Study reviews discussion of medical errors at hospital conferences

By Liese Greensfelder on December 04, 2003

Doctors participating in internal medicine hospital conferences designed to review adverse medical events do not often discuss related medical errors, according to a study led by researchers from the San Francisco VA Medical Center (SFVAMC).

Study results showed that participants in surgery conferences were more likely than those in internal medicine conferences to discuss medical errors as errors and to attribute errors to a particular cause. Conference leaders for both groups missed opportunities to use explicit language in error discussion, according to the researchers.  The study appears in the December 3 issue of the Journal of the American Medical Association.

“The findings of the study point to a culture that has difficulty acknowledging and dealing with error,” says the study’s senior author, Seth Landefeld, MD, associate chief of staff for geriatrics at SFVAMC, professor and chief of geriatrics at University of California, San Francisco, and senior scholar in the Department of Veterans Affairs National Quality Scholars Program.

“It’s indicative that in our medical profession we are, by and large, thinking of ourselves within the framework of the individual actor and not stepping back and seeing the entire stage within which we operate. What we need to do is ask how we can better prepare this stage and help it function with as low an error rate as possible,” he says.

According to a report published by the Institute of Medicine in 2000, each year an estimated 44,000 to 98,000 people die in U.S. hospitals as a result of medical errors. Academic medical centers hold regular in-house meetings that are designed to address adverse events suffered by patients and to determine whether the events may have resulted from errors.

Protected by law from being used to bolster claims of medical malpractice, the aim of these morbidity and mortality conferences is to educate physicians and reduce or prevent recurrence of errors, according to the study authors.  What was largely unknown was to what extent adverse events and errors were actually discussed in these conferences, says Edgar Pierluissi, MD, the study’s lead author who was a Department of Veterans Affairs National Quality scholar and a fellow in the division of geriatrics at SFVAMC at the time of the study.

To find out how conference participants discussed errors, Pierluissi trained four internal medicine residents and five surgery residents to assess and rate case presentations at morbidity and mortality conferences. After listening to each case discussion, the observers determined whether an adverse event had occurred in the case, whether an error had occurred, whether the error caused the adverse event, and the scope of the discussion of the event. The observers attended 85 surgery department conferences and 66 medicine department conferences at four Bay Area teaching hospitals during 2000 and 2001.

The observers discovered several differences in the manner in which cases were treated at the two kinds of conferences. In surgery conferences, attended principally by surgeons, the focus was generally on individual cases, and audience participation was encouraged. At medicine conferences, attended principally by doctors of internal medicine, the focus was on presentations by invited speakers, with relatively little discussion of cases. Study results showed 72 percent of the cases presented at surgery conferences included adverse events and 37 percent of the cases presented at medicine conferences included adverse events.

The differences between the manner in which the groups discussed errors was also notable. In surgery conferences, when an error had caused an adverse event, the error was discussed as an error 77 percent of the time, while in medicine conferences, errors were discussed as such 48 percent of the time. Thirty-eight percent of errors that were discussed in medicine conferences were attributed to a particular cause, compared to 79 percent of errors discussed in surgery conferences. Four of the ten cases involving errors that were discussed in medicine conferences involved errors related to providers other than doctors of internal medicine, such as radiologists or surgeons.  Nine of the 80 cases involving errors discussed at surgery conferences involved providers other than surgeons.

“Despite the challenges of measuring something as hard to define as an error, we believe we have accurately measured what goes on in these conferences,” says Pierluissi. “The differences we found were such that a typical medicine conference attendee would listen to only seven error discussions each year, compared to a surgery conference attendee, who could expect to hear 44 error discussions in a year.”

The study’s authors speculate that many of the differences they found between medicine and surgery conferences may be due to the fact that the American College for Graduate Medical Education requires that surgery morbidity and mortality conferences present and discuss all deaths and complications that occur on a weekly basis. Internal medicine is not subjected to a similar requirement.

“There are many reasons that we don’t like to talk about errors,” Landefeld says. “In general, there’s a human tendency to not want to acknowledge mistakes.” But open discussions of errors can yield enormous benefits, he says. “Error reporting is the first step toward a systematic approach of error reduction. An open, non-blaming, non-punitive discussion of errors helps create a process through which people can identify the causes of errors and what we might do to prevent them in the future. We need to do everything possible to develop attitudes and behaviors to reduce errors.”

The four hospitals surveyed in the study were San Francisco General Hospital Medical Center, SFVAMC, UCSF Medical Center and Stanford University Medical Center. San Francisco General Hospital Medical Center and SFVAMC are both affiliated with UCSF.

UCSF has been a national leader in developing strategies to reduce medical errors, according to Robert Wachter, MD, chief of medical service at UCSF Medical Center and chair of UCSF’s Patient Safety Committee. “This study is an important step. The only way we are going to make progress in our epidemic of medical errors is by facing up to the reality of errors, and the very human tendency to shy away from discussing them. We need to come up with new approaches, and this study helps by showing us some of the key issues that need to be addressed.”

Yet, Wachter says, the conferences that were surveyed in the study took place three years ago, at a time when the medical community was just beginning to respond to the Institute of Medicine’s 2000 report.  “If the authors came back to these conferences today,” he says, “they would find that many of the changes they favor have already been instituted.”

In 2000, Wachter and Kaveh Shojania, MD, an assistant professor of medicine at UCSF, co-authored a widely-distributed federal report outlining evidence-based patient safety practices. The two are also editors of a new online medical errors journal. Sponsored by the federal Agency for Healthcare Research and Quality, the journal serves as a national web-based mortality and morbidity conference.

Other co-authors of the JAMA study are Melissa A. Fischer, MD, University of Massachusetts Medical School, Worcester, MA, and Andre R. Campbell, MD, associate professor of surgery at UCSF.  Pierluissi is currently medical director for performance improvement at San Mateo Medical Center in San Mateo, California.

The study was supported in part by grants to UCSF from the National Institute on Aging, the John A. Hartford Foundation, and Dartmouth College, and by a grant to Dartmouth College from the Pfizer Foundation.

Additional information:

Agency for Healthcare Research and Quality’s AHRQ WebM&M, an online journal and forum on patient safety and health care quality: AHRQ Web M & M

To Err is Human: Building a Better Health System (2000) Institute of Medicine: Institute of Medicine