Examination of Witnesses (Questions 369-379)|
TUESDAY 4 FEBRUARY 2003
369. Good morning ladies and gentlemen. Thank
you very much for coming along. Could you please introduce yourselves
for the record and after that, if you have any opening comments
to make either collectively or individually, now is the time to
do that before we get onto the questions.
(Mrs Howard) I am Janet Howard. I am
a health protection nurse working in Shropshire. I am also the
Co-ordinator of the Community Network of the Infection Control
(Mrs Perry) I am Christine Perry. I am Senior Infection
Control Nurse at the United Bristol Health Care Trust. I am Chairperson
of the Infection Control Nurses Association.
(Professor Little) I am Paul Little. I am a GP and
professor of primary care research at Southampton University.
My area of interest is in the management of acute infections.
I do not pretend to be an expert on surveillance.
(Mrs Williams) I am Gini Williams. I am a lecturer
in TB and Public Health at City University. I am a trustee for
the charity TB Alert and I am nurse consultant to the International
Union of TB and Lung Disease.
370. As there are no opening comments to make
I will take the first question. You are aware of the document
Getting Ahead of the Curve which identifies under-reporting
of infection as a barrier to effective surveillance. What, in
your opinion, are the problems in the primary care setting that
need to be overcome to ensure high quality, comprehensive reporting?
(Mrs Williams) In terms of tuberculosis most of the
reporting is done by specialist centres as TB is actually diagnosed
most often in a chest clinic. It is not necessarily an issue for
primary care; the issue is for primary care to recognise potential
symptoms of TB and then refer them to the specialist services.
(Professor Little) I suppose I want to go back a step.
I personally found the arguments in Getting Ahead of the Curve
not very convincing. One of the problems for me is the understanding
of what we mean by surveillance. There are a whole range of potential
ways that you can perform surveillance. There is passive surveillance
which occurs at the moment when the GP or the nurse will send
a sample of somebody they are particularly worried about, but
generally things are managed syndromically. There is also passive
surveillance in terms of syndromic surveillance. By that I mean
the sort of information that we are getting from NHS Direct about
the pattern of symptoms. That information has only just started
to come on line but it is incredibly useful, in my opinion. Then
there is the type of sentinel system that the Royal College has
where specific spotter practices will perform syndromic surveillance
but when there is a rise in influenza like illness there will
be targeted sampling as well. Going further on there would be
structured clinical and microbiological surveillance with GPs
sending a lot of specimens routinely according to defined protocols.
When the CMO advises us that we need an effective surveillance
system I am not quite sure what he is talking about. I think there
are advantages and disadvantages in all those systems, but in
my view, using syndromic surveillance actually is extremely promising.
For example, we can use the information from NHS Direct on a daily
basis; we can target when there is a peak in influenza like illness
in the south of England. That could then be fed into the local
CCDC and there could be targeted microbiological surveillance
if necessary. You have a trigger system. Is anything odd going
on? If there is some odd can we target sampling there? That would
seem to me a very efficient use of resources. Whether we should
augment the other forms of surveillance I am not quite sure; it
depends on what you want surveillance to do. I did not find the
arguments for why we should augment surveillanceif that
is what was intended by the CMOvery convincing. The arguments
were about antibiotic resistance, spotting outbreaks and new infections.
I could deal with each of those briefly if you would like me to.
371. You mentioned NHS Direct and things like
flu. We are all very used to flu. Do people actually ring NHS
Direct about flu because people know what to do about it and I
would have thought it might be rather under-reported to NHS Direct.
Or is that not the case? I am genuinely curious.
(Professor Little) You are right. And this is one
of the things we have to bear in mind. The situation from a primary
care perspective is that most people do not report the common
infections because they self-manage perfectly happily. That will
apply to flu, sore throats, tonsilitis, chest infections, urinary
infection, you name it. We cannot actually afford for all those
people to be reporting their acute infections. We see about one
in nine; something like that, one in nine or one in ten. We cannot
afford to do anything to the system that is going to bring all
those people into medical care and medical advice. We would just
372. Has some research been done so that we
know what the multiplier is? Do we know the level of under-reporting?
(Professor Little) Not terribly recent. Most of that
is old research. I do not think we know currently. It might be
a matter of one in five rather than one in ten, but most people
manage their own things themselves perfectly happily. We do see
the things that are more severe. The chest infections we see will
be more severe than most chest infections managed by people themselves,
but there is huge overlap and we cannot afford to be seeing all
the people who have flu and chest infections and sore throats.
373. Do we want to know that accurately?
(Professor Little) I think that is a very good question.
I am not sure we do need to know. I think the question is: what
are the arguments for enhancing surveillance? We need to be very
careful that we are not going to bring people who are happily
self-caring into the system.
374. It sounds that the combination of sentinel
practice observations plus NHS Direct make a really rather powerful
combination, useful for most purposes.
(Professor Little) I personally believe that is true.
You would probably spot an outbreak fairly quickly with that system.
You will see rapidly whether there is an increase in gastro-enteritis
in Truro, so let us ask the GPs for a week or so to send samples
in from all the acute gastro-enteritis or food poisoning patients.
375. The thing about NHS Direct is not actually
whether you are seeing true levels but you are looking for perturbations
(Professor Little) That is right. You are looking
for: "something funny is going on". It may not be very
representative but actually there is something different going
on now and does it need to be investigated.
