Survey Unveils Scope of Unreported Radiation Errors

Staying Out of the New York Times

Nick Mulcahy

October 07, 2011

October 7, 2011 (Miami Beach, Florida) — Only a subset of near misses and errors that are observed by radiation oncology team members are reported, according to a study presented here at the American Society for Radiation Oncology (ASTRO) 53rd Annual Meeting.

The study comes from an anonymous, multi-institutional survey that was distributed to clinical staff at 4 academic radiation oncology centers, all of which have in-house incident reporting systems.

Gaps between observed and reported events were noted at each center. In the survey, staff collectively admitted to failing to sometimes report 36.4% of minor near misses, 14.2% of major near misses, 24.5% of minor errors, and 4% of major errors.

Interestingly, nearly all of the survey participants (96%) said that they had a responsibility to report these events.

Response to the survey was high (n = 268; 76% of total staff), and included participation from all professional groups — attending physicians, residents, dosimetrists, physicists, radiation therapists, and nurses — said the study's lead author, Kendra Harris, MD, MSc.

Dr. Harris is a radiation oncology resident at Johns Hopkins Medical Center in Baltimore, Maryland, which was one of the participating institutions. The others were North Shore-Long Island Jewish Health System, New Hyde Park, New York; Washington University School of Medicine, St. Louis, Missouri; and the University of Miami, Florida.

The subject of radiation errors and their sometimes disastrous outcomes was clearly on the minds of ASTRO organizers this year. The meeting program included a panel entitled Quality Assurance in Radiosurgery: Staying Out of The New York Times. The reference is to a 2010 front-page story that described the "wonders" of radiation in oncology but also detailed case histories of its "brutality."

In their survey, the most commonly cited barrier to reporting errors was concern about professional sanctions, the authors report.

Significant differences between the professional groups were seen in terms of concern about getting colleagues in trouble (P = .0084), liability (P = .0248), and embarrassment in front of colleagues (P = .0027). Notably, all of these concerns were most frequently reported by attending physicians and residents.

"It seems like embarrassment is a big barrier [to reporting]," Dr. Harris told Medscape Medical News.

Clinicians tend to be embarrassed.

Another oncologist at the meeting agreed. "Clinicians tend to be embarrassed" about errors and near misses, said Christopher Koprowski, MD, chair of radiation oncology at the Helen Graham Cancer Center of Christiana Care Health System in Newark, Delaware. He was also "not surprised" by the survey's findings.

"Our culture is judgmental. It's not a learning culture yet, but we're headed in that direction," Dr. Koprowski told Medscape Medical News in an interview after having seen the poster presentation. Establishing a learning culture is part of the answer to reducing errors, he said.

All the professional groups in the survey reported that communication failures were the most common source of error in their departments. In the participant feedback section, the most common sources of concern about errors were setup errors (n = 15), complex stereotactic treatments (n = 10), and computer-related events (n = 13).

Tools for Improvement

Another study presented at ASTRO picked up where the survey by Dr. Harris's team left off.

That study, entitled Patient Safety in Radiation Oncology: Tools for Improvement, is also from Johns Hopkins, and looks at 2 safety-improvement mechanisms that were implemented in the institution's Department of Radiation Oncology in 2006/07 — an incident reporting system and a failure mode and effects analysis (FMEA).

The incident reporting system has logged more than 200 reports of near-miss errors to date, report Eric Ford, PhD, assistant professor of medical physics, and colleagues. "FMEA is a prospective error-prevention methodology in which failure modes are identified and scored according to severity, occurrence rate, and detectability, and are combined into a risk priority number," they explain.

They note that "improvements are warranted" in the field, saying that "the rate of serious injury during radiotherapy is approximately 1000 times higher than in industries such as commercial aviation and modern anesthesiology, which are often cited as examples of complex but ultrasafe enterprises."

Dr. Ford's team found that 97% of all near-miss errors could have been caught by at least 1 of 15 quality-control mechanisms in the FMEA. Those mechanisms include items such as electronic safeguards and checklists (e.g., obtaining patient consent, physicist review of chart, review of orders). The poster listed a number of errors and showed which of the 15 quality-control checks would catch the problem. For example, a "wrong dose" would be caught by both "physics weekly chart check" and "checklist."

"All of this stuff in the aggregate helps," said Dr. Koprowski about the quality-control mechanisms. "Still, 3% of the incidents cannot be detected by quality-control checks."

"This is what we all do," he added about the quality-control measures. Dr. Koprowski also explained that at his center, approximately 100,000 fields are done annually that require manual positioning by staff. "This leads to human error," he noted.

Dr. Harris has disclosed no relevant financial relationships. One coauthor reports receiving a research grant from Varian Medical; another coauthor reports being a shareholder in ViewRay and Fulcrum Medical. Dr. Ford and coauthors have disclosed no relevant financial relationships.

American Society for Radiation Oncology (ASTRO) 53rd Annual Meeting: Abstracts 2711 and 2712. Presented October 3, 2011.

processing....