Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls

By J. Sterne, Ian R. White, J. Carlin, Michael Spratt, Patrick Royston, Michael G. Kenward, Angela M. Wood, James R. Carpenter. British medical journal, 2009

6,901 citations recorded by the source corpus.

Most studies have some missing data. Jonathan Sterne and colleagues describe the appropriate use and reporting of the multiple imputation approach to dealing with them

Subfields: Sociology and Political Science, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/7b0365a2e7ca7aaecd6ed55b78b544e06d6076d6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Sociology and Political Science

In Statistics and Probability

Citations in the corpus

Prognosis Research Strategy (PROGRESS) 3: Prognostic Model Research (2013, background): “Important issues to be addressed include dealing with missing data [26], modelling continuous prognostic factors [27,28], the complexity of the model [29], and checking the model assumptions.”
Multiple imputation using chained equations: Issues and guidance for practice (2011, background): “Little and Rubin’s framework [1] is often used to classify the missing data as being (i) missing completely at random (MCAR—the probability of data being missing does not depend on the observed or unobserved data), (ii) missing at random (MAR—the probability of data being missing does not depend on ”
External validation of multivariable prediction models: a systematic review of methodological conduct and reporting (2014, background): “The majority (64%) of studies were assumed to have conducted complete-case analyses to handle missing values, despite methodological guidance to do the contrary [40-44].”
Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): A randomised, placebo-controlled trial (2012, neutral): “All analyses were on an intention-to-treat basis and, because almost all randomised patients were followed up, there was no need to use imputation methods for missing data.(11) The primary endpoint was all-cause mortality, and the observed reduction in mortality with tranexamic acid was both statist”
Comparison of Performance of Data Imputation Methods for Numeric Dataset (2019, methodology): “MICE is increasingly popular method for doing multiple imputations (Patric and White 2011; Sterne et al. 2009; van Buuren and Groothuis-Oudshoorn 2011; White, Royston, and Wood 2011).”
The proportion of missing data should not be used to guide decisions on multiple imputation (2019, methodology): “Thus, where conclusions are being drawn from a small number of complete cases, we recommend sensitivity analyses to explore a range of plausible analysis and imputation models, as well as the impact of deviations from MAR [9,43].”
When and how should multiple imputation be used for handling missing data in randomised clinical trials – a practical guide with flowcharts (2017, methodology): “Missing data may seriously compromise inferences from randomised clinical trials, especially if missingness is not at random and if missing data are not handled appropriately [3, 4].”
Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies: The CHARMS Checklist (2014, background): “Numerous recommendations for reporting missing data in medical research have been proposed [103,104,114,115].”