CHAPTER 5 SURVEILLANCE |
5.1 As noted above,
both prudent use of antimicrobials and infection control are informed
by surveillance. Surveillance provides the information on which
to base policies, and by which to assess their effectiveness.
Surveillance is not cost-free; it costs money to gather and transmit
information, and it costs money to collect and analyse it. Witnesses
have proposed various ways to improve surveillance in the United
Kingdom, of disease in general and resistance in particular, both
by getting new information, and by adding value to information
Duty to report
5.2 At present, in
the United Kingdom, certain clinical conditions are notifiable
under the Public Health (Control of Disease) Act 1984 and
the Public Health (Infectious Diseases) Regulations 1988.
Our witnesses point to various shortcomings of this regime.
5.3 First, there is
no requirement to report resistance as such; all current reporting
is voluntary. Dr Mayon-White observes that voluntary informal
reporting "strains slightly the rules of medical confidentiality"
(Q 177). The PHLS would like mandatory reporting, not of
every resistance"That would be far too broad and you
would be overwhelmed with data" (Q 91)but of
certain key resistances either known elsewhere but not in the
United Kingdom, or unknown but anticipated (Q 92). They offered
a list of eight resistances; Dr Mayon-White proposed a list
of just four (Q 195). See Box 8.
|Notifiable diseases under the Public Health (Control of Disease) Act 1984, s.10
| * Cholera
| * Plague
| Relapsing fever
| * Smallpox
|Also required to be notified under the 1984 Act, s.11
| Food poisoning
|Required to be notified under the Public Health (Infectious Diseases) Regulations 1988 (S.I.1988 No. 1546)
| Acute encephalitis
| Acute poliomyelitis
| Dysentery (amoebic or bacillary)
| * Leprosy
| * Malaria
| Meningococcal septicaemia (without meningitis)
| Ophthalmia neonatorum
| Paratyphoid fever
| Scarlet fever
| Typhoid fever
|* Viral haemorrhagic fever
| (including Dengue fever,
| Ebola virus and Lassa fever)
| Viral hepatitis
| Whooping cough
|* Yellow feverProcedure for notification
| Notification under the Act and Regulations is to be made by a doctor to the "proper officer of the local authority" (normally the CCDC), and by him to the District Health Authority. The proper officer must report cases of the diseases marked *, and "any serious outbreak of any disease", to the Chief Medical Officer immediately; and must make weekly and quarterly returns of all notifiable diseases (except leprosy) to the Registrar General.
|Resistances proposed to be notifiable
| PHLS list (Q 92)
| Staphylococcus aureus
| Neisseria meningitidis
| Enterobacteriaceae (not Proteus)
| Streptococcus pyogenes
| Streptococcus pneumoniae
| Mycobacterium tuberculosis
| Vancomycin (i.e. VRSA)
| All â-lactams
| Isoniazid and rifampicin -
| (i.e. MDR-TB)
| Dr Mayon-White's list (Q 195)
| Resistant pneumococci
5.4 Secondly (AMM p 4),
notifiable diseases are required to be reported to the "proper
officer" appointed by the local authority; in practice this
is usually the CCDC of the health authority. There is no requirement
to report to PHLS. Thirdly, the regulations refer to diseases
according to a very antiquated classification. In Dr Mayon-White's
view, they ought to be brought in line with modern microbiology,
and refer to diseases by their causative organism (Q 177).
Finally, one of the notifiable diseases is "food-poisoning".
As Dr Mayon-White explained, "The food poisoning statistics
are a ragbag of conditions", embracing everything from E. coli
to indigestion (Q 183).
5.5 The Department
of Health informed us, "The Government is currently reviewing
the provisions of the Public Health (Control of Disease) Act 1984
(and the regulations made under it) with the intention of bringing
forward new legislation when Parliamentary time allows. Proposals
for the new legislation will be set out in a discussion document
which the Government expects to publish in the Spring. One of
the provisions expected to feature in the new legislation is a
requirement on laboratories to notify certain specified test results.
This would be in recognition of the role laboratory testing now
plays in the diagnosis of disease. It would also complement the
existing system of notification of actual or suspected cases of
certain diseases undertaken by registered medical practitioners.
The definitive list of results to be reported would be determined
in consultation with experts once the legislation has completed
its passage through Parliament, but the Department will consider
the inclusion of test results which can indicate some causes of
hospital-acquired infection or resistance to (certain forms of)
antibiotics. These are likely to include, for instance, test results
indicating MRSA infection..." (letter 14.1.98, cp Q 777).
5.6 Dr Christopher
Bartlett of PHLS, Director of the Communicable Disease Surveillance
Centre, recommended that mandatory reporting of certain resistances
be backed up by "an untoward event reporting system, in which
any unusual or unexpected resistance marker in any micro-organism
is reported. The advantage of this is that you can identify new
antimicrobial resistance events at an early stage and implement
investigations to try and sort out the determinants. Untoward
event reporting has been shown to be very effective in disease
surveillance" (Q 95).
Scotland and Northern
5.7 The PHLS is an
agency of the Department of Health, and operates only in England
and Wales. Scotland has microbiological laboratories, which report
on a voluntary basis to the Scottish Centre for Infection and
Environmental Health (SCIEH) of the NHS in Scotland; the SCIEH
is "broadly analogous" to the PHLS Communicable Disease
Surveillance Centre. PHLS-type services in Northern Ireland are
provided by the Northern Ireland Public Health Laboratory in Belfast.
Surveillance of TB and HIV/AIDS is integrated across the United
5.8 Dr Alasdair MacGowan,
Chairman of the Working Party on Antimicrobial Resistance Surveillance
of the British Society for Antimicrobial Chemotherapy (BSAC),
told us, "There is a problem, in terms of integration of
the various national parts of the United Kingdom, as to what we
should be doing" (Q 97). Professor Brian Duerden,
Deputy Director of PHLS, agreed: "We do need to improve the
links with Scotland and Northern Ireland to get the national picture".
5.9 The Chief Medical
Officer for England and Wales considers liaison to be good, but
sees scope for improvement, e.g. in compatibility of definitions
and data-collection (Q 818). The Scottish Office say, "There
is evidence that reporting may be more comprehensive in Scotland"
than in England and Wales;
a system for electronic laboratory reporting, "fully compatible"
with the PHLS system, is being established in Scotland as recommended
in the Pennington Report; and a recent review has recommended
retaining separate Scottish reference laboratories for Salmonella,
Campylobacter, E. coli 0157, MRSA and meningococci,
rather than merging these functions with PHLS, on ground of both
clinical and cost effectiveness (p 565). The Chief Medical
Officer for Northern Ireland tells us that DHSS (NI) are currently
working on setting up a Regional Communicable Disease Epidemiology
Unit, with a view inter alia to "strengthening links
with PHLS" (p 546).
5.10 Surveillance is
a prime target for applications of information technology, for
both analysis and transmission of data; and many of our witnesses
call for investment in this area (e.g. AMM Q 47, Mayon-White
Q 177). The Chief Medical Officer agreed; he noted the importance
of confidentiality of patient-specific data (Q 777).
Linking and feedback:
the ICARE model
5.11 Value can be added
to surveillance data by using information technology to link them
with other data. Resistance data can be linked with records of
antimicrobial usage; with clinical data from doctors submitting
specimens for analysis (AMM p 5); and with clinical outcome
data, to test the correlation between in vitro and in
vivo (WHO Q 138; Davey p 156, Q 276). SmithKline
Beecham are enthusiastic about the capacity of information from
this "triangle" (usage, clinical outcome and resistance)
to effect "the transformation of antibiotic use" (p 106).
|LINKING AND FEEDBACK IN THE USA
|National Nosocomial Infection Surveillance (NNIS)
|NNIS began in 1970, as a voluntary system to collect demographic and resistance data from certain wards in participating hospitals. It now involves 250 hospitals, all of 100 beds or more; it is still expanding, and becoming more representative of the generality of US hospitals. NNIS is confidential; CDC returns to each hospital its own results, in relation to the overall distribution.
|ICARE takes NNIS data and adds information about antibiotic usage. It currently involves 40 hospitals, each of which receives a nominal $3-4000 to support data collection; it covers 13 "bug-drug" combinations, chosen for their clinical importance. Like NNIS, ICARE is confidential; CDC returns to each hospital its own results, in relation to the overall distribution. In feeding back results to each hospital, the ICARE team at CDC try to identify interventions which can bring down rates of resistance. "One size doesn't fit all": two hospitals may have equally high levels of MRSA; in one case the cause may turn out to be overuse of cephalosporins, in the other proximity to a nursing home with poor infection control. Dr John McGowan, Professor of Infectious Diseases at Emory University, commented that, as a means of tackling resistance, ICARE has the advantage, over general guidelines, of being highly specific and taking full account of local circumstances, including constraints on resources. One limiting factor for ICARE, admitted by CDC, is the usage data; if pharmacy information exists at all, it is often set up to inform billing rather than to analyse usage. Another, pointed out by Dr McGowan, is the absence of community surveillance, which for some bug-drug combinations would be crucial.
5.12 Value can also
be added by feeding surveillance data, with analysis and comparisons,
back to those who provided it. According to the BSAC, "The
PHLS at present does not regularly return analysis of the sensitivity
data it collects on a routine basis to those who provide it"
(p 79). The Minister for Public Health said that the PHLS
is committed "in principle" to improving the situation
5.13 At CDC in the
USA we learned about the US National Nosocomial Infection Surveillance
(NNIS) system, and about Project ICARE (Intensive Care Antibiotic
Resistance Epidemiology): see Box 9. We were most impressed
by ICARE, and wondered how the concept might be transplanted to
the United Kingdom. One way forward might be for a group of United
Kingdom hospitals to ask to participate in ICARE itself; alternatively,
a free-standing United Kingdom project might be supported from
the NHS R&D Budget. Dr Mary Cooke, for the Department
of Health (Q 786), pointed out that UK NINSS (see above,
paragraph 4.28) is less developed than US NNIS, but said that
incorporating prescribing information would make it more useful.
5.14 The PHLS is roughly
half funded by payments from the NHS and other "customers",
with the balance of its budget provided by the Department of Health.
The departmental contribution has been falling, and is projected
to fall further:
The Minister for Public Health acknowledged
that the PHLS is "an indispensable resource in protecting
the public", and that the NHS must be "competent and
adequately resourced" to deal with the problems of resistance;
and she acknowledged the need for improved surveillance. However,
she gave no indication that the Department will find more money
for the PHLS (QQ 768, 779).
Crisis in clinical
5.15 In microbiology
as in other medical specialties, clinical research is led by the
clinical academic staff who combine a medical school appointment
with an NHS contract. However, clinical academic posts in microbiology
are currently being lost: we have heard that chairs are vacant
at St Thomas' Hospital, the Royal London Hospital and King's College
Hospital in London, and in Sheffield and Liverpool. The Chief
Medical Officer acknowledged this problem, and is anxious to resolve
it (Q 814); he pointed to the paradox that this shortage
of specialists has arisen at a time of great scientific excitement
for the discipline. Glaxo Wellcome identify a decline in the capacity
of UK universities to conduct research and produce well-trained
graduates in microbiology; this, they say, is why most of their
antibacterial research is done in Italy (p 408). SmithKline
Beecham likewise identify a "technology gap" and a "lack
of trained people" (p 485).
5.16 SmithKline Beecham
blame underfunding of the discipline "for many years".
The AMM blame the Research Assessment Exercise (RAE), "which
has laid greater value on the more fundamental aspects of research
than those of a more practical and immediately applicable nature"
(AMM p 376; cp Amyes p 544). They point out that, if
funds are forthcoming for the applied research into the epidemiology
and control of resistant organisms which many of our witnesses
say is needed, they will be of limited use without "the very
staff with the capability to lead such research and who have the
credibility to ensure the implementation of any changes in practice".
5.17 We ourselves expressed
concern about the state of clinical academic medicine generally,
in our report Medical Research and the NHS Reforms in 1995
(3rd Report 1994-95, HL Paper 12). The Committee
of Vice-Chancellors and Principals (CVCP) responded by setting
up an independent Task Force, chaired by Sir Rex Richards,
whose report Clinical Academic Careers was published in
July 1997. We met Sir Rex to discuss his report in December 1997,
as recorded in our 3rd Report of this Session (HL Paper 47).
The Task Force found that medicine generally came poorly out of
the RAE; they blamed the conflicting demands of teaching, research
administration and clinical service, coupled with disparity of
reward between clinical academics and NHS staff (3rd Report
1997-98, QQ 5-6). The CVCP has not yet responded to the Richards
Report; but the NHS Executive and the Higher Education Funding
Council for England (HEFCE) have already taken steps to improve
liaison in connection with the RAE. In particular, as announced
in June 1997 in a letter to Vice-Chancellors and others (3rd Report
1997-98, para. 8), they have set up a task group on how the
RAE should treat health services research, and another on the
implications for the RAE of the links between teaching, research
and patient care.
A national strategy
5.18 Professor Finch
says, "At present there is no national systematic monitoring
arrangement prospectively studying trends in resistance, that
might provide robust data on which to make firm judgements [e.g.
of the impact of OTC antibiotics]. Current data is selective in
terms of sampling and is rarely denominator controlled" (p 189).
The ICNA say, "There is currently no standardised national
data collection system in operation in the United Kingdom which
allows for comparison over time and between centres" (p 126).
The AMM swell the chorus of disapproval: "The PHLS network
of 48 laboratories and some NHS and university-run laboratories
do collect bacterial strains of interest on a voluntary basis
to send to central PHLS reference facilities or report to CDSC.
Overall this has been a rather haphazard process and needs to
be done systematically" (p 5, cp Q 5).
They go on, "Systematic collection of epidemiological data
on resistance should be initiated immediately...The costs to the
NHS are likely to be modest compared with many other actions"
5.19 Cost is an issue.
The AMM told us, "Most United Kingdom laboratories are not
resourced to perform surveillance". According to the BSAC,
"Much of the data published in medical journals derives from
pharmaceutically sponsored studies [see ABPI p 177, and Box 11
below]...Although the pharmaceutical industry in conjunction with
private companies, NHS and university laboratories has put considerable
effort into producing good quality surveillance data so far, this
has not resulted in an ongoing surveillance programme. The Government,
the NHS R&D Programme and charities have not funded large-scale
surveillance schemes" (p 79).
5.20 The Royal College
of Pathologists define the need in terms of institutions (p 456).
"We commend the apparently outmoded concept of well-found
diagnostic laboratories alongside appropriate clinical facilities;
these should be strategically sited in selected hospitals, adequately
staffed...and funded...able to carry out appropriate surveillance...with
results nationally and internationally co-ordinated...ready to
define problems when and where they arise; and able to make persuasive
applications for appropriate research funding..."
5.21 The PHLS are not
defensive of the status quo. Dr Bartlett would like
to see "support for the development of surveillance systems
within an overall national surveillance strategy. At least there
needs to be developed a consensus beyond the PHLS, working with
colleagues in the NHS and academia, including clinicians and public
health colleagues also" (Q 107).
5.22 Work on a strategy
for resistance surveillance is in fact in hand. "The BSAC
has recently set up a Working Party on Resistance Surveillance.
It has proposed a multi-level approach to surveillance in the
United Kingdom. Discussions have started with the PHLS as to how
surveillance in the United Kingdom can be improved by a collaborative
arrangement. In addition the BSAC is actively seeking partnerships
with the pharmaceutical industry
and the Wellcome Trust, and has a proactive approach to implementing
its concepts of surveillance during 1998 and beyond" (BSAC
p 76). Details of the BSAC's proposals are given in their
written evidence (p 80). The ABPI support them, and call
on the Government to resource them (p 178). The Minister
for Public Health said, "We support a strategic approach
to this", but was unable to make any commitment as to resources
Though see the evidence of the Scottish Microbiology Association
(p 470) and Dr Brian Watt (p 542). Back
Cp also Amyes and Young, pp 374 and 376. Back
Including Zeneca and SmithKline Beecham. Back