Intended for healthcare professionals


Severe acute respiratory syndrome revisited

BMJ 2003; 326 doi: (Published 19 April 2003) Cite this as: BMJ 2003;326:831

SARS and other ‘medical’ major incidents: lessons from the Furness Legionnaire’s outbreak

Zambon’s editorial1 reminds us of the challenge such outbreaks might
pose for healthcare systems. We have been analysing the local
organisational response to the Legionnaire’s disease outbreak in Barrow-in
-Furness in summer 2002. Some features may be relevant to large-scale
‘medical’ incidents such as SARS (sudden acute respiratory syndrome),
influenza epidemics and the effects of biological terrorism.

First, recognition of the outbreak may be delayed, especially when
symptoms are non-specific. Although Legionella was identified promptly,
the lack of a single catastrophic trigger such as an explosion (as would
be usual in a ‘typical’ major incident) and the initial difficulty
predicting the scale of the outbreak led to some confusion as to quite
whether, and when, a major incident should be declared. A further
difference is the duration of the crisis, which necessitated careful
manpower planning to protect staff from overwork. This would be more
pronounced if staff themselves were incapacitated, as might well occur
with SARS.

Second, the low mortality rate was attributed partly to the
widespread use of an early warning scoring system for the timely
identification and intensive care referral of deteriorating patients2. We
have still to explore whether this effect was due to the scoring system
itself or the close involvement of
ICU staff on general wards.

Third, the hospital’s incident plan was simply not designed for this
type of incident. Paradoxically, this seems to have been beneficial in
that it allowed experienced clinical and managerial staff the freedom to
improvise as events demanded. Further, despite the presence of a central
incident room, our data suggest a loose organisational hierarchy with
relatively ‘low status’ employees able to make decisions. These
characteristics are evident in safety-critical ‘high reliability
organisations’3 and the challenge for major incident planning is to
prevent such vital human factors being stifled by protocol and

1 Zambon M Sudden acute respiratory syndrome revisited(editorial)
BMJ 2003; 326: 831-2 (19 April)

2 Morgan RJM, Williams F, Wright MM An early warning scoring system for detecting developing critical illness. Clin Intens Care 1997; 8:

3 Roberts KH Some characteristics of one type of high reliability
organisation Organization Science 1990; 1: 160-76

Andrew Smith Head of R&D

Cathy Wild Researcher

Morecambe Bay Hospitals Trust,
Royal Lancaster Infirmary,
LA1 4RP,

John Law

Department of Sociology,
University of Lancaster,
LA1 4YT,

Competing interests:  
None declared

Competing interests: No competing interests

23 April 2003
Andrew F Smith
Consultant Anaesthetist
Cathy Wild, John Law
Royal Lancaster Infirmary, LA1 4RP, UK