Intended for healthcare professionals


Improving patient safety through training in non-technical skills

BMJ 2009; 339 doi: (Published 23 September 2009) Cite this as: BMJ 2009;339:b3595
  1. Rhona Flin, professor of applied psychology1,
  2. Rona Patey, consultant anaesthetist2
  1. 1School of Psychology, University of Aberdeen, King’s College, Old Aberdeen AB24 2UB
  2. 2Aberdeen Royal Infirmary
  1. r.flin{at}

    As in aviation, education should occur early in the core curriculum

    To reduce iatrogenic injury, healthcare organisations have been encouraged to adopt approaches from high risk industries—most notably aviation—that focus on human factors. The best known of these methods is crew resource management (CRM) training, designed to reduce human error by enhancing non-technical skills such as situation awareness, decision making, and teamwork.

    Although CRM programmes are widely used in aviation, and mandated in many countries, measurable effects on safety outcomes remain elusive, partly because commercial aircraft accidents are infrequent. Although some studies have reported changes in behaviour, a meta-analysis of the effectiveness of CRM training only found significant improvements in trainees’ attitudes to safety.1

    Varieties of CRM training are being adopted for multiprofessional groups of clinical staff who have had no previous education in non-technical skills. Some courses focus on teamwork,2 others cover a range of topics.3 Although training in multidisciplinary teams has some benefits, it is not the ideal method for teaching basic CRM concepts.

    The timing of such training is important for successful application in health care. Undergraduate and early professional education in non-technical skills would provide course participants with a basic understanding of the psychological and physiological factors influencing human performance. A common vocabulary would facilitate discussion of unsafe behaviours, improve team communication, and help to develop solutions for reducing risks to patients. A study of behaviour observed during surgical procedures before and after team members had attended a CRM training course and received support coaching showed significant improvements in error rates (technical) and non-technical skills.4 But investigators noted that “considerable cultural resistance to adoption was encountered, particularly among medical staff.”

    The introduction of CRM training to health care may ultimately help to improve the safety culture by shifting the norms of acceptable behaviour. But the antipathy encountered in the surgical study4 and the need for support coaching, suggests that the non-technical skills courses are unlikely to change professional behaviours unless they are properly integrated into educational and safety management systems. Healthcare systems have been advised not to mindlessly import these programmes: “the aviation experience should be used as a template for developing data driven actions.”5

    Importantly, in aviation, CRM courses are a core component of professional education and they are not the pilot’s first introduction to human factors. A student pilot who wants to gain a licence must pass several theoretical examinations, including one on human performance.6 The syllabus covers basic physiology (for example, vision, fatigue) and basic psychology (including safety culture, perception, decision making, memory, and stress).7 For more advanced pilot licences, the syllabuses contain topics on teamwork and leadership. From the very start of a pilot’s professional education, the message is conveyed that technical skills alone are not enough to ensure a safe flight.

    Once pilots are employed, their organisation must provide CRM skills training and assessment of the non-technical skills takes place as part of regular competence assurance (licensing) checks. The CRM courses are usually specially designed to cover a syllabus defined by the regulator, which is based on non-technical skills identified from analyses of accidents, flight deck observations, aviation psychology research, and confidential incident reporting.5 8 Thus a feedback loop transfers safety intelligence into continuing professional development through the CRM training.

    Healthcare professionals also need to understand the physiological, psychological, and social factors that may affect their ability to deliver safe clinical treatment. The Parliamentary Inquiry into Patient Safety pointed out that “there are serious deficiencies in the undergraduate medical curriculum which are detrimental to patient safety” and recommended training in non-technical skills.9 We would agree that it is too late to start delivering this training after undergraduate education has been completed and professional attitudes are almost fully formed. Like the pilots’ training, education in non-technical skills should be delivered early to healthcare students as part of their core curriculum. Failure to include CRM principles and practices in the undergraduate course results in such teaching being undervalued when it is offered after qualification. Similarly, in the postgraduate curriculums, the absence of training in non-technical skills misses the opportunity to provide repeat practice and develop an integrated assessment strategy that includes human factors.10

    There are indications that this deficit is beginning to be tackled. The WHO curriculum on patient safety for medical students includes a module on human factors, and this training package is now being adapted for other professional groups.11 At Aberdeen University’s medical school, a new module on non-technical skills has been introduced to the final year course, and during clinical attachments students are asked to identify use of these skills in practice when watching senior doctors. In specialties such as surgery and anaesthesia, scientific studies of behaviours in members of operating theatre teams are now producing suitable data on which to base postgraduate courses in non-technical skills.12


    Cite this as: BMJ 2009;339:b3595


    • Competing interests: None declared.

    • Provenance and peer review: Commissioned, not externally peer reviewed.


    View Abstract