{"id":{"repo_id":"mcmaster","oai_identifier":"oai:macsphere.mcmaster.ca:11375/33572"},"canonical_url":"https://search.ndltd.org/etd/mcmaster/oai:macsphere.mcmaster.ca:11375/33572","repository":{"repo_id":"mcmaster","name":"McMaster University","base_url":"https://macsphere.mcmaster.ca/server/oai/request"},"display":{"title":"Statistical Methods for Analyzing EQ-5D Data in Clinical Trials","abstract":"Patient-reported outcomes (PROs) are increasingly recognized as important outcomes in clinical trials. Among preference-weighted PRO measures for assessing health-related quality of life (HRQoL), the EQ-5D is one of the most widely used instruments and is routinely collected in randomized clinical trials (RCTs). The EQ-5D data, if analyzed properly, can produce important evidence for two purposes: estimating the treatment effect between arms and deriving health state utilities (HSUs) to support economic modeling. Although there is a guidance book for reporting and analyzing EQ-5D data, it primarily focuses on cross-sectional analyses. In contrast, EQ-5D data collected alongside RCTs are typically longitudinal in nature, requiring statistical methods that appropriately address key analytical challenges, including non-normal distributions, within-subject correlation over time, and missing data. However, methodological guidance on analyzing EQ-5D data in trials remains limited. This thesis aims to generate empirical evidence by comparing commonly used statistical models for estimating treatment effects and deriving HSUs using EQ-5D data from RCTs. The PhD project began with a systematic review of statistical methods used to analyze EQ-5D data in RCTs. Based on the findings, we conducted empirical comparisons of the most commonly used and recommended methods using individual-level data from multiple RCTs across different disease areas, focusing on treatment-effect estimation, missing data imputation, as well as HSU estimation. We further evaluated the implications of statistical model selection on both clinical and economic decision-making by examining its impact on statistical significance and clinical relevance of treatment-effect estimates, as well as on economic model outputs and cost-effectiveness conclusions. Across analyses, the empirical evidence supports linear mixed-effects models (LMMs) as a reasonable and pragmatic approach for analyzing EQ-5D data in RCTs, providing evidence-based methodological recommendations that may inform future guidance and improve the quality of EQ-5D analysis and reporting.","abstract_html":"Patient-reported outcomes (PROs) are increasingly recognized as important outcomes in clinical trials. Among preference-weighted PRO measures for assessing health-related quality of life (HRQoL), the EQ-5D is one of the most widely used instruments and is routinely collected in randomized clinical trials (RCTs). The EQ-5D data, if analyzed properly, can produce important evidence for two purposes: estimating the treatment effect between arms and deriving health state utilities (HSUs) to support economic modeling. Although there is a guidance book for reporting and analyzing EQ-5D data, it primarily focuses on cross-sectional analyses. In contrast, EQ-5D data collected alongside RCTs are typically longitudinal in nature, requiring statistical methods that appropriately address key analytical challenges, including non-normal distributions, within-subject correlation over time, and missing data. However, methodological guidance on analyzing EQ-5D data in trials remains limited. This thesis aims to generate empirical evidence by comparing commonly used statistical models for estimating treatment effects and deriving HSUs using EQ-5D data from RCTs. The PhD project began with a systematic review of statistical methods used to analyze EQ-5D data in RCTs. Based on the findings, we conducted empirical comparisons of the most commonly used and recommended methods using individual-level data from multiple RCTs across different disease areas, focusing on treatment-effect estimation, missing data imputation, as well as HSU estimation. We further evaluated the implications of statistical model selection on both clinical and economic decision-making by examining its impact on statistical significance and clinical relevance of treatment-effect estimates, as well as on economic model outputs and cost-effectiveness conclusions. Across analyses, the empirical evidence supports linear mixed-effects models (LMMs) as a reasonable and pragmatic approach for analyzing EQ-5D data in RCTs, providing evidence-based methodological recommendations that may inform future guidance and improve the quality of EQ-5D analysis and reporting.","abstract_has_math":false,"creators":["Yan, Jiajun"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":"Health Research Methodology","school":null,"contributors":[],"advisors":["Xie`, Feng"],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-18T19:31:48Z","subjects":["EQ-5D","randomized clinical trial","statistical analysis","patient-reported outcomes","model comparison"],"languages":[],"rights":["Attribution-NonCommercial-NoDerivatives 4.0 International"],"rights_urls":["http://creativecommons.org/licenses/by-nc-nd/4.0/"],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://doi.org/10.71548/7670"],"render_values":[{"text":"https://doi.org/10.71548/7670","href":"https://doi.org/10.71548/7670","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/11375/33572","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/11375/33572","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://macsphere.mcmaster.ca/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Amacsphere.mcmaster.ca%3A11375%2F33572","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Xie`, Feng"]},{"key":"dc:contributor.department","label":"Department","values":["Health Research Methodology"]},{"key":"dc:creator","label":"Author","values":["Yan, Jiajun"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-09-21T14:24:19Z"]},{"key":"dc:date.issued","label":"Date","values":["2026"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["EQ-5D","randomized clinical trial","statistical analysis","patient-reported outcomes","model comparison"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["Attribution-NonCommercial-NoDerivatives 4.0 International"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://creativecommons.org/licenses/by-nc-nd/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/11375/33572","https://doi.org/10.71548/7670"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Patient-reported outcomes (PROs) are increasingly recognized as important outcomes in clinical trials. Among preference-weighted PRO measures for assessing health-related quality of life (HRQoL), the EQ-5D is one of the most widely used instruments and is routinely collected in randomized clinical trials (RCTs). The EQ-5D data, if analyzed properly, can produce important evidence for two purposes: estimating the treatment effect between arms and deriving health state utilities (HSUs) to support economic modeling. Although there is a guidance book for reporting and analyzing EQ-5D data, it primarily focuses on cross-sectional analyses. In contrast, EQ-5D data collected alongside RCTs are typically longitudinal in nature, requiring statistical methods that appropriately address key analytical challenges, including non-normal distributions, within-subject correlation over time, and missing data. However, methodological guidance on analyzing EQ-5D data in trials remains limited. This thesis aims to generate empirical evidence by comparing commonly used statistical models for estimating treatment effects and deriving HSUs using EQ-5D data from RCTs. The PhD project began with a systematic review of statistical methods used to analyze EQ-5D data in RCTs. Based on the findings, we conducted empirical comparisons of the most commonly used and recommended methods using individual-level data from multiple RCTs across different disease areas, focusing on treatment-effect estimation, missing data imputation, as well as HSU estimation. We further evaluated the implications of statistical model selection on both clinical and economic decision-making by examining its impact on statistical significance and clinical relevance of treatment-effect estimates, as well as on economic model outputs and cost-effectiveness conclusions. Across analyses, the empirical evidence supports linear mixed-effects models (LMMs) as a reasonable and pragmatic approach for analyzing EQ-5D data in RCTs, providing evidence-based methodological recommendations that may inform future guidance and improve the quality of EQ-5D analysis and reporting."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Doctor of Philosophy (PhD)"]},{"key":"dc:description.sponsorship","label":"Dc Description Sponsorship","values":["the EuroQol Research Foundation"]},{"key":"dc:title","label":"Title","values":["Statistical Methods for Analyzing EQ-5D Data in Clinical Trials"]},{"key":"others:access_status","label":"Others Access Status","values":["embargo"]}]}],"canonical_facts":{"dc:contributor.advisor":["Xie`, Feng"],"dc:contributor.department":["Health Research Methodology"],"dc:creator":["Yan, Jiajun"],"dc:date.accessioned":["2026-09-21T14:24:19Z"],"dc:date.issued":["2026"],"dc:description.abstract":["Patient-reported outcomes (PROs) are increasingly recognized as important outcomes in clinical trials. Among preference-weighted PRO measures for assessing health-related quality of life (HRQoL), the EQ-5D is one of the most widely used instruments and is routinely collected in randomized clinical trials (RCTs). The EQ-5D data, if analyzed properly, can produce important evidence for two purposes: estimating the treatment effect between arms and deriving health state utilities (HSUs) to support economic modeling. Although there is a guidance book for reporting and analyzing EQ-5D data, it primarily focuses on cross-sectional analyses. In contrast, EQ-5D data collected alongside RCTs are typically longitudinal in nature, requiring statistical methods that appropriately address key analytical challenges, including non-normal distributions, within-subject correlation over time, and missing data. However, methodological guidance on analyzing EQ-5D data in trials remains limited. This thesis aims to generate empirical evidence by comparing commonly used statistical models for estimating treatment effects and deriving HSUs using EQ-5D data from RCTs. The PhD project began with a systematic review of statistical methods used to analyze EQ-5D data in RCTs. Based on the findings, we conducted empirical comparisons of the most commonly used and recommended methods using individual-level data from multiple RCTs across different disease areas, focusing on treatment-effect estimation, missing data imputation, as well as HSU estimation. We further evaluated the implications of statistical model selection on both clinical and economic decision-making by examining its impact on statistical significance and clinical relevance of treatment-effect estimates, as well as on economic model outputs and cost-effectiveness conclusions. Across analyses, the empirical evidence supports linear mixed-effects models (LMMs) as a reasonable and pragmatic approach for analyzing EQ-5D data in RCTs, providing evidence-based methodological recommendations that may inform future guidance and improve the quality of EQ-5D analysis and reporting."],"dc:description.degree":["Doctor of Philosophy (PhD)"],"dc:description.sponsorship":["the EuroQol Research Foundation"],"dc:identifier.uri":["https://hdl.handle.net/11375/33572","https://doi.org/10.71548/7670"],"dc:rights":["Attribution-NonCommercial-NoDerivatives 4.0 International"],"dc:rights.uri":["http://creativecommons.org/licenses/by-nc-nd/4.0/"],"dc:subject":["EQ-5D","randomized clinical trial","statistical analysis","patient-reported outcomes","model comparison"],"dc:title":["Statistical Methods for Analyzing EQ-5D Data in Clinical Trials"],"dc:type":["Thesis"],"others:access_status":["embargo"]},"updated_at":"2026-11-18T19:31:48Z"}