Surely the OUTCOME measure is something like dead/alive or functional
status. Severity of illness is more of an INPUT measure?
Faced with a similar problem a few years ago we were forced back onto
APACHE, but it certainly wasn't very good. Apart from all the theoretical
problems, you needed things like blood gas results to compute the APACHE
score (which I suppose you won't have if this is a trial about avoiding
hospital admission). But APACHE is at least explicitly about acute illness.
Incidentally, you will need some clear thinking about what you mean by a
valid measure of severity. Without being too Zen about it, the problem is
that a severe acute illness is one which might have killed you, so severity
is risk of dying. But if you didn't die, it can't have been severe, right? I
don't think probabilistic scores completely resolve this koan.
(Irwin D, Jessop EG Severity of sickness at admission to hospital in
Colchester 1985 and 1990. Public Health 1993; 107: 171 - 175.)
edmund jessop
EG Jessop
West Surrey Health Authority
Ridgewood Centre, Old Bisley Road
CAMBERLEY
Surrey GU16 5QE
01276 605545
[log in to unmask]
-----Original Message-----
From: p=NHS NATIONAL
INT;a=NHS;c=GB;dda:RFC-822=EVIDENCE-BASED-HEALTH(a)JISCMAIL.AC.UK;
Sent: 28 June 2001 09:32
To: p=NHS NATIONAL
INT;a=NHS;c=GB;dda:RFC-822=EVIDENCE-BASED-HEALTH(a)JISCMAIL.AC.UK;
Subject: outcome measures
Dear all
am looking for an outcome measure to measure severity of illness of acutely
ill older people that may require admission to hospital (partly to compare
case-mix between control and intervention groups in a trial).
am aware of the Duke severity of illness index (DUSOI index) and Shiels' et
al's work on testing this scale on non-acutely ill patients in the UK.
I suspect there isn't a useful scale that has been tested/validated on the
acutely ill older population. Am I right?
be grateful for pointers as hitting blind alleys at a rate of knots
now.........
Owen Dempsey
clinical research fellow
Centre for research in primary care
Uni of Leeds
|