Citation

Trucco P, Cavallin M. Safety Sci. 2006; 44(6): 491-513.

Affiliation

Department of Management, Economics and Industrial Engineering, Politecnico di Milano, Piazza Leonardo da Vinci, 32 - 20133 Milan, Italy (paolo.trucco@polimi.it)

Copyright

(Copyright © 2006, Elsevier Publishing)

DOI

10.1016/j.ssci.2006.01.003

PMID

unavailable

Abstract

Similarly to the industry sector in the late '80s, nowadays leading organisations in the healthcare sector acknowledge the fact that human errors and system failures can never be totally eliminated; accordingly, hospitals are moving into the challenge of designing "fault tolerant" systems within care management processes. This new perspective leads analysis in a new direction: from a merely retrospective approach to a joint prospective-retrospective one, based on a complete Clinical Risk Management (CRM) process. Current Clinical Risk Assessment (CRA) methods show their inadequacy when healthcare organisations try to set safety targets or to assess safety performance improvements on a quantitative basis. It is here that the call for further methodological developments clearly emerges. This paper deals with the description of a new CRA method, called CREA (Clinical Risk and Error Analysis), showing improved features with respect to the state of the art. CREA implements not only a quantitative risk analysis of error modes, but also a quantitative assessment of critical organisational factors affecting patient safety, based on Vincent's framework [Vincent, C., Taylor-Adams, S., Stanhope, N., 1998. Framework for analysing risk and safety in clinical medicine. British Medical Journal 316, 1154-1157]. Providing a consistent method for the integration of data analysis and expert judgement, CREA presents a higher level of accuracy and reliability with respect to FMEA/FMECA or HFMEA methods.The method has been tested in a vascular surgery department, where over 2500 surgical operations (5% of which take place in the Emergency Unit) and 15,000 outpatient services are performed in a year. The experimental study concerns risk analysis in drug administration, one of the most common and frequent clinical processes during a general hospitalisation.