Use of multiple imputation in the epidemiologic literature.

The authors attempted to catalog the use of procedures to impute missing data in the epidemiologic literature and to determine the degree to which imputed results differed in practice from unimputed results. The full text of articles published in 2005 and 2006 in four leading epidemiologic journals was searched for the text imput. Sixteen articles utilizing multiple imputation, inverse probability weighting, or the expectation-maximization algorithm to impute missing data were found. The small number of relevant manuscripts and diversity of detail provided precluded systematic analysis of the use of imputation procedures. To form a bridge between current and future practice, the authors suggest details that should be included in articles that utilize these procedures.

[1]  M. Marmot,et al.  Self-reported economic difficulties and coronary events in men: evidence from the Whitehall II study. , 2005, International journal of epidemiology.

[2]  R. Ness,et al.  Periconceptional multivitamin use reduces the risk of preeclampsia. , 2006, American journal of epidemiology.

[3]  S. Greenland.  Absence of confounding does not correspond to collapsibility of the rate ratio or rate difference. , 1996 .

[4]  R. Martorell,et al.  Maternal and childhood nutrition and later blood pressure levels in young Guatemalan adults. , 2005, International journal of epidemiology.

[5]  Kiros Berhane,et al.  Bayesian modeling of air pollution health effects with missing exposure data. , 2006, American journal of epidemiology.

[6]  T. Raghunathan,et al.  Education, income, occupation, and the 34-year incidence (1965-99) of Type 2 diabetes in the Alameda County Study. , 2005, International journal of epidemiology.

[7]  B. Eskenazi,et al.  Association of DDT and DDE with Birth Weight and Length of Gestation in the Child Health and Development Studies, 1959–1967 , 2005 .

[8]  W Vach,et al.  Biased estimation of the odds ratio in case-control studies due to the use of ad hoc methods of correcting for missing values for confounding variables. , 1991, American journal of epidemiology.

[9]  C. R. Weinberg,et al.  Imputation for Exposure Histories with Gaps, under an Excess Relative Risk Model , 1996, Epidemiology.

[10]  T. Cole,et al.  Statistical issues in life course epidemiology. , 2006, American journal of epidemiology.

[11]  J. Cerhan,et al.  Organochlorines in Carpet Dust and Non-Hodgkin Lymphoma , 2005, Epidemiology.

[12]  G. Tomlinson,et al.  The validity of the certification of manner of death by Ontario coroners. , 2006, Annals of epidemiology.

[13]  S Greenland,et al.  A critical look at methods for handling missing covariates in epidemiologic regression analyses. , 1995, American journal of epidemiology.

[14]  M. Boyle,et al.  Childhood and early adult predictors of risk of incident back pain: Ontario Child Health Study 2001 follow-up. , 2005, American journal of epidemiology.

[15]  Thomas Lumley,et al.  Ambient air pollution and asthma exacerbations in children: an eight-city analysis. , 2006, American journal of epidemiology.

[16]  Ken P Kleinman,et al.  Much Ado About Nothing , 2007, The American statistician.

[17]  A. Sigurdson,et al.  An application of a weighting method to adjust for nonresponse in standardized incidence ratio analysis of cohort studies. , 2005, Annals of Epidemiology.

[18]  S. Willich,et al.  Pragmatic randomized trial evaluating the clinical and economic effectiveness of acupuncture for chronic low back pain. , 2006, American journal of epidemiology.

[19]  Kaare Christensen,et al.  Age trajectories of grip strength: cross-sectional and longitudinal data among 8,342 Danes aged 46 to 102. , 2006, Annals of epidemiology.

[20]  J. Schwartz,et al.  Individual-level modifiers of the effects of particulate matter on daily mortality. , 2006, American journal of epidemiology.

[21]  Ian R White,et al.  Comparison of imputation and modelling methods in the analysis of a physical activity trial with missing outcomes. , 2004, International journal of epidemiology.

[22]  P. G. van der Velden,et al.  The importance of estimating selection bias on prevalence estimates shortly after a disaster. , 2006, Annals of epidemiology.