Of the 21 studies investigating the relationship between obesity and mortality,17- 19,21- 24,26,29- 33,35- 39,41,4210 studies reported obesity as a significant predictor of death18,21,22,24,29,31,33,36,40,42through multivariate modeling by logistic regression or a Cox proportional hazards model. One half of the cancer and stem cell transplant studies (n = 5) found obesity to be a significant predictor of mortality.22,24,36,40,42 None of the elective surgery studies (n = 4) investigated mortality as an outcome,28,34,43,44 probably owing to the low mortality rates associated with these procedures. In 11 studies, there were no significant associations between obesity and mortality.17,19,23,26,30,32,35,37- 39,41Study quality, as assessed by the Newcastle-Ottawa Scale, was higher, on average, among studies reporting obesity as a significant predictor of mortality. In 3 studies, the incidence of mortality was compared by weight group only,17,26,41 with no analysis for covariates. Only 1 study was prospective.18 Most studies were large cohort and multicenter studies. Most analyses controlled for covariates and confounders, although the retrospective data collection limited the availability of relevant confounding data in several studies.