Human factors and cardiac surgery: a multicenter study.

OBJECTIVE: To study the role of human factors on surgical outcomes, with a series of 243 arterial switch operations performed by 21 surgeons taken as a model. METHODS: The following data were collected: patient-specific and procedural variables, self-assessment questionnaires, and a written report from a human factors researcher who observed the operation. The relationship of patient-specific variables to outcomes (death and death and/or near miss) was used to develop a multivariable baseline model to analyze the role of human factors after adjustment for these variables. RESULTS: The overall mortality was 6.6% with 24.3% of cases resulting in death and death and/or near misses. The self-assessment questionnaires were found to be unhelpful. Major and minor human failures were extracted from the written report. Major negative events were potentially life-threatening failures, whereas minor events were failures that, in isolation, were not expected to have serious consequences. Major events were closely related to death (P <.001) and death and/or near misses (P <.001). Appropriate compensation, however, sharply reduced the risk of death (P =.003). The total number of minor events was also closely related to both death and death and/or near misses (P <.001). CONCLUSION: The study highlights the role of human factors in negative surgical outcomes. Even in the most eventful circumstances, however, appropriate human factors defense mechanisms can lead to a successful outcome.

The human factor in cardiac surgery: errors and near misses in a high technology medical domain.

In this review, we discuss human factors research in cardiac surgery and other medical domains. We describe a systems approach to understanding human factors in cardiac surgery and summarize the lessons that have been learned about critical incident and near-miss reporting in other high technology industries that are pertinent to this field.

The role of structured observational research in health care.

Structured observational research involves monitoring of healthcare domains by experts to collect data on errors, adverse events, near misses, team performance, and organisational culture. This paper describes some of the results of structured observational studies carried out in health care. It evaluates the strengths, weaknesses, and future challenges facing observational researchers by drawing lessons from the human factors and neonatal arterial switch operation (ASO) study in which two human factors specialists observed paediatric cardiac surgical procedures in 16 UK centres. Lessons learned from the ASO study are germane to other research teams embarking on studies that involve observational data collection. Future research needs robust observer training, clear measurable criteria to assess each researcher's domain knowledge, and observational competence. Measures of inter-rater reliability are needed where two or more observers participate in data collection. While it is important to understand the factors that lead to error and excellence among healthcare teams, it is also necessary to understand the characteristics of a good observer and the key types of error that can occur during structured observational studies like the human factors and ASO project.

Diagnosing "vulnerable system syndrome": an essential prerequisite to effective risk management.

Investigations of accidents in a number of hazardous domains suggest that a cluster of organisational pathologies-the "vulnerable system syndrome" (VSS)-render some systems more liable to adverse events. This syndrome has three interacting and self-perpetuating elements: blaming front line individuals, denying the existence of systemic error provoking weaknesses, and the blinkered pursuit of productive and financial indicators. VSS is present to some degree in all organisations, and the ability to recognise its symptoms is an essential skill in the progress towards improved patient safety. Two kinds of organisational learning are discussed: "single loop" learning that fuels and sustains VSS and "double loop" learning that is necessary to start breaking free from it.

Dynamic modelling in a study of surgical error management.

We demonstrate the use of dynamic longitudinal models to investigate error management in cardiac surgery. Case study data were collected from a multicentre study of the neonatal arterial switch operation (ASO). Information on two types of negative events, or 'errors', observed during surgery, major and minor events, was extracted from case studies. Each event was judged to be recovered from (compensated) or not (uncompensated). The aim of the study was to model compensation given the occurrence of past events within a case. Two models were developed, one for the probability of compensating for a major event and a second model for the probability of compensating for a minor event. Analyses based on dynamic logistic regression models suggest that the total number of preceding minor events, irrespective of compensation status, is negatively related with the ability to compensate for major events. The alternative use of random effects models is investigated for comparison purposes.

Pitfalls of adverse event reporting in paediatric cardiac intensive care.

AIMS: To evaluate the pitfalls of incident reporting in a complex medical environment. METHODS: Retrospective review of 211 incident reports in a paediatric cardiac intensive care unit (CICU). Two adverse event reporting databases were compared: database A (DA), the hospital's official reporting system, is non-anonymous and reports are predominantly made by nurses; database B (DB) is anonymous and reports are submitted by a CICU consultant who collects data from daily ward rounds. Both databases classify adverse events into incident type (drug errors, ventilation, cannulae/indwelling lines, chest drains, blood transfusion, equipment, operational) and severity (0 = no, 1 = minor, 2 = major, 3 = life threatening consequences). RESULTS: Between 1 April 1998 and 31 July 2001 there were 211 adverse events involving 178 patients (11.87%), among 1500 patients admitted to CICU. A total of 112 incidents were reported in DA, 143 in DB, and 44 in both. In isolation, both databases gave an unrepresentative picture of the true frequency and severity of adverse events. Under-reporting was especially notable for less severe events (grade 0, or near misses) CONCLUSION: Incident reporting in the medical field is highly variable, and is heavily influenced by profession of the reporters as well as anonymity. When adverse event reporting is based predominantly on the observations of a single professional group, the data are grossly inaccurate.