From: "Jon Brassey" <[log in to unmask]>
> We've been asked an interesting question which on an obvious level is unanswerable. However, I
> think it's a very interesting type of question which has significant impact on evidence-based
> practice. The question is:
>
> "If every diabetic was treated in primary care according to best practice, how many would still
> need to go to secondary care"
Yes, an interesting question.
Type I -- lets exclude them from discussion
Prevalence is about 1:800 and many are children.
In my opinion GPs should not be looking after these patients.
In the fast moving world of modern medicine it is impossible
to keep up to speed on this subject for 3-4 patients (even
in a massive practice, with one GP seeing them all, it will
only be about 30 patients). Practice nurses can not hope to
know what a diabetic specialist nurse knows etc. etc..
Type IIs
The answer, IMO, to the question is "about the same as now"
(assuming that we GPs continue to refer for the same things
at the same rate and don't have to change behaviour to
fit with new quality initiatives!)
Firstly, best care does not really "prevent" problems it only
"delays" problems. This is usually fudged over in to-days politically
driven medicine as the sound-bite of "x number of heart attacks
prevented" is so seductive.
Secondly, best care is not that effective ;-)
Heresy I hear people shout.
Well however much they re-package and hype the UKPDS results
the NNTs over 10 years were pretty poor with the
intensive vs conventional treatment trial failing to produce
any significant results except in microvascular disease.
It will be interesting to see good long term studies that combine
intensive lipid, blood pressure and glucose treatments -- maybe we will
make a radical difference, maybe this metabolic disease is not
amenable to our current manipulations. Maybe we will have lots
of physically fit diabetics who have dementia and need other
forms of care - who knows, time will tell.
David Jobson
Retired GP
The 1998 UKPDS Studies NNTs for the three trials
Intensive vs conventional
Intensive vs Metformin
Tight BP control
NNT 10y
NNT 10y
NNT 8.4yr
Any diab related end point
31
12
11
Diab relate Death
ns
ns
20
All cause Mortal
ns
19
ns
MI
ns
ns
Stroke
ns
36
27
PVD
ns
ns
Microvasc dis
42
ns
21
i.e. 21 major endpoints, but only 8 sig results--but 3 of those include the others
could say 18 major endpoints and only 5 sig results
|