Hospitals' efforts to improve patient safety rely on several methods of monitoring and evaluating the occurrence of adverse events: including incident reports from members of the health care team, automated surveillance of clinical data, and review of medical records.
A group of Massachusetts researchers report in the July 15 Annals of Internal Medicine that surveying patients about their experiences can provide additional important information. They found that nearly 23 percent of patients reported experiencing a complication during or immediately after their hospital stay, compared with about 11 percent of patients whose adverse events were identified through medical record review.
"Every approach to monitoring adverse events or complications has its limitations, even record review, which has been regarded as the "gold standard," says lead author Joel Weissman, PhD, of the Massachusetts General Hospital (MGH) Institute for Health Policy. "Our research demonstrates that patients themselves can be a valuable source of information about unexpected complications that occur as a result of medical care, both during their hospital stay and after they are discharged."
While many hospitals regularly survey patients after discharge, those surveys are usually focused on patients' satisfaction with their care and not on whether they experienced injuries or complications. The current study was designed to evaluate whether patients can accurately report adverse events they experienced, the types of events patients were most likely to report and how well patient reports matched what in the medical record. It consisted of two primary phases: a telephone survey of patients admitted to 16 Massachusetts hospitals during six months in 2003 and a review of the medical records of survey participants who gave written permission for the review.
Weissman explains, "Adverse events are complications or injuries to patients -- some of which may due to preventable errors, and some which are neither preventable nor error-related. For example, an allergic reaction to a drug is an adverse drug event. If the allergy was known, administering that drug was a preventable error. But if the allergy was unknown, it was not preventable, although still an adverse event. Patients may be more likely to know about complications than about errors in their care."
About 2,600 patients participated in the telephone survey, which took place 6 to 12 months after hospital discharge. The 20-minute interview assessed several aspects of their clinical care and specifically asked about any negative effects, complications or injuries they had experienced during or after their hospitalization. Events patients reported were subsequently reviewed by two physician co-authors, who evaluated and scored them by severity and preventability.
Medical records review, approved by almost 1,000 patients, was conducted by nurses trained to identify adverse events according to specific criteria. Two different physician reviewers classified and scored the medical record events.
The study identified 281 patients who experienced some sort of adverse event -- 229 were identified in the interview and 105 in the medical record, but only 53 were noted by both sources. Less than 10 percent of events identified by either method were serious or life-threatening, and under a third were determined to be probably or definitely preventable. Most of the events that took place after patients were discharged from the hospital were related to their care but did not become evident until they left the hospital.
"This study shows very clearly that additional tools can and should be added to hospitals' efforts to evaluate patient safety," says study co-author Nancy Ridley, MS, of the Massachusetts Department of Public Health. "Patient involvement plays an important role in the safety and delivery of health care services, and we encourage hospitals to enlist patients in these efforts wherever possible."
Adds Saul Weingart, MD, PhD, of Dana-Farber Cancer Institute, a study co-author and one of the physicians who reviewed patient-reported events, "We need to learn more about how patients can help clinicians ensure safe care in the hospital and in ambulatory settings. It's pretty clear that they can teach us important things about improving patient safety, if we only ask them."
While conducting the kind of telephone surveys used in this study might be costly, the researchers note, so is medical record review. Adding safety-oriented questions to existing satisfaction-focused patient questionnaires might be a first step, but utilizing multiple methods to track adverse events will probably give the most accurate results. "What is most important is to learn from our mistakes and determine how to prevent them in the future," Weissman says. "We hope that our research will give patient safety advocates another tool to accomplish that aim."
An associate professor of Medicine at Harvard Medical School, Weissman is currently on leave from his MGH position and working as a senior health policy advisor in the Massachusetts Executive Office of Health and Human Services. He is also a professor of Family and Community Medicine at the University of Massachusetts Medical School.
Additional co-authors of the Annals report are Eric C. Schneider, MD, MSc, and Arnold M. Epstein, MD, MA, Harvard School of Public Health; Jo Ann David-Kasdan, RN, MS, and Sandra Feibelmann, MPH, MGH; and Constantine Gatsonis, PhD, Brown University. At the time the study was conducted, co-authors Catherine Annas, JD, and Leslie Kirle, MPH, were with the Mass. Department of Public Health and the Massachusetts Hospital Association, respectively. The study was supported by a Cooperative Agreement from the Agency for Healthcare Research and Quality to the Mass. Department of Public Health.
Cite This Page: