What question did this study set out to answer?

To review the mechanisms of missing data in clinical epidemiological research and evaluate conventional and advanced analytical approaches for handling missing values, with a focus on multiple imputation.

March 1, 2017SHILAP Revista de lepidopterologíaOpen Access

Missing data and multiple imputation in clinical epidemiological research

View Full Paper

Ask AI

Bookmark

Authors

APAlma B PedersenThe University of Melbourne EMEllen M. MikkelsenFox Chase Cancer Center DCDeirdre Cronin‐FentonAarhus University

Discussion

Member takes

Implication

Methodological review evaluates missing data handling in clinical epidemiology, highlighting multiple imputation's ability to yield unbiased estimates under missing-at-random assumptions.

Key Points

To review the mechanisms of missing data in clinical epidemiological research and evaluate conventional and advanced analytical approaches for handling missing values, with a focus on multiple imputation.
Categorized missing data into missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR) frameworks.
Evaluated conventional strategies—including complete-case analyses, missing indicator methods, single value imputation, and sensitivity analyses—relative to multiple imputation implemented under MAR assumptions.
Conventional strategies such as complete-case analysis yield unbiased results only under the MCAR assumption, which is rarely met in clinical epidemiology, and typically produce less precise estimates.
Multiple imputation appropriately models the uncertainty of missing data under MAR assumptions, providing valid and unbiased association estimates that refine coefficients for both missing and completely observed variables.

Cite This Study

Pedersen et al. (2017) studied this question.

synapsesocial.com/papers/69d77ebadb9d5e1bf4b8b240 https://doi.org/10.2147/clep.s129785

View Full Paper

Ask AI

Bookmark

Implication

Methodological review evaluates missing data handling in clinical epidemiology, highlighting multiple imputation's ability to yield unbiased estimates under missing-at-random assumptions.

Missing data and multiple imputation in clinical epidemiological research

Discussion

Member takes

Implication

Key Points

Cite This Study

Missing data and multiple imputation in clinical epidemiological research

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider