-9434-7907</orcidid><orcidid>https://orcid.org/0000-0002-1425-1623</orcidid><orcidid>https://orcid.org/0000-0003-0238-1831</orcidid><orcidid>https://orcid.org/0000-0003-1340-6694</orcidid></search><sort><creationdate>20210304</creationdate><title>Relative efficiency of equal versus unequal cluster sizes in cluster randomized trials with a small number of clusters</title><author>Liu, Jingxia ; Xiong, Chengjie ; Liu, Lei ; Wang, Guoqiao ; Jingqin, Luo ; Gao, Feng ; Chen, Ling ; Li, Yan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c496t-d539abf4f640fdd25d278be8687be1f5bf6ee6b8190efa1bf90e33d5f9a32e973</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Bias</topic><topic>Bias-corrected sandwich estimator</topic><topic>Cluster Analysis</topic><topic>cluster randomized trial (CRT)</topic><topic>Computer Simulation</topic><topic>Efficiency</topic><topic>generalized estimating equation (GEE)</topic><topic>Humans</topic><topic>intracluster correlation coefficient (ICC)</topic><topic>Randomized Controlled Trials as Topic</topic><topic>relative efficiency (RE)</topic><topic>Sample Size</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Liu, Jingxia</creatorcontrib><creatorcontrib>Xiong, Chengjie</creatorcontrib><creatorcontrib>Liu, Lei</creatorcontrib><creatorcontrib>Wang, Guoqiao</creatorcontrib><creatorcontrib>Jingqin, Luo</creatorcontrib><creatorcontrib>Gao, Feng</creatorcontrib><creatorcontrib>Chen, Ling</creatorcontrib><creatorcontrib>Li, Yan</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Journal of biopharmaceutical statistics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Liu, Jingxia</au><au>Xiong, Chengjie</au><au>Liu, Lei</au><au>Wang, Guoqiao</au><au>Jingqin, Luo</au><au>Gao, Feng</au><au>Chen, Ling</au><au>Li, Yan</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Relative efficiency of equal versus unequal cluster sizes in cluster randomized trials with a small number of clusters</atitle><jtitle>Journal of biopharmaceutical statistics</jtitle><addtitle>J Biopharm Stat</addtitle><date>2021-03-04</date><risdate>2021</risdate><volume>31</volume><issue>2</issue><spage>191</spage><epage>206</epage><pages>191-206</pages><issn>1054-3406</issn><eissn>1520-5711</eissn><abstract>To calculate sample sizes in cluster randomized trials (CRTs), the cluster sizes are usually assumed to be identical across all clusters for simplicity. However, equal cluster sizes are not guaranteed in practice, especially when the number of clusters is limited. Therefore, it is important to understand the relative efficiency (RE) of equal versus unequal cluster sizes when designing CRTs with a limited number of clusters. In this paper, we are interested in the RE of two bias-corrected sandwich estimators of the treatment effect in the Generalized Estimating Equation (GEE) models for CRTs with a small number of clusters. Specifically, we derive the RE of two bias-corrected sandwich estimators for binary, continuous, or count data in CRTs under the assumption of an exchangeable working correlation structure. We consider different scenarios of cluster size distributions and investigate RE performance through simulation studies. We conclude that the number of clusters could be increased by as much as 42% to compensate for efficiency loss due to unequal cluster sizes. Finally, we propose an algorithm of increasing the number of clusters when the coefficient of variation of cluster sizes is known and unknown.</abstract><cop>England</cop><pub>Taylor & Francis</pub><pmid>32970522</pmid><doi>10.1080/10543406.2020.1814795</doi><tpages>16</tpages><orcidid>https://orcid.org/0000-0002-9434-7907</orcidid><orcidid>https://orcid.org/0000-0002-1425-1623</orcidid><orcidid>https://orcid.org/0000-0003-0238-1831</orcidid><orcidid>https://orcid.org/0000-0003-1340-6694</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1054-3406 |
ispartof | Journal of biopharmaceutical statistics, 2021-03, Vol.31 (2), p.191-206 |
issn | 1054-3406 1520-5711 |
language | eng |
recordid | cdi_proquest_miscellaneous_2446667770 |
source | MEDLINE; EBSCOhost Business Source Complete |
subjects | Bias Bias-corrected sandwich estimator Cluster Analysis cluster randomized trial (CRT) Computer Simulation Efficiency generalized estimating equation (GEE) Humans intracluster correlation coefficient (ICC) Randomized Controlled Trials as Topic relative efficiency (RE) Sample Size |
title | Relative efficiency of equal versus unequal cluster sizes in cluster randomized trials with a small number of clusters |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T02%3A44%3A16IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Relative%20efficiency%20of%20equal%20versus%20unequal%20cluster%20sizes%20in%20cluster%20randomized%20trials%20with%20a%20small%20number%20of%20clusters&rft.jtitle=Journal%20of%20biopharmaceutical%20statistics&rft.au=Liu,%20Jingxia&rft.date=2021-03-04&rft.volume=31&rft.issue=2&rft.spage=191&rft.epage=206&rft.pages=191-206&rft.issn=1054-3406&rft.eissn=1520-5711&rft_id=info:doi/10.1080/10543406.2020.1814795&rft_dat=%3Cproquest_pubme%3E3087511581%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3087511581&rft_id=info:pmid/32970522&rfr_iscdi=true |