(Mrs Perry) I would agree that we need to separate
out the differences between the different types of surveillance.
Certainly we do need that on-going surveillance to look for the
peaks and the trends, but there may be times that we do need that
increased surveillance that picks up on the risk factors so that
we are able to target our actions on those risk factors to be
able to prevent infections. The on-going surveillance will not
necessarily pick up those risk factors. That is where you would
need the further information on where people have been exposed
to infection and be able to target your action.
(Mrs Howard) I would like to build on what my GP colleague
is saying. Initially there will need to be a change in legislation
to allow other professionals to actually notify and report. That
is the first issue. I think there needs to be a big involvement
with colleagues in primary care and other colleagues about how
we plan the notification and surveillance systems. We need to
be able to give clear evidence as to why we are particularly asking
people to report. Do we really need to know some things? We need
to understand that there is a trend happening. If we do not give
clear evidence my feeling is that we will end up with reporting
fatigue whereby the very important ones that we require enhanced
surveillance forsuch as TB, ecoli 0157we
will not actually find out about. We want to get the information
that will help us investigate them properly. The other issue for
primary care is, I think, that there is no really prompt system
for feedback. There is no encouragement to tell us about things.
The other big issue for the future I think will be expert backup
within the primary care trusts for the IT systems. When we have
IT systems that will talk to one another we are going to need
real expert prompt and on-going backup for the systems. I think
those are the key issues if we are to make things work as they
talk about in Getting Ahead of the Curve, certainly from
376. Can I just come back a little on the surveillance
business and what it means. It obviously means different things
to different people. It would be a shame to throw out bacteriological
or microbiological surveillance on the basis that we do not know
what to do with it and there is too much of it. Those are obviously
practical reasons, but it would be a shame if we did not pick
up TB or ecoli 0157 simply because we were relying on syndromic
surveillance. I just wondered where you see microbiological surveillance
fitting in. I am aware that amongst the myriad of patients with
diarrhoea and vomiting there will be a small proportion of ecoli
0157 and unless you do the test you will not know, and yet you
may be spreading it around if you do not know. I just wonder where
you feel that biological backup is needed in the surveillance
(Professor Little) I think it is a very important
question and there is no easy answer to that. Currently what happens
is "passive surveillance": you have somebody you are
particularly worried about. Nasty ecoli will give you a nasty
disease and a clinically more severe disease and presumably a
more prolonged disease. I think it would be helpful for GPs to
have guidance about the circumstances in which something unusual
and the situations in which it would be useful to send a sample.
To do it routinely for all the normal gastro-enteritis and normal
urinary or chest infections which can be managed syndromically
would seem like a waste of resources. It is not just the fact
that it is resources, the question is the opportunity cost and
whether GPs would do it properly, but is also the issue that if
you give patients the message that these samples are very important
you are implying that they need to come to see the doctor to have
samples when they have an infection, so you are potentially altering
the threshold of attendance and re-attendance by getting the large
iceberg that we do not see into the medical system. I think syndromic
surveillance would pick up a lot of nasty outbreaks. The other
situationwhich happens at the moment but perhaps we could
improve the guidanceis to improve guidance to GPs about
what constitutes a severe clinical infection and then target the
microbiological tests to those individuals.
377. Not all instances of ecoli 0157 are severe;
some are quite mild. It is just the odd one that may be severe.
It may not be in your practice but in the practice next door.
It is quite tricky.
(Professor Little) I am sure it is. The question is
what are the options, the costs and the disadvantages of all the
378. Could you just describe a little bit about
what goes on in the sentinel practices with regard to sending
in microbiological samples more routinely on typical infections,
rather than only those which appear a little worse than usual?
(Professor Little) I think you are going to be speaking
to Douglas Fleming and he will be able to give you the precise
details. My understanding is that by and large it is clinical
surveillance. GPs fill out clinical pro formas and then they will
only send microbiological investigations in certain situations
for example when there is a trigger or Douglas tells them they
need to. They do not routinely send samples. I suppose, again,
it depends on what you want it for. If you want general information
about what is happening nationally and whether we are getting
more reports in routine samples of ecoli then that sort of system
would give that information. It will not give you information
about what is happening in a locality. It is not a wide enough
system or a comprehensive enough system that you are going to
find out the problem about the nasty ecolis.
379. If the NHS Direct or other sources of information
shows that there is an increase in symptoms which are unusual,
will the sentinel practices then be asked to home in on this and
do more testing than usual?
(Professor Little) At the moment I do not think that
information is being fed back. I suppose it is just collecting
that information, although I think surveillance has started. The
question is how you then act on that information. Whether it would
be cascaded to the local CCDC who then contact the practice or
whether there would be some other cascading system, you would
need to feed it back and have to support the GPs and the practice
nurses in the sending of samples. At the moment I do not know
how that happens. In principle that information could be provided
at a very local level.
(Mrs Howard) As yet in practice within our unit we
do not actually use a lot of the information from NHS Direct because
it has not been captured as such. When that type of information
comes in from other informal reporting systems that we have, what
we then do is we look at it and if it was necessary we would be
writing or faxing out to the GPs in the area and saying that we
are aware that there are cases of whatever in your area. We would
ask them to have a higher index of suspicion, et cetera. We would
give them the information to help them when they see patients
on a regular basis. At the moment that is how we manage it. That
system is in place I would imagine within most CCDC units around