Yoel Donchin

A look into the nature and causes of human errors in the intensive care unit

OBJECTIVES The purpose of this study was to investigate the nature and causes of human errors in the intensive care unit (ICU), adopting approaches proposed by human factors engineering. The basic assumption was that errors occur and follow a pattern that can be uncovered. DESIGN Concurrent incident study. SETTING Medical-surgical ICU of a university hospital. MEASUREMENTS AND MAIN RESULTS Two types of data were collected: errors reported by physicians and nurses immediately after an error discovery; and activity profiles based on 24-hr records taken by observers with human engineering experience on a sample of patients. During the 4 months of data collection, a total of 554 human errors were reported by the medical staff. Errors were rated for severity and classified according to the body system and type of medical activity involved. There was an average of 178 activities per patient per day and an estimated number of 1.7 errors per patient per day. For the ICU as a whole, a severe or potentially detrimental error occurred on the average twice a day. Physicians and nurses were about equal contributors to the number of errors, although nurses had many more activities per day. CONCLUSIONS A significant number of dangerous human errors occur in the ICU. Many of these errors could be attributed to problems of communication between the physicians and nurses. Applying human factor engineering concepts to the study of the weak points of a specific ICU may help to reduce the number of errors. Errors should not be considered as an incurable disease, but rather as preventable phenomena.

Healthcare climate: a framework for measuring and improving patient safety.

Objectives:Reviews of patient safety efforts suggest that technical/administrative change must be augmented by global factors such as organizational culture and climate. The objective was to outline a comprehensive model for healthcare climate and test one of its elements, the nursing subclimate, in terms of several patient safety outcomes. Design:Measure organizational climate in nursing units, followed by random sampling of patient safety practices in each unit 6 months later. Setting:Sixty-nine inpatient units in three hospitals that make up the entire tertiary care system in one metropolitan area. Subjects:A total of 955 nurses. Interventions:None. Measurements and Main Results:A two-part Nursing Climate Scale referring to hospital- and unit-level climates, followed by five randomly timed observations of patient safety practices covering routine and emergency care in each unit. Climate scales met the criteria of internal reliability, within-unit agreement, and between-unit variability, using standard statistics of climate research. Both the hospital and unit nursing climates exhibited significant variation, which predicted the routine medication safety scores (Z = 2.65 and 2.93 accordingly, p < .01), with similar results for emergency safety scores. A significant interaction (Z = 2.78, p < .01) indicated that best/worst safety is obtained when the unit and hospital climates are aligned (for better or worse) and that positive unit climate can compensate for the detrimental effect of poor hospital climate. Furthermore, climates strength increased its predictive power with regard to patient safety practices (Z = 3.64 for medication and 2.28 for emergency safety; p < .01). The small number of participating hospitals limits organization-level analyses. Conclusions:The nursing climate identifies units where the likelihood of adverse events is greater or lower than the hospitals average. Such information can guide prevention efforts in selected units. These data encourage the development of additional climate subscales subsumed under the healthcare climate model (e.g., physicians subclimate).

Transparency in Health CarePreoperative Briefing in the Operating Room: Shared Cognition, Teamwork, and Patient Safety

Contemporary preoperative team briefings conducted to improve patient safety focus mainly on supplying identification details regarding the patient and the surgical procedure. Drawing on cognitive theory principles, in this study a briefing protocol was developed that presents a broader perspective model of the patient and the planned procedure. In addition to customary identification details and drug sensitivities, the new briefing also includes review of significant background information, needed equipment, planned surgery stages, and so forth. The briefing content was developed following 130 continuous, nonstructured observations conducted in gynecologic and orthopedic operating rooms. The briefing form was designed as a large poster hung in a visible position on the operating room wall. The poster guides the team members (ie, nurses, surgeons, and anesthesiologists) in their conduct. Briefing is conducted orally, and no written records are required. The number of nonroutine events (ie, situations that, if not corrected, might lead to patient harm) observed in the 130 surgeries conducted without briefing was compared with the number of events in 102 surgeries in which briefing was conducted. There was a 25% reduction in the number of nonroutine events when briefing was conducted and a significant increase in the number of surgeries in which no nonroutine event was observed. Team members evaluated the briefing as most valuable for their own work, the teamwork, and patient safety. Following the study, the new briefing format was accepted and adopted for routine use. Team briefings designed to supply a broader-perspective surgery model may be an easy-to-apply tool to reduce the number of nonroutine events during surgery and increase patient safety.

Cardiac Vagal Tone Predicts Outcome in Neurosurgical Patients

ObjectiveTo evaluate the relationship between presurgical levels of cardiac vagal tone and outcome in neurosurgical patients. DesignProspective series. SettingRespiratory ICU in a university hospital. PatientsFifty-one adults admitted to the respiratory ICU between 1982 and 1985. Forty-two patients were scheduled for elective neurosurgery, and nine patients suffered from head trauma. InterventionsTen minutes of electrocardiographic (EKG) data were recorded before medical intervention. Neurosurgical patients scheduled for surgery had EKG data recorded 24 hrs before their operation. Trauma patients had EKG data recorded immediately after arrival in the respiratory ICU. Measurements and Main ResultsCardiac vagal tone was evaluated using a vagal tone index, quantified from the EKG. Cardiac vagal tone monitored before surgical intervention significantly distinguished between the outcome groups only for the elective neurosurgical patients. Age, gender, heart rate, Glasgow Coma Scale scores, and tumor location, size, and malignancy were not related to outcome in the elective neurosurgery group. However, within the trauma group, low Glasgow Coma Scale scores were significantly related to poor outcome. ConclusionsCardiac vagal tone may offer important predictive value by alerting the physician to the functional consequence of head injury. Information relating to autonomic nervous system functioning, such as the vagal tone index used in this study, may provide additional information that will complement the computed tomography scan results. This study demonstrates that the vagal tone index is a predictive factor that may be efficiently extracted from the heart rate pattern routinely monitored in ICUs.

The Nature and Causes of Human Errors in a Medical Intensive Care Unit

The articel presents the main outcomes and conclusions of a two year research effort directed to study the causes of human errors in a Respiratory Intensive Care Unit (ICU). In the course of the study, doctors and nurses recorded errors in treatment routines that were committed during their daily work. Over a period of 4 months we collected 554 errors, which were independently judged for their criticality. In addition, 46, twenty-four hour, observations were conducted, of all activities at a patient bed. A total of 8178 activities were recorded over the 46 observations. We also performed: a detailed human factors analysis of the patient bed as a work station. It was found that the dominant cause of errors are problems related to complete and clear documantation and transfer of information between staff members. Additional causes were lack of standatization in equipment composition and layout, as well as absence of adequate marking and labeling. These problems seems” to be equally relevant to other ICUs visited by the team. Remedial steps are presently being implemented.

Spectral analysis of fetal heart rate in sheep: The occurrence of respiratory sinus arrhythmia

Respiratory sinus arrhythmia is a pattern of rhythmic variation in the heart rate that occurs at the frequency of respiration and is mediated principally by the vagus nerve. Spectral analysis can decompose the variance of a series of sequential measures into constituent frequencies to measure and verify whether there is respiratory sinus arrhythmia in utero in the fetal lamb. Recordings of heart period were obtained from electrodes implanted under fetal skin in six chronic preparations. Respiratory rate and heart period were recorded immediately after delivery and daily for the next 5 days. Respiratory sinus arrhythmia was clearly demonstrated in the neonatal lambs, and the same frequency of respiratory sinus arrhythmia was observed in the fetus and in the newborn lamb (0.8 to 0.1 Hz). There was a reproducible pattern of change in respiratory sinus arrhythmia from 27 days before delivery until term, with a decline in the amplitude of respiratory sinus arrhythmia 4 to 8 days before delivery. We conclude that respiratory sinus arrhythmia was demonstrated in fetal sheep and may serve as an indicator of the integrity of the central nervous system in the fetus and the neonate.

An analysis of work activity in the operating room: Applying psychological theory to lower the likelihood of human error

We report on our observations in the operating room, where we examined the work activity in hopes of understanding why human errors can occur even with simple tasks (e.g., loading a syringe with the wrong solution). We employed a human factors analysis guided by our understanding of human cognition (memory, attention, action planning, etc.) with the goal of improving safety. By applying psychological theory (human cognition) to this real-world environment, we suggest where human error is prone to occur due to factors such as non-optimal procedures and design layout. We speculate that such weaknesses can contribute to adverse events and offer low-cost solutions aimed at minimizing the likelihood of such errors occurring.

Discussion Panel Establishing in house Human Factors expert teams to enable comprehensive design of medical work units

The present discussion panel addresses the need and possible approaches for providing integrative and inclusive human factors design of medical work units. An associated question is whether such a design perspective can be achieved without the instantiation of in house human factors teams. While recognition of the general importance and possible contribution of human factors to efficiency and safety of health care is on the rise and is accompanied by a rapidly growing body research and publications; to date its focus has been mainly on individual systems and isolated work procedures. An important overlooked requirement is for a coherent and inclusive design of the global work unit (operating theater, hospital ward, neonatal unit, etc, etc), much the same way in which the overall configuration of an airplane cockpit or a process control room are considered. Furthermore, can such an inclusive perspective of work units be achieved, unless health care institutes establish in house human factors teams? These are the topics to be evaluated.

Human Factors and Safety in Medical Systems: Where have we Reached? Where should we go?

Over the last two decades there has been a growing recognition in the need for a systematic study of adverse events, errors and difficulties in health care. The systematic investigation of this topic and the resultant database have grown and diversified exponentially. We believe that the time has come to evaluate the achievements of this first wave of research, and discuss directions for the next stage. The participants in the panel have been major contributors to this area of work. They differ in their background and specific research interests, which reflect aspects of the overall medical system. The panel addresses several key questions: Merit of different information sources, retrospective versus prospective studies, implementation methods, and education needs. The discussion examines communalities and differences of philosophy, principles and approaches, with an eye on guiding future work.