| CHAPTER 8: PUBLIC HEALTH ISSUES:
INFECTIOUS DISEASES |
| |
| 1996 Outbreak of E. Coli O157 Infection in
Central Scotland and the Pennington Group Report |
| |
| The possibility of an outbreak of food
poisoning caused by E. coli O157 was first identified on the afternoon of Friday 22
November 1996, when the Public Health Department of Lanarkshire Health Board became aware
of several cases of infection (some of which had been confirmed by microbiological
testing) in residents of Wishaw in the central belt of Scotland. This outbreak was to
escalate, to involve residents of Forth Valley Health Board as well, and eventually to
become the largest outbreak of infection caused by this organism in the United Kingdom to
date, claiming more lives than any previous E. coli O157 outbreak world-wide. By
the end of the outbreak, a total of 582 people had been affected (379 in Lanarkshire and
118 in Forth Valley), 160 had been hospitalised and 21 had died. |
| |
| As part of the Government’s response to
what was clearly emerging as a serious outbreak of E. coli O157 infection, the
Secretary of State for Scotland established an Expert Group, under the Chairmanship of
Professor Hugh Pennington, on 28 November 1996. The remit of this Group was ‘To
examine the circumstances which led to the outbreak in the central belt of Scotland and to
advise on the implications for food safety and the general lessons to be learned’.
The Group was asked to examine the current knowledge of E. coli O157, taking
account of scientific research in the area, and the adequacy of existing arrangements for,
and guidance on, the handling of food poisoning outbreaks. The Pennington Group convened
from the beginning of December 1996 until the end of March 1997 and met key individuals
involved in the management and control of the central Scotland outbreak, the outbreak in
Tayside which occurred shortly afterwards, and also outbreaks of E. coli O157
infection in Scotland in the recent past. The Group gathered advice and information from a
wide range of sources and presented their interim report and priority recommendations to
the Secretary of State on 31 December 1996; a final report was presented on 8 April 1997.
In all, the Group made 32 wide-ranging recommendations which covered: |
| |
- farms and E. coli O157 in livestock
- practices in hygiene in slaughterhouses
- practices in hygiene in meat production premises and butcher
shops
- food hygiene at the point of consumption
- enforcement
- surveillance
- research
- arrangements for handling and control of outbreaks of
foodborne disease
|
| |
| The Government accepted all the
recommendations in the final report. Many of these have now been implemented, and work
continues on those that are outstanding. Action that has taken place to date includes the
following: |
| |
- revised guidance has been issued to raise awareness among
farm workers of the problems caused by E. coli O157
- action has been taken to ensure that animals presented for
slaughter are clean and to raise standards in meat plants, particularly abattoirs and
cutting plants
- the Codes of Practice under the Food Safety Act 1990 have
been reviewed and Code of Practice 9 has been amended to ensure that high risk premises
(those handling both raw and cooked meats) are visited at least every six months by
environmental health departments. Code of Practice 16 (which deals with the Food Hazard
Warning System) has been amended so that at a local level the risk of a particular food
hazard is assessed not only in relation to the number of people affected, but also in
relation to the virulence of the organism, the extent of distribution of the affected
food, the vulnerability of the consumer group involved and the confidence in public
recall; there is clearer guidance on when central government should be notified of a local
hazard and guidance on the release of information to the media and to the public, making
it clear that consideration of the public health interest should be paramount.
|
| |
| To implement the recommendations of the
interim report, research has been commissioned (including methods of typing E. coli O157
and studies of its prevalence and incidence in farm livestock) and the Guidelines on
the Investigation and Control of Outbreaks of Foodborne Disease have been reviewed. A
Working Party was established under the Chairmanship of Professor Cairns Smith of Aberdeen
University to review, and if necessary revise, this guidance in the light of the
experience of the central Scotland outbreak. This task was almost completed by the end of
1997, with the revised guidance due to go out for consultation in early 1998. |
| |
| In addition, all local authorities and health
boards in Scotland were invited to review their arrangements for outbreak control and to
confirm to The Scottish Office that comprehensive plans were in place. The
Government’s proposals to take forward the Pennington recommendations with respect to
the implementation of HACCP (Hazard Analysis and Critical Control Point) programmes and
the licensing of high risk premises have been the subject of consultation, with final
action still to be decided. |
| |
| Infectious Intestinal Diseases in 1997 |
| |
| Gastrointestinal infections continued to
represent a major cause of morbidity in Scotland in 1997. Table 8.1 lists the numbers of such infections reported by
Scottish laboratories to the Scottish Centre for Infection and Environmental Health
(SCIEH) in 1996 and in 1997. |
| |
| Escherichia coli O157 Infection |
| Despite the central Scotland outbreak, the
rate of reporting of E. coli O157 infection in Scotland changed little between 1996
and 1997, from 9.87 to 8.23 reports per 100,000 population. There were, however,
exceptions in some health board areas: in Borders, the rate rose by 79%, from a high level
of 22.6 per 100,000 population in 1996, when all cases were apparently sporadic, to 40.5
per 100,000 in 1997, when there were several outbreaks. The rate in Lanarkshire has
declined markedly since 1996 but not to pre-outbreak levels. In contrast, the rate in Forth Valley, also affected in the central Scotland outbreak,
is still high (22.9 per 100,000), partly due to an outbreak in the second quarter of 1997.
In total 12 outbreaks of E. coli O157 infection involving more than one family were
identified in 1997, compared with seven in 1996. |
| |
| During 1997 there were considerable changes
in the relative frequency of reported phage types (PT) of E. coli O157. Phage type
2 accounted for 35% of isolates typed in 1997 compared with 63% in 1996 when more than 300
isolates of this phage type were reported from the central Scotland outbreak. In contrast E.
coli O157 PT28 rose from 23% of isolates in 1996 to 37% in 1997, becoming the most
commonly reported phage type. Phage type 8, responsible for the 1997 outbreak in Forth
Valley, accounted for 16% of isolates compared with 2% in 1996. |
| |
| Salmonellosis |
| Identifications of Salmonella species
reported by the Salmonella Reference Laboratory increased by 2.5% between 1996 and 1997 (Table 8.2). Reports of S.enteritidis
rose by 16%; there was, however, a 39% fall in the number of reports of S.typhimurium
DT 104, countering the upward trend of previous years. This phage type is associated with
antibiotic resistance. The greatest increases in reporting rates of Salmonella in
mainland health board areas were in Borders, with a 95% increase, and in Tayside, where
the increase was 31%. In both areas this was largely accounted for by outbreaks. In 1997
Borders had the highest rate of salmonella reports in Scotland, 151 per 100,000
population, more than twice the national average of 65.3 per 100,000. |
| |
| In 1997 17 outbreaks of Salmonella
infection involving more than one family were identified, compared with 14 in 1996. An
outbreak