Letters Section Editor: Jody W. Zylke, MD, Senior Editor.
Author Affiliations: Shock Trauma Center Department of Anesthesiology, University of Maryland School of Medicine, Baltimore (Dr Galvagno; firstname.lastname@example.org); and Johns Hopkins University Bloomberg School of Public Health (Ms Baker) and Division of Acute Care Surgery, Trauma, Emergency Surgery, and Critical Care, Johns Hopkins University School of Medicine (Dr Haider), Baltimore, Maryland.
In Reply: Dr Delgado and colleagues are concerned about bias resulting from not including transport times in the analysis. Although time was missing for 57.8% of data, we performed a sensitivity analysis using total prehospital time as a variable to create a separate propensity-matched data set. The effect estimate did not change considerably (odds ratio, 1.07 [95% CI, 1.04-1.17]; P < .001 for patients transported to level I trauma centers). We did not present the results from this analysis in the article because with such a high proportion of missing data, the results were likely to be biased. Nevertheless, the absence of distance and time information is one limitation in our study. The propensity analysis also addresses the concern about unknown assignment mechanism. It was not the case that patients transported by ground EMS—of which a significant proportion may have been transported from urban areas—were more severely injured and had higher in-hospital death rates as reported by Rogers et al.1 Unadjusted severity was significantly higher for patients transported by helicopter: 24.6% of patients transported by helicopter had an ISS of 25 to 34 vs 19% of patients transported by ground (P < .001). The unadjusted in-hospital death rate was significantly higher for patients transported by helicopter (12.6%) compared with patients transported by ground (11%; P < .001).
Galvagno SM, Baker SP, Haider AH. Helicopter vs Ground Transportation for Patients With Trauma—Reply. JAMA. 2012;308(6):563-565. doi:10.1001/jama.2012.7778