Patient Safety Event Reports from Family Physicians, Staff and Patients in Family Practice Offices: A Pilot Study

Presenter: John Hickner, MD, MS

Institution: American Academy of Family Physicians National Research Network, Leawood KS

Co-Authors: Bob Phillips, MD, MSPH; Deborah Graham, MSPH; Jennifer Kappus; Susan Dovey, PhD; James Galliher, PhD; Nancy Elder, MD, MSPH

Introduction: Reporting patient safety events (medical errors) confidentially or anonymously to a central database for analysis is an important initial step in improving patient safety. Several investigators have reported studies of errors reported by family physicians, but there are no published studies of errors reported simultaneously by family physicians, their staff and their patients. We hypothesized that each of these groups would identify and report different kinds of patient safety events to an anonymous reporting system.

Methods: During a ten-week period from April to July 2003, we solicited error reports from physicians, staff and patients of five family physician offices and five family practice residency clinics. The 10 clinics represented rural, suburban, and urban locations throughout the United States. All practices had physicians who were members of the American Academy of Family Physicians (AAFP) National Research Network. From the 10 participating sites, approximately 401 physicians and staff either signed a consent form to participate or, for the sites that did not require consents, attended the study training session. This represented 86 percent of staff and physicians in these offices.

Institutional Review Board approval was obtained from each of the sites, a process that took three months to accomplish.

Physicians and staff submitted reports either via mail forms, or via the AAFP web-based errors reporting system, AAFP Patient Safety Reports. Patients were invited to use either of these methods or to submit reports verbally using an automated voice response system via a toll free phone number.

We used a taxonomy we developed in two prior medical errors reporting study to classify the kinds of reports we received. (Dovey, 2002) Errors were coded by three research team members with additional group reviews of coding at regular intervals by the investigators. This is a report of the first 374 errors reported.

Results: During the 10-week reporting period we received 729 reports. Of these reports, 325 reports were from physicians, 404 from staff, and 125 from patients. Although error reports were about similar events, we noted a difference in the reporting pattern among physicians, staff and patients. Physicians were more likely to report chart completeness and availability errors and medication errors. Staff reported chart completeness and availability errors, but they were more likely than physicians to report appointment errors. Despite active solicitation for medical errors, patients reported very few medical errors. Examples of types of patient reports include errors related to patient flow, maintenance of the physical building, and communications with patients.

Discussion: We found that, with active solicitation, family physicians and their staff will report medical errors to an anonymous national database operated by their primary professional organization, the American Academy of Family Physicians. Despite some differences, the error reports we received from staff are generally similar to those we received from physicians in this study and the family physicians who participated in our prior studies of medical errors. We must find a better method of soliciting error reports from patients.