As you know, QoL is measured in a lot of different ways. You end up, aftyer some
years, deciding that you are trying to measure some surrogate of happiness or
trying to describe the multiple dimensions of each person by writing their life
history. Neither easily lends itself to numerical comparison.

One has, therefore, to compromise.  Most of the general physical measures I
have used, both paper and pencil; 'are you able to....' or 'how long does it
take you to...' or even 'how do you consider the quality of your life in
relation to...' and the imposed ones 'now Mrs Jones let me see you make a cup of
tea...' seem to correlate reasonably well at a group level. Some odd things like
sensory deprivation and incontinence seem to be on a different dimension and
some of the mental scores can be odd - others seem to correlate with the
physical dimension.

Lots of different measures are said to be 'disease specific', but looking at
them many seem to have all the old favourite questions. If you look at the
individual questions many can be abandoned and give a very similar score, again
at a group level.

Years ago someone (I can't remember who, but I quote them regularly) said that
any scoring sytem was only useful for comparing groups - that no score could be
used for making decisions about individual treatment and I think that is right.
The 'single assessment process' for elderly people (see the NSF) is a worry in
this regard, as my impression is that a pencil and paper approach is central to
its 'objectivity'.

So its like BMI, most of the standard systems don't do what they say on the
tin, but they're not bad as long as you read the instructions. As as NICE
appraisal committee member can I also make a plea for things that translate
fairly easily into QALYs.

Dr Norman J Vetter
Department of Epidemiology
cardiff University
Heath Park Campus
Cardiff, CF14 4XN
UK

Tel: +44 (0)29 20 742311
       +44 (0)29 20 742318
Fax: +44 (0)29 20 742898

Web: www.normanvetter.com

>>> Matt Williams <[log in to unmask]> 09/02/2005 16:30:23 >>>
Dear All,

I'm a bit stuck on a problem with QoL.

I started by thinking about mortality, where a figure (40%) seems to
have a clear referent (if I take 1000 of you, 400 are dead in x years).
In fact, this interpretation seems to give a semantics to the mortality
figure which is grounded in an objective reality (after all, you're
either dead or not).

QoL doesn't seem quite so simple. Despite the fact that we agree that it
exists, and that it can vary, and that some things would make it better
or worse, I'm not sure what a QoL of 40/100 would mean - even if the 100
point scale were well validated, etc.

I've only been able to think of two ways of dealing with this:
1: Ground everything in functional aspects - how far can you walk, how
often are you sick, etc.

2: Accept we don't really know what someone's absolute QoL is, but look
instead at whether it is likely to go up or down from here. So, if you
have flu, and take some paracetamol, I don't know what your QoL is, and
I don't know what it is going to be, but I can strongly suspect that it
is going to be better than it is now.

The underlying problem is that there doesn't seem to be an objective
referent (as in mortality) for us to hang the interpretation on.

Does anyone else have any ideas?

Thanks,
Matt
--
Dr. M. Williams MRCP(UK)
Clinical Research Fellow
Cancer Research UK
+44 (0)207 269 2953
+44 (0)7384 899570