PNNL

Abstract

It is now well understood in health care, as stated in an Institute of Medicine (IOM) report (2001),1 that safety is a system property and that patients are injured because of poor system designs. In a subsequent report,2 the IOM advocated an applied research agenda on patient safety that would include, among other things, hazard analysis to assess the validity and efficiency of integrating retrospective and prospective techniques. 2(p. 20) Since July 1, 2001, The Joint Commission has required that hospitals conduct proactive risk assessments on their self-identified high-risk processes. In health care, proactive risk assessment usually takes the form of Failure Mode and Effects Analysis (FMEA). Simply put, a proactive risk assessment answers three questions: (1) What can go wrong? (2) How likely is it? and (3) What are the consequences? An applied research firm, four community hospitals, and a community health care alliance in south-central Washington State as members of the Tri-Cities Patient Safety Coalition (TCPSC) have collaborated in applying proactive risk assessment methods to high-risk health care processes. The four participating hospitals are Kadlec Regional Medical Center (Richland), Kennewick General Hospital (Kennewick), Lourdes Health Network (Pasco), and Prosser Memorial Hospital (Prosser). In TCPSC hospitals’ experience, FMEA may be difficult to apply to certain health care processes.3 In cases when there is lack of specific process-related detail or when a process is being considered at a high level, FMEA may not be the best assessment tool. Other forms of proactive risk assessment, such as fault tree analysis (FTA), event tree analysis (ETA), and hazard identification, are used in high-risk industries such as aerospace, nuclear energy, and chemistry.7 Although not yet in common use in assessing health carerelated processes, these methods can yield useful results. The advantage of using FTA and ETA over FMEA is that they explicitly consider the effect of combinations of failures, whereas FMEA considers the effect of one failure at a time and combinations of events define scenarios that lead to various undesired outcomes. Defining the set of process- or system-event combinations (scenarios) can provide a valuable picture of the risk. This article presents three case studies as a tutorial on three different kinds of proactive risk assessments performed by the TCPSC to assess the risk of adverse events associated with a process or system: (1) an ETA of different specific situations involving hospital patients who act out in unsafe ways, (2) a hazard identification of the use of patient-owned equipment inside the hospital, and (3) an interhospital FMEA of specific interhospital transfers of patients. Table 1 (page 367) provides a summary the features of and comments on the three kinds of proactive risk assessments on the basis of the experiences described in this article. A permanent TCPSC team including an engineer [G.C.] from the applied research firm and representatives from each of the five member organizations [the remaining authors, four of whom have nursing backgrounds and are in middle management] conducted all three proactive risk assessments. In each case, the engineer led the team’s sessions. In addition, the team drew on the resources of the participating hospitals for appropriate expertise. For example, for hazard identification for the use of patient-owned equipment, one hospital donated the time of its environment of care nurse, another hospital donated the services of its clinical engineer, and a third hospital donated the services of a registered nurse for the duration of the assessment. The interhospital FMEA, which included process walkdowns at each of the member hospitals, required participation of a larger team. The team, meeting at least monthly (but in some cases, weekly) took two to six months to perform each of the assessments.