Impact of missing data on bias and precision when estimating change in patient-reported outcomes from a clinical registry

被引:88
作者
Ayilara, Olawale F. [1 ]
Zhang, Lisa [1 ]
Sajobi, Tolulope T. [2 ]
Sawatzky, Richard [3 ]
Bohm, Eric [4 ]
Lix, Lisa M. [1 ]
机构
[1] Univ Manitoba, Dept Community Hlth Sci, S113-750 Bannatyne Ave, Winnipeg, MB R3E 0W3, Canada
[2] Univ Calgary, Cumming Sch Med, Dept Community Hlth Sci, Calgary, AB, Canada
[3] Trinity Western Univ, Sch Nursing, Langley, BC, Canada
[4] Univ Manitoba, Dept Surg, Winnipeg, MB, Canada
基金
加拿大健康研究院;
关键词
Auxiliary variables; Maximum likelihood estimation; Missing data; Registry; Mixed-effects model; MULTIPLE-IMPUTATION; MAXIMUM-LIKELIHOOD; LONGITUDINAL DATA; INFORMATION; SCORE; HIP;
D O I
10.1186/s12955-019-1181-2
中图分类号
R19 [保健组织与事业(卫生事业管理)];
学科分类号
摘要
BackgroundClinical registries, which capture information about the health and healthcare use of patients with a health condition or treatment, often contain patient-reported outcomes (PROs) that provide insights about the patient's perspectives on their health. Missing data can affect the value of PRO data for healthcare decision-making. We compared the precision and bias of several missing data methods when estimating longitudinal change in PRO scores.MethodsThis research conducted analyses of clinical registry data and simulated data. Registry data were from a population-based regional joint replacement registry for Manitoba, Canada; the study cohort consisted of 5631 patients having total knee arthroplasty between 2009 and 2015. PROs were measured using the 12-item Short Form Survey, version 2 (SF-12v2) at pre- and post-operative occasions. The simulation cohort was a subset of 3000 patients from the study cohort with complete PRO information at both pre- and post-operative occasions. Linear mixed-effects models based on complete case analysis (CCA), maximum likelihood (ML) and multiple imputation (MI) without and with an auxiliary variable (MI-Aux) were used to estimate longitudinal change in PRO scores. In the simulated data, bias, root mean squared error (RMSE), and 95% confidence interval (CI) coverage and width were estimated under varying amounts and types of missing data.ResultsThree thousand two hundred thirty (57.4%) patients in the study cohort had complete data on the SF-12v2 at both occasions. In this cohort, mixed-effects models based on CCA resulted in substantially wider 95% CIs than models based on ML and MI methods. The latter two methods produced similar estimates and 95% CI widths. In the simulation cohort, when 50% of the data were missing, the MI-Aux method, in which a single hypothetical auxiliary variable was strongly correlated (i.e., 0.8) with the outcome, reduced the 95% CI width by up to 14% and bias and RMSE by up to 50 and 45%, respectively, when compared with the MI method.ConclusionsMissing data can substantially affect the precision of estimated change in PRO scores from clinical registry data. Inclusion of auxiliary information in MI models can increase precision and reduce bias, but identifying the optimal auxiliary variable(s) may be challenging.
引用
收藏
页数:9
相关论文
共 35 条
[1]  
[Anonymous], 2004, Multiple imputation for nonresponse in surveys
[2]   Practical and statistical issues in missing data for longitudinal patient-reported outcomes [J].
Bell, Melanie L. ;
Fairclough, Diane L. .
STATISTICAL METHODS IN MEDICAL RESEARCH, 2014, 23 (05) :440-459
[3]   Nearest neighbor imputation algorithms: a critical evaluation [J].
Beretta, Lorenzo ;
Santaniello, Alessandro .
BMC MEDICAL INFORMATICS AND DECISION MAKING, 2016, 16
[4]   A comparison of inclusive and restrictive strategies in modern missing data procedures [J].
Collins, LM ;
Schafer, JL ;
Kam, CM .
PSYCHOLOGICAL METHODS, 2001, 6 (04) :330-351
[5]   MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM [J].
DEMPSTER, AP ;
LAIRD, NM ;
RUBIN, DB .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01) :1-38
[6]   Weight loss, adverse events, and loss to follow-up after gastric bypass in young versus older adults: A Scandinavian Obesity Surgery Registry study [J].
Dreber, Helena ;
Thorell, Anders ;
Torgerson, Jarl ;
Reynisdottir, Signy ;
Hemmingsson, Erik .
SURGERY FOR OBESITY AND RELATED DISEASES, 2018, 14 (09) :1319-1326
[7]   Analyzing Incomplete Item Scores in Longitudinal Data by Including Item Score Information as Auxiliary Variables [J].
Eekhout, Iris ;
Enders, Craig K. ;
Twisk, Jos W. R. ;
de Boer, Michiel R. ;
de Vet, Henrica C. W. ;
Heymans, Martijn W. .
STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL, 2015, 22 (04) :588-602
[8]  
Franklin PD, 2018, NEJM CATAL
[9]   Addressing Missing Data in Patient-Reported Outcome Measures (PROMS): Implications for the Use of PROMS for Comparing Provider Performance [J].
Gomes, Manuel ;
Gutacker, Nils ;
Bojke, Chris ;
Street, Andrew .
HEALTH ECONOMICS, 2016, 25 (05) :515-528
[10]   Missing data imputation using statistical and machine learning methods in a real breast cancer problem [J].
Jerez, Jose M. ;
Molina, Ignacio ;
Garcia-Laencina, Pedro J. ;
Alba, Emilio ;
Ribelles, Nuria ;
Martin, Miguel ;
Franco, Leonardo .
ARTIFICIAL INTELLIGENCE IN MEDICINE, 2010, 50 (02) :105-115