You are here
Multiple imputation of baseline data in the cardiovascular health study.
Tuesday,2012-11-06 12:06 America/Los_Angeles — eenright
| Title | Multiple imputation of baseline data in the cardiovascular health study. |
| Publication Type | Journal Article |
| Year of Publication | 2003 |
| Authors | Arnold, AM, Kronmal, RA |
| Journal | Am J Epidemiol |
| Volume | 157 |
| Issue | 1 |
| Pagination | 74-84 |
| Date Published | 2003 Jan 01 |
| ISSN | 0002-9262 |
| Keywords | African Continental Ancestry Group, Aged, Aged, 80 and over, Analysis of Variance, Cardiovascular Diseases, Cause of Death, Cohort Studies, Data Collection, Data Interpretation, Statistical, Epidemiologic Studies, Female, Humans, Linear Models, Male, Mathematical Computing, Models, Statistical, Predictive Value of Tests, Proportional Hazards Models, Risk Factors, Software, Survival Analysis, United States, Ventricular Remodeling |
| Abstract | <p>Most epidemiologic studies will encounter missing covariate data. Software packages typically used for analyzing data delete any cases with a missing covariate to perform a complete case analysis. The deletion of cases complicates variable selection when different variables are missing on different cases, reduces power, and creates the potential for bias in the resulting estimates. Recently, software has become available for producing multiple imputations of missing data that account for the between-imputation variability. The implementation of the software to impute missing baseline data in the setting of the Cardiovascular Health Study, a large, observational study, is described. Results of exploratory analyses using the imputed data were largely consistent with results using only complete cases, even in a situation where one third of the cases were excluded from the complete case analysis. There were few differences in the exploratory results across three imputations, and the combined results from the multiple imputations were very similar to results from a single imputation. An increase in power was evident and variable selection simplified when using the imputed data sets.</p> |
| DOI | 10.1093/aje/kwf156 |
| Alternate Journal | Am J Epidemiol |
| PubMed ID | 12505893 |
| Grant List | N01-HC-15103 / HC / NHLBI NIH HHS / United States N01-HC-35129 / HC / NHLBI NIH HHS / United States N01-HC-85079 / HC / NHLBI NIH HHS / United States N01-HC-85086 / HC / NHLBI NIH HHS / United States |