[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Views 94
Citations 0
Comment & Response
August 2016

Question About the Analysis of Missing Data and Identification of an Error

Author Affiliations
  • 1Mikurashima Clinic, Mikurashima, Tokyo, Japan
  • 2Department of Urology, Kagawa University, Kida, Kagawa, Japan
  • 3Department of Plastic and Reconstructive Surgery, Saga University Hospital, Saga, Japan
JAMA Surg. 2016;151(8):781-782. doi:10.1001/jamasurg.2016.0127

To the Editor We read the study by Li et al1 with great interest and have been inspired to consider new prophylactic strategies against Clostridium difficile infection because the surgical patients in our departments are similar to those in their study. However, we have 2 questions for the authors.

First, the authors mentioned that missing data on preoperative demographics and clinical characteristics in the Veterans Affairs Surgical Quality Improvement Program were rare, but item 12(c) in the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement recommends that one should “[e]xplain how missing data were addressed.”2(p346) Precisely how much data were missing? We would like to know whether these missing data had the prior condition for multiple imputation3 and why they were missing.