This page is a repository of odds and ends as well as a place to store content that I will link to on occasion.

Posted 1/16/2017

Published as a Rapid Response to the Makary BMJ Paper estimating 250,000 preventable deaths in the US each year attributable to medical error

While Makary and Daniel should be commended for drawing attention to the issue of medical errors in their Analysis paper in the BMJ (1), there are several issues that should be addressed regarding their statistical methods and reporting of their primary outcome. In Table 1, they list the Healthgrades study on the top line (2). This is a white paper which has not undergone any peer review that I am able to locate. In table 1, this Healthgrades study listed the “% of admissions with a preventable lethal adverse outcome” and “Extrapolation to 2013 US admissions” to be 0.71 and 251,454, respectively. The last line on that same table lists those exact same numbers for the “Point estimate from all data” for the same outcomes. Given that the Healthgrades study analyzed 37 million patients compared to 2,341, 838, and 795 for the other three studies (3,4,5), 99.98% of the statistical power from this analysis came from this one Healthgrades study.

Essentially, the authors are not reporting any new information in this analysis. They are simply republishing the results of a non-peer reviewed white paper. While these studies may very well be the combined total of all of our published literature on this subject, the authors do not make a note that practically all of their statistical power came from this one database study. As others have already pointed out online (6) the other three studies (which provided much greater detail and rigor than the Healthgrades paper) combined for a grand total of 35 patient deaths. Should we really be extrapolating studies with a total of 35 patients to represent all medical errors in a country with over 323 million people? (7)

Medical errors are certainly an important area to address with rigorous research. However, we must be held to high statistical standards and avoid such large and sweeping extrapolations. The media will not understand the nuance of the fact that the main goal of this analysis was to point out the lack of research into medical errors. They will only see that 250,000 patients die per year in the US due to medical errors. This is needless sensationalism and needs to be countered with more rigorous research so as to not scare the patients who entrust their healthcare and their lives to us.

Stephen Carroll, DO
Emergency Medicine Physician
US Army
San Antonio Military Medical Center
Fort Sam Houston, TX
Twitter: @embasic

1. Makary MA, Daniel M. Medical error-the third leading cause of death in the US. BMJ. 2016;353:i2139.

2.HealthGrades quality study: patient safety in American hospitals. 2004. http://www.

3.Department of Health and Human Services. Adverse events in hospitals: national incidence among Medicare beneficiaries. 2010.

4. Classen D, Resar R, Griffin F, et al. Global “trigger tool” shows that adverse events in hospitals may be ten times greater than previously measured. Health Aff 2011;30:581-9doi: 10.1377/hlthaff.2011.0190.

5. Landrigan CP, Parry GJ, Bones CB, Hackbarth AD, Goldmann DA, Sharek PJ. Temporal trends in rates of patient harm resulting from medical care. N Engl J Med 2010;363:2124-34. doi:10.1056/NEJMsa1004404 pmid:21105794.

6. Are there really 250,000 preventable deaths per year in US Hospitals? Skeptical Scalpel Blog. Accessed May 8,, 2016. Available:…

7. Population Clock- Accessed May 8, 2016. Available: