At 17:49 5/9/96 GMT, Dr Adrian Midgley wrote:
>>Andrew [log in to unmask]
> wrote at 3:26 pm on 5/9/96
>about "Re: Managing Wait lists":
>-----------------------------
>
>>This would only hold if the demographic profile of every practice was
>similar.
>>
>>eg. practice X covers a new housing estate which will (may) have a large
>>number of healthy youngish adults but practice Y covers an inner city area.
>>The people in the first percentile of X's list are unlikely to be as
>>'absolutely' urgent as those in Y's first percentile.
>>or to re-phrase - X ranks an in-growing toenail at 52%, Y would rank it at
>>70% simply because he/she has other more serious things to deal with. Should
>>the toenail consultant then put X's patients ahead of Y's?
>>I suspect a similar point applies to the morbidity rates for each practice.
>
>Thank you.  That is a valid point.  Is it possible to quantify it, for
>instance by reference to data at the Brompton which might allow an
>indication of the variability of urgency distribution by practice referring
>to you;

Good idea but tricky. As we are an ex-SHA (heart & chest) we get referrals
from all over the country so the number of referrals per practice is small &
erratic. Consequently it is hard to tell whether one practice consistently
referrs for urgent care more frequently than another. Some clever data
manipulation may yield something over time.
I suspect (& hope) that many NHS organisations are now reaching the point of
having enough data to start doing some interesting trend analysis.

>Other ways of establishing the set point to compare one practices urgency
>hierarchy to anothers would include a panel (commissioning group??) making
>a direct comparison between patients at slightly different levels in the
>various practices' own rank order.
>
>For instance,
>hypothesis: Practice X patients at the 25th centile (I would avoid the
>first centile and the hundredth, instinctively) are of corresponding
>urgency to practice Y patients at the 15th centile.
>
>comparison: PRactice X patients from  23 24 25 26 27 centile levels
>versus Practice Y patients from 13 14 15 16 17 centile levels.
>
>expected ranking - generally alternating.
>13  14  15  16  17
>  23  24  25  26  27
>If definitely not alternating then redo until it is near enough.
>
>Any estimates of the range of multipliers likely to be needed?

sorry, define 'multipliers'.
Do you means factors to be used in generating the ranking or in performing
the comparison?

>------------------------
>Of course, the fundholding situation at present means something very similar
>is likely to be happening if one but not both X and Y are fundholding.
>
>You are in a good position to contemplate the effect of the inter-hospital
>comparison made possible by this concept if it proved possible to elaborate
>and apply it.

True but this is only the first stage of the process (assuming that
admission is required), having seen the patient the consultant may decide
that an 'urgent' case is routine & so the patient slips down the admission
wait list.

>I am heartened that nobody has told me that it was all worked
>out/tried/discarded years ago...
I believe that the National Case Mix Office are working on Health Benefit
Groups which intersect with this idea. Having said that I am not sure what
is meant by 'urgency' -
        minimising damage to the patient,
        minimising long term costs for looking after the damaged patient,
        patient A will benefit more compared to patient B from being classed
as urgent,

------------------------------------------------------------------------
Andrew Capey				[log in to unmask]
Corporate Data Manager
Royal Brompton Hospital			Tel: 0171 351 8726
Sydney Street, London SW3 6NP, UK	Fax: 0171 351 8743
------------------------------------------------------------------------



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%