Peter J. Fabri, MD, PhD, FACS
Associate Dean, Graduate Medical Education
University of South Florida College of Medicine
Tampa, FL

How to design a pilot system for tracking adverse and near-miss events in surgical patients

Karl Bilimoria, MD; Daniel Abbott, MD; and Jeffrey Wayne, MD
Department of Surgery Northwestern University

Abstract

The frequency of medical errors has become a pressing issue among health care providers.

In order to track adverse and near-miss events, we designed a Web-based reporting system by which surgical residents and attendings can report these events anonymously. The adverse and near-miss events are classified in detail and compiled into a comprehensive database that will be queried in order to analyze these errors and follow patterns of patient morbidity and mortality. Based on the data, we intend to implement plans to correct systems issues or educational deficits, which may reduce the likelihood of adverse and near-miss events recurring in the future.

Introduction

As a chief resident was presenting a case of hemorrhage in a postoperative patient on heparin, the minds of many in the room were trying to recall the details of a similar presentation at a Morbidity and Mortality (M&M) conference a few months ago. Unfortunately, nobody could recollect that particular case. Could it be that the same error was made twice resulting in significant patient morbidity? The inability to recollect reinforced our thoughts regarding the need for a more sophisticated method by which to track errors.

But what about the mishap that almost happened, the so-called near-miss or close call? Errors that are prevented by a vigilant nurse, detail-oriented pharmacist, or senior resident also need to be tracked and analyzed because today's near-miss may be tomorrow's catastrophe. Since the Institute of Medicine released its 1999 report on the frequency of patient care errors, a renewed emphasis has been placed on tracking and evaluating such events (Kohn and Corrigan 2000). Our department set out to design a system to track adverse events as well as close calls.

Objectives

First, we needed to establish a system by which the cases discussed in Morbidity and Mortality (M&M) conference could be categorized and organized, and we wanted to incorporate data on near-miss events into this system. In addition, it was important to gather opinions on how these errors happened and what could have been done to prevent them. We also assumed that there were a number of events that health care workers failed to report for fear of the various consequences. The overall goal was to analyze these adverse and near-miss events through a comprehensive database in hopes of identifying patterns of errors. Subsequently, solutions could be devised to remedy errors in hopes of protecting patients in the future.

Methods

We set out to design an online Web-based reporting system by which residents and attendings could easily enter adverse and near-miss events in a timely fashion. One can access the hospital's Web page and select the Surgical Patient Safety icon which leads them to the project's homepage (Figure 1). This homepage would offer two reporting options. First, the weekly non-anonymous M&M report that every service currently submits would be switched to the online system (Figure 2). The second route would be an anonymous form which could be completed for near-miss events or adverse events that for whatever reason were not reported through the M&M report page (Figure 3).

The only way to gather as many reports as possible was to allow health care workers to divulge this information without fear of embarrassment, punishment, or retribution. Allowing for maximal anonymity was crucial. As such, access to this Web page is granted by entering a User Name and Password which is known to everyone in the surgery department. Furthermore, the Web page can be accessed from any computer on the Internet, as long as the person has access to the hospital's Web site. Additionally, the site visitor can provide as much or as little information as they choose. Only a description of the event is required to file the report. We also made it known that the reports would be reviewed by the Surgical Patient Safety team only (discussed later in this article), and there would be no instance where we would discuss the specific case outside the team.

Furthermore, the database needed protection from legal subpoenas. The legal details of the M&M process were surprising. In Illinois, the only part of the M&M process that is protected by law is the actual conference. That is, only the one hour discussion itself is free from legal scrutiny. The reporting of events is not protected. Any list of the complications discussed is not protected. Any actions taken or reports generated based on the M&M discussion are not protected. As such, our database of events would not be sheltered. If the hospital or individual attendings were ever sued, the plaintiffs' attorneys could theoretically request a copy of our adverse event and near-miss database. As such, we structured this project to be a modification to the M&M process. Our hospital and departmental Quality Assurance (QA) committees gave us the charge to conduct a research project to track the nature of near-miss and adverse events. This approach obviously required Internal Review Board (IRB) approval, but in the end, all data and discussions would be protected through both the M&M process and research aspect.

Finally we were able to begin designing our online system. The focus of the project was to track error and identify patterns of error. The database could be queried by service, attending, gender, age, date of event, and error type (judgment, technique, nature of disease, systems). Any of these variables could be inclusively or exclusively combined to generate a report. In addition, the specific nature of the error needed to be categorized in order to sort the database effectively. As such, we devised a classification system which grouped error (Table 1). There are numerous classification systems in literature; however, our system was developed by looking back at our department's M&M records from the last two years.

A mockup of the Web-based system was given to our department's information technology group, and a custom software package was subsequently designed. The software allowed residents to enter the data with ease, and allowed the investigators to query the database using any combination of variables. The brevity of the forms to be completed by residents and attendings and ease of use were important principles. As such, most information fields were made optional and the number of fields was limited. The software engineers also made the Web page user-friendly with easy access and drop-down menus.

The results of the reporting will be analyzed monthly by the Surgical Patient Safety committee consisting of two residents and an attending. Every month, the subcommittee will assess the reported cases. The charts will be reviewed under the protection of an IRB Waiver of Authorization for conducting research that presents minimal risk to the patients; the research could not be reasonably conducted without the waiver. We will also review and correlate the month's cases with similar cases logged in the past. The subcommittee will seek out opinions on each case from relevant associated departments including floor, intensive care, and operating room nursing, along with anesthesia and pharmacy. We will have predetermined, constant representation from each of these departments. Potential solutions will be discussed within the sub-committee and the department of surgery. Attempts will be made to identify and correct systems errors. If an educational deficit in knowledge is identified, a presentation on the topic will be given to house staff and attendings, if necessary, in hopes of alleviating the problem. To demonstrate that we are acting on the issues submitted by our residents and attendings, aggregate data and implemented solutions will be presented periodically at our M&M conference.

Conclusions

We have designed a Web-based, anonymous reporting system for adverse and near-miss events. Systematic, standardized, and reliable reporting of weekly morbidity and mortality (M&M) cases will facilitate recognition of severe, recurrent, or otherwise significant events that can be constructively investigated and can be addressed with educational discourse. Reporting of adverse events not reported as M&M cases can contribute to the definition of the real incidence of adverse events threatening patient safety. The process will lead to recognition of educational, procedural, and systematic deficits that can be addressed to reduce similar instances in the future.

As the project proceeds, we intend to open the reporting of events to floor, operating room, and surgical intensive care unit nurses, anesthesia residents and attendings, and pharmacists. In addition, extending this reporting system to the other hospitals in our system will allow us to track all adverse events with which our residents are associated. Finally, should this pilot program succeed, a multi-institutional trial may be undertaken in order to track a larger spectrum of errors and to draw potential solutions based on the experiences others have already encountered.

Bibliography

Barach, P. and Berwick, D. (June 2003). "Patient safety and the reliability of health care systems." Ann Intern Med 138(12): 997-8.

Bates, D. and Gawande. A. (2003). "Improving patient safety with information technology." N Eng J Med 348: 2526-34.

Berwick, D. (2003). "Errors of Today and Tomorrow." N Eng J Med 348(25): 2570-2.

Etchells, E., O'Neil, C., et al. (2003). "Patient safety in surgery: error detection and prevention." World J Surg 27: 936-42.

Gawande, A., Studdert, D., et al. (2003). "Risk factors for retained instruments and sponges after surgery." N Eng J Med 348: 229-35.

Gawande, A., Thomas, E., et al. (1999). "The incidence and nature of surgical adverse events in Colorado and Utah." Surgery 126(1): 66-75.

Kohn, L. and Corrigan, J. (2000). To err is human: building a safer health system. Washington, DC, National Academy Press.

Krizek, T. (2000). "Surgical Error." Arch Surg 135: 1359-1366.

Landrigan, C., Rothschild, J., et al. (2004). "Effect of reducing interns' work hours on serious medical errors in intensive care units." N Eng J Med 351(18): 1838-48.

Leape, L. (2002). "Reporting of adverse events." N Eng J Med 347(20): 1633-38.

Leape, L., Berwick, D., et al. (2002). "What practices will most improve patient safety? Evidence-based medicine meets patient safety." JAMA 288(4): 501-7.

Lockley, S., Cronin, J., et al. (2004). "Effect of reducing interns' weekly work hours on sleep and attentional failures." N Eng J Med 351(18): 1829-37.

Martin, R., Brennan, M., et al. (2002). "Quality of complication reporting in the surgical literature." Ann Surg 235(6): 803-13.

Shojania, K., Duncan, D., et al. (2002). "Safe but sound. Patient safety meets evidence-based medicine." JAMA 288(4): 508-13.

Vincent, C. (2003). "Understanding and responding to adverse events." N Eng J Med 348(11): 1051-6.

Volpp, K. and Grande, D. (2003). "Residents' suggestions for reducing errors in teaching hospitals." N Eng J Med 348(9): 851-5.

Wanzel, K., Jamieson, C., et al. (2000). "Complications on a general surgery service: incidence and reporting." Canadian J Surg 43(2).