(Re-posted from www.ACR.org)

The more than 32,000 members of the ACR are committed to providing the highest quality patient care. As part of this process, the ACR constantly refines practice standards designed to eliminate all foreseeable medical errors in the face of advancing technology and increasing demand for medical imaging services.

The ACR welcomes input from other organizations that may aid in this endeavor. However, without careful and logical analysis, the incomplete, inaccurate information contained in a recent United States Pharmacopeia (USP) (http://www.usp.org) study regarding medication error rates in imaging facilities may unnecessarily alarm patients and may cause many patients who require imaging care to mistakenly avoid getting it.

The USP data were obtained by voluntary reports from radiology departments, hospitals, and other facilities. However, the report does not make clear how representative these voluntary data are of all radiology facilities in the United States or how the number of errors reported compares to the incredible number of imaging procedures performed annually.

The USP report assigns 2,030 total errors over 5 years (2000-2004) to radiology facilities, with 12% (242) of that number termed harmful. The report also states that less than 1% of the errors cited may have been connected to any sentinel event.

By comparison, approximately 2.5 billion (http://apps2.acr.org/mcpr/pr/med_err_stat.html) imaging procedures were performed nationwide during the course of the USP study. The USP error figures (even taken as is), when compared to the documented number of procedures annually, represent an overall medication error rate for radiology facilities of .00008%. (http://apps2.acr.org/mcpr/pr/med_err_stat.html) This number is more than 3,700 times better than the lowest hospitalwide medication error rate (.3%) shown by actual peer-reviewed studies cited in a recent Institute of Medicine's report on health care error rates.^1,2

"The ACR works for the day when all medical errors are eliminated. However, this report is deeply flawed and fails to relate the extremely low frequency of such errors in relation to the more than 570 million (http://apps2.acr.org/mcpr/pr/level_util.html) medical imaging procedures performed in the United States each year. This incredibly vague report does not provide evidence that imaging facilities operated by trained, certified physicians and certified nonphysician personnel present a significantly increased risk over other medical facilities in regard to medication usage," said James P. Borgstede, MD, chair of the ACR Board of Chancellors.

The USP report fails to delineate which medical specialists actually perform exams and incorrectly attributes nearly half of the 2,030 errors cited in the report to radiology. Cardiac catheterization labs are listed as the areas where the highest percentage (40%) of medication errors cited in the report occurred. The report attributes all 823 errors incurred in these labs to "radiology." While interventional radiology is a growing subspecialty of radiology, and the number of these procedures performed by radiologists is increasing, radiologists currently perform less than 1% of cardiac catheterization procedures nationwide.

The paper also lumps many events which may be unrelated to the actual performance of imaging procedures into its analysis. For example, injections that may have been ordered by physicians in other hospital departments, yet out of necessity (particularly in regards to emergency room situations) were carried out in the radiology department independent of the actual imaging process were included. This methodology further skews the results of the analysis.

"The massive flaws in the report's methodology and its failure to provide context as to the scope of any errors reported seriously undermine the study's conclusions. To take this report as evidence of systemic failures in radiology facilities which represent an increased risk to patients is not only inaccurate, but irresponsible and potentially dangerous," said Borgstede.

The College has previously implemented guidelines to address how the already extremely low number of medication errors that might occur in imaging departments can be reduced even further. The ACR looks forward to working with health care organizations and governmental bodies to further address this issue.

1. Kohn LT, Corrigan JM, Donaldson, eds. To Err Is Human. Washington, DC: National Academy Press (IOM) 2000: 33-35.
2. Leape et al 1998