In message <00a401c43d7c$07928b60$b21d9ed9@oemcomputer>, Fay Wilson
<[log in to unmask]> writes

>Good to see a consideration of first principles. We are already getting
>patients asking for certain information not to be recorded on the computer
>but the problem is we don't keep paper records. now (maybe we should). The
>problem is that everyone assumes that the record is complete and therefore
>may miss significant matters if they don't ask every time (which sort of
>negates the reason for having records). Maybe we should stop and ask what is
>the purpose of medical records.

I suppose the *purpose* of medical records depends on the settings in
which medicine is practised - including the administrative setting.

To make a stab at some broad categories:-
1. to record consultations with a health professional, actions taken
(medication, referral, lab requests etc.) and agreements reached - an
aide memoir.
2. to allow this information to be used by other health professionals
within the same organisation (practice, hospital clinic, district nurses
attending a patient at home). This does not presuppose any sharing of
records between organisations.
3. To allow use of the record within an organisation to audit the
management of groups of patients within the organisation.

These were the pre-NPfIT purposes - and confidentiality was relatively
simple: information only went out of the organisation if it was
specifically permitted. (There were problems - I'm not saying the system
was perfect, or even very functional..)

Once you start looking at the sharing of full records between
organisations - and in particular between organisations serving
different purposes such as NHS and Social Services - you have a very
different utility, confidentiality and security model - and the records,
like it or not, start to serve very different purposes - whether or not
this is the intention.

So

4. To allow management of the NHS. (we've already got this to some
extent - but the amount of *detail* is limited: there is a difference
between recording that John Smith had a hernia repair for central
returns and being able to access the total record).
5. To allow instant access to the record at the time and place of need -
i.e. an emergency record summary.
6. to allow the care of an individual patient across different
organisations and places of care (which is what the current
re-organisation of the NHS is all about).

Of course, what you need in the record is different in each of these
cases. You don't need to put the same information in a record in the
practice as you do in the referral letter: the referral letter is likely
to contain more of a differential diagnosis and expected action, for
instance.
If the record is held within an organisation , the same applies.
If a record is held in common, things become extremely complicated - who
entered it, how does this entry relate to other entries, which health
professional can over-ride or change management (is the rheumatoid
arthritis needing NSAI more important than the warfarin for AF or a
Starr-Edward valve? who takes responsibility? and who arranges any
follow-up needed - including when to stop?)

This isn't technical security - or even confidentiality - but have the
issues - and their impact on NCRS - been thought through?

Coming back to the thread -
Does "security" include data stability, attribution and standards? Or
are we talking solely about security in the MI5 sense?
What does NPfIT understand by "Confidentiality"? Don't forget, the NHS
is all one big, happy family.. Sharing information with one member
implies consent to share it with the whole lot.. Or has this changed?
If 80% of the people consulted in the Which survey agree to sharing
medical records between health professionals, does this mean that the
views of 20% can and should be disregarded?

Once the minor details of how to structure the record so that parts can
be excluded without in any way putting the patient at risk, devising the
requirements and rules for permission to access some or all of the
record, and developing a structure that will allow implementation of
these rules in a real time setting (locum called in at short notice?
Doesn't only happen in general practice), it might be time to go on to
the research and administration aspects..

The only thing keeping me from total despair is a funny feeling that the
records will be so patchy that they won't actually be that useful!

MaryH
>--
>Fay

--
Mary Hawking