Handling missing data in clinical research
- PMID: 36150546
- DOI: 10.1016/j.jclinepi.2022.08.016
Handling missing data in clinical research
Abstract
Because missing data are present in almost every study, it is important to handle missing data properly. First of all, the missing data mechanism should be considered. Missing data can be either completely at random (MCAR), at random (MAR), or not at random (MNAR). When missing data are MCAR, a complete case analysis can be valid. Also when missing data are MAR, in some situations a complete case analysis leads to valid results. However, in most situations, missing data imputation should be used. Regarding imputation methods, it is highly advised to use multiple imputations because multiple imputations lead to valid estimates including the uncertainty about the imputed values. When missing data are MNAR, also multiple imputations do not lead to valid results. A complication hereby is that it not possible to distinguish whether missing data are MAR or MNAR. Finally, it should be realized that preventing to have missing data is always better than the treatment of missing data.
Copyright © 2022 The Author(s). Published by Elsevier Inc. All rights reserved.
Similar articles
-
Mechanism-aware imputation: a two-step approach in handling missing values in metabolomics.BMC Bioinformatics. 2022 May 16;23(1):179. doi: 10.1186/s12859-022-04659-1. BMC Bioinformatics. 2022. PMID: 35578165 Free PMC article.
-
Predictors of clinical outcome in pediatric oligodendroglioma: meta-analysis of individual patient data and multiple imputation.J Neurosurg Pediatr. 2018 Feb;21(2):153-163. doi: 10.3171/2017.7.PEDS17133. Epub 2017 Dec 1. J Neurosurg Pediatr. 2018. PMID: 29192869 Review.
-
Missing data and multiple imputation in clinical epidemiological research.Clin Epidemiol. 2017 Mar 15;9:157-166. doi: 10.2147/CLEP.S129785. eCollection 2017. Clin Epidemiol. 2017. PMID: 28352203 Free PMC article.
-
Multiple imputation for non-response when estimating HIV prevalence using survey data.BMC Public Health. 2015 Oct 16;15:1059. doi: 10.1186/s12889-015-2390-1. BMC Public Health. 2015. PMID: 26475303 Free PMC article.
-
Review: a gentle introduction to imputation of missing values.J Clin Epidemiol. 2006 Oct;59(10):1087-91. doi: 10.1016/j.jclinepi.2006.01.014. Epub 2006 Jul 11. J Clin Epidemiol. 2006. PMID: 16980149 Review.
Cited by
-
A Principled Approach to Characterize and Analyze Partially Observed Confounder Data from Electronic Health Records.Clin Epidemiol. 2024 May 21;16:329-343. doi: 10.2147/CLEP.S436131. eCollection 2024. Clin Epidemiol. 2024. PMID: 38798915 Free PMC article.
-
Optimizing perinatal wellbeing in pregnancy with obesity: a clinical trial with a multi-component nutrition intervention for prevention of gestational diabetes and infant growth and neurodevelopment impairment.Front Med (Lausanne). 2024 Apr 12;11:1339428. doi: 10.3389/fmed.2024.1339428. eCollection 2024. Front Med (Lausanne). 2024. PMID: 38681052 Free PMC article.
-
Moderators of treatment effect of non-steroidal anti-inflammatory drugs for patients with (sub) acute low back pain: Protocol for a systematic review with individual participant data meta-analysis.MethodsX. 2024 Apr 11;12:102713. doi: 10.1016/j.mex.2024.102713. eCollection 2024 Jun. MethodsX. 2024. PMID: 38660043 Free PMC article.
-
Fatigue Item Response among Hemoglobin-Normalized Patients with Paroxysmal Nocturnal Hemoglobinuria: PEGASUS Trial Results at 16 and 48 Weeks.J Clin Med. 2024 Mar 15;13(6):1703. doi: 10.3390/jcm13061703. J Clin Med. 2024. PMID: 38541928 Free PMC article.
-
Comparison of the effects of imputation methods for missing data in predictive modelling of cohort study datasets.BMC Med Res Methodol. 2024 Feb 16;24(1):41. doi: 10.1186/s12874-024-02173-x. BMC Med Res Methodol. 2024. PMID: 38365610 Free PMC article.
MeSH terms
LinkOut - more resources
Full Text Sources
Miscellaneous