Connect

Member Research and Reports

Member Research and Reports

UAB and Johns Hopkins Researchers Discuss the Role of Multiple Imputation

Dr. Peng Li, research associate in the department of biostatistics at the University of Alabama at Birmingham, and Dr. Elizabeth A. Stuart, professor in the departments of mental health; biostatistics; and health policy and management of the Johns Hopkins Bloomberg School of Public Health — along with co-author Dr. David B. Allison, distinguished professor and director of UAB’s Nutrition Obesity Research Center (NORC) and office of energetics — recently examined the role of multiple imputation (MI) in clinical trial results.

LiP_UAB_ASPPH

EStuart_Hopkins_ASPPH
[Photo: Dr. Peng Li (above) and Dr. Elizabeth A. Stuart]

Acknowledging that investigators regularly encounter incomplete and insufficient data in research — and discussing the causes of such missing data — the authors note that it is often necessary to substitute estimated values for missing data in lieu of disregarding potentially valuable information gleaned from studies. They review MI’s benefits and limitations, recognizing that any approach to handling missing data will involve assumptions. Therefore, the researchers also offer cautions to consider when reviewing results based on the method.

“Multiple Imputation: A Flexible Tool for Handling Missing Data” was published in November in The Journal of the American Medical Association.

Journal article: http://jama.jamanetwork.com/article.aspx?articleid=2468879