Correspondence |

ResponseResponse FREE TO VIEW

Renee D. Stapleton, MD, PhD, FCCP; Andrew G. Day, MSc; Benjamin T. Suratt, MD, FCCP
Author and Funding Information

From the Department of Medicine (Drs Stapleton and Suratt), Division of Pulmonary and Critical Care Medicine, University of Vermont; and Clinical Evaluation Research Unit (Mr Day), Kingston General Hospital, and Department of Community Health and Epidemiology (Mr Day), Queen’s University.

Correspondence to: Renee D. Stapleton, MD, PhD, FCCP, Department of Medicine, University of Vermont College of Medicine, 149 Beaumont Ave, HSRF 222, Burlington, VT 05405; e-mail: renee.stapleton@uvm.edu

Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.

© 2012 American College of Chest Physicians

Chest. 2012;141(6):1638-1639. doi:10.1378/chest.12-0420
Text Size: A A A
Published online

To the Editor:

We thank Dr De Jong and colleagues for their interest in our article.1 Dr De Jong and colleagues argue that our finding that critically ill patients who are obese survive as often as their counterparts who are of normal weight is an example of the Simpson paradox, where the confounder severity of illness would reverse the link between obesity and mortality. We agree that one major limitation of our study, which is stated in the article, is that no severity of illness assessment tool specifically for obese patients exists, and any existing assessment tool, including APACHE (Acute Physiology and Chronic Evaluation) II, may not accurately reflect mortality risk in patients who are extremely obese because there may be hidden factors not accounted for in that assessment.1 We also addressed the potential confounding of severity of illness, stating that an alternate hypothesis is that patients who are extremely obese have a lower threshold for ICU admission compared with patients who are of normal weight, meaning that the disease severity is less than perceived. Furthermore, we agree that future development of an obesity-specific severity of illness tool would be welcome.

However, the existence of the Simpson paradox is not supported by data. Although we cannot preclude the theoretical possibility of the Simpson paradox, there is no evidence to support that it has (or has not) occurred in these data.

It is important to note that our observed associations do not imply causality. We concluded that neither obesity caused lower mortality in the ICU nor the association between obesity and outcome would persist if we were able to more accurately control for severity of illness. Nevertheless, our conclusion that critically ill patients who are extremely obese survive at least as often as patients of normal weight is a correct statement of association and not intended to imply causality or even association conditional on severity. We believe that our conclusions and observed associations have value even in the absence of causal inference because there are important policy and clinical implications regardless of a true causal link.

Finally, Dr De Jong and colleagues ask whether it would be possible to match a posteriori the cause of admission in addition to the multivariate analysis. We included primary admission diagnosis in our multivariate analysis; performing a matched analysis would not alter the results appreciably.

Martino JL, Stapleton RD, Wang M, et al. Extreme obesity and outcomes in critically ill patients. Chest. 2011;1405:1198-1206. [CrossRef] [PubMed]




Martino JL, Stapleton RD, Wang M, et al. Extreme obesity and outcomes in critically ill patients. Chest. 2011;1405:1198-1206. [CrossRef] [PubMed]
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543