首页> Statistical methods in medical research> Propensity score analysis with partially observed covariates: How should multiple imputation be used?

Propensity score analysis with partially observed covariates: How should multiple imputation be used?

导出 OA

翻译打印收藏纠错

第一作者： Clémence,Leyrat

第一单位： 1 Department of Medical Statistics, London School of Hygiene and Tropical Medicine, UK.

作者： Clémence,Leyrat ^[1] ; Shaun R,Seaman ^[2] ; Ian R,White ^[3] ; Ian,Douglas ^[4] ; Liam,Smeeth ^[4] ; Joseph,Kim ^[5] ; Matthieu,Resche-Rigon ^[6] ; James R,Carpenter ^[7] ; Elizabeth J,Williamson ^[8]

作者单位： 1 Department of Medical Statistics, London School of Hygiene and Tropical Medicine, UK. ^[1] 2 MRC Biostatistics Unit, Cambridge Institute for Public Health, Cambridge, UK. ^[2] 2 MRC Biostatistics Unit, Cambridge Institute for Public Health, Cambridge, UK.;3 London Hub for Trials Methodology Research, MRC Clinical Trials Unit, UCL, London, UK. ^[3] 4 Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, UK. ^[4] 1 Department of Medical Statistics, London School of Hygiene and Tropical Medicine, UK.;5 IMS Health, Real-World Evidence Solutions, UK. ^[5] 6 SBIM Biostatistics and Medical Information, Hôpital Saint-Louis, France.;7 ECSTRA Team (Epidémiologie Clinique et Statistiques pour la Recherche en Santé), UMR 1153 INSERM, Université Paris Diderot, France. ^[6] 1 Department of Medical Statistics, London School of Hygiene and Tropical Medicine, UK.;3 London Hub for Trials Methodology Research, MRC Clinical Trials Unit, UCL, London, UK. ^[7] 1 Department of Medical Statistics, London School of Hygiene and Tropical Medicine, UK.;8 Farr Institute of Health Informatics, London University College, London, UK. ^[8]

关键词 Missing covariates Rubin’s rules chained equations inverse probability of treatment weighting missingness pattern