Unpredictable bias when using the missing indicator method or complete case analysis for missing confounder values: an empirical example

M.J. Knol, K.J.M. Janssen, A.R.T. Donders, A.C.G. Egberts, E.R. Heerdink, D.E. Grobbee, K.G.M. Moons, M.I. Geerlings

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

OBJECTIVE: Missing indicator method (MIM) and complete case analysis (CC) are frequently used to handle missing confounder data. Using empirical data, we demonstrated the degree and direction of bias in the effect estimate when using these methods compared with multiple imputation (MI).

STUDY DESIGN AND SETTING: From a cohort study, we selected an exposure (marital status), outcome (depression), and confounders (age, sex, and income). Missing values in "income" were created according to different patterns of missingness: missing values were created completely at random and depending on exposure and outcome values. Percentages of missing values ranged from 2.5% to 30%.

RESULTS: When missing values were completely random, MIM gave an overestimation of the odds ratio, whereas CC and MI gave unbiased results. MIM and CC gave under- or overestimations when missing values depended on observed values. Magnitude and direction of bias depended on how the missing values were related to exposure and outcome. Bias increased with increasing percentage of missing values.

CONCLUSION: MIM should not be used in handling missing confounder data because it gives unpredictable bias of the odds ratio even with small percentages of missing values. CC can be used when missing values are completely random, but it gives loss of statistical power.

Original languageEnglish
Pages (from-to)728-36
Number of pages9
JournalJournal of Clinical Epidemiology
Volume63
Issue number7
DOIs
Publication statusPublished - Jul 2010

Keywords

  • Age Factors
  • Bias
  • Biomedical Research
  • Cohort Studies
  • Confounding Factors (Epidemiology)
  • Data Collection
  • Depressive Disorder
  • Female
  • Humans
  • Income
  • Male
  • Marital Status
  • Middle Aged
  • Outcome Assessment (Health Care)
  • Research Design
  • Sex Factors
  • Journal Article
  • Research Support, Non-U.S. Gov't

Fingerprint

Dive into the research topics of 'Unpredictable bias when using the missing indicator method or complete case analysis for missing confounder values: an empirical example'. Together they form a unique fingerprint.

Cite this