David Wright,

Intensive Therapy Unit; Western General Hospital, Edinburgh

Keywords: Intensive Care Unit; Incident Reporting

When accidents or mistakes are looked at in detail, certain key factors can be seen to have influenced events. The technique of identifying such factors – critical incidents - was developed by Flanagan and others in the 1940s to help analyse aircraft accidents.

By the 1970s, critical incident studies had been used to investigate anaesthetic mishaps . Modelled on this work, we have been looking at critical incidents in an intensive care unit (ICU) since 1989. We have used a questionnaire which was based on which had been developed by Williamson and colleagues for use in anaesthetic practice.

We defined a critical incident as an occurrence that might have led (if not discovered in time) or did lead, to an undesirable outcome. The incident must have been caused by an error made by a member of staff or by a failure of equipment. It had to be described in detail by a person involved in or who had discovered the incident. It had to have occurred while a patient was under the care of the ICU staff (though not necessarily in ICU). It had to be clearly preventable.

Such studies need a form for describing the incidents, a system for collecting the forms, classification and analysis of the incidents, a review of the incidents and action as a result of this, and feedback to staff, which includes summaries of the incidents or any action suggested to lessen the likelihood of future incidents.

The essence of such studies is anonymity and absence of criticism or blame, and these should be constantly emphasised.

The study started in 1989 and our first year's experience was published in 1991 . By early 1999, over 100 incident forms had been completed, and this paper reviews 710 reports from the period of January 1989 to February 1999.

The majority of incidents was due to errors by staff, and was not caused by equipment failure. These human errors fell naturally into four groups: those related to ventilation, those related to vascular lines, those related to drug administration and a miscellaneous group.

When the incidents were analysed and classified, a number of contributing factors emerged. Such factors were inexperience, shortage of staff, night time, poor communication. Also, a number of factors could be seen as contributing to the detection of incidents, such as regular checking, the presence of alarms on equipment, and the presence of experienced staff.

Nurses completed more than 90% of the forms and the vast majority of incidents recorded had no serious sequelae for patients.

Earlier studies confirm how important human error is in the generation of critical incidents . Other studies show that most incidents reported do not have serious long-term sequelae. ICU studies have previously reported 12 or 24 months of incidents. Our current report of 10 years of incidents allows certain points to be considered over a longer period of time. These points include variation years by year in the number of reports submitted, variation year by year in the overall groups of incidents (ventilation, vascular line, drugs, miscellaneous) and whether some incidents occur time and time again despite their being highlighted and preventive action suggested.

In addition, the type of preventive action that is suggested can be classified and analysed.

It is important to appreciate that such studies do not record the true incidence of critical incidents. What they do do is to provide a relatively simple way of drawing attention to things that have gone wrong. They allow these individuals who submit forms an opportunity to analyse their own behaviour and attitudes, and they provide a stimulus for action to try and prevent further incidents. They encourage mutual trust and a mature approach to improving quality, which enhance the way staff feel about themselves and their unit.

Dr David Wright, Intensive Therapy Unit, Western General Hospital, Edinburgh, UK. Tel +44 131 537 1666

Dr David Wright is an anaesthetist, and an ICU consultant, at the Western General Hospital, Edinburgh.