Two physicians at Penn State Milton S. Hershey Medical Center have started an online petition to get The BMJ to retract a study published in May that found medical errors to be the No. 3 cause of death in the U.S.
Shyam Sabat, MD, a neurologist, and Virginia Hall, MD, an OB-GYN, co-signed the petition on change.org. Dr. Sabat also wrote a post on the Pennsylvania Medical Society's blog, calling the study "a shoddy piece of scientific and statistical work."
The original study, written by Martin Makary, MD, with Baltimore-based Johns Hopkins, was a meta-analysis of four studies that analyzed medical death rate data from 2000 to 2008. However, according to Dr. Sabat, one of the studies was much larger than the other three. "These other three studies don't have enough statistical power to be clubbed with the first study," he wrote.
Additionally, Dr. Sabat took issue with the fact that the study only looked at Medicare population mortality rates and then applied it to all U.S. inpatient admissions "without any correction of any form."
The blog continues: "How a reputed group such as the BMJ could not see through these simple but outrageous statistical blunders is anyone's guess." He ends the post with a demand for a retraction and an apology.
As of this article's publication, 169 people have signed the change.org petition.
In a statement emailed to Penn Live, a BMJ editor stood by the study. "We do not believe there are any grounds to retract the paper, and are not considering this course of action."
Dr. Makary also stood by his study and the methods used, stating that "death from medical care gone wrong should be measured rigorously" but that a strong data collection infrastructure to do so "does not exist."
In a statement emailed to Becker's, he continued, "In the meantime, the science of safety relies on estimates derived from the best scientific studies in the literature. The most frequently cited incidence rate for medical error has been the 1999 IOM 'To Err is Human' estimate that uses data from 1984 and 1992. While no estimate is perfect, we offer a more updated estimate that uses better scientific studies, including recent studies from the New England Journal of Medicine, Health Affairs, and the independent Medicare OIG report. Other studies, such as the analysis by John James in the Journal of Patient Safety, suggest that our estimate may be too low."
Note: This story was updated July 11 to include Dr. Makary's statement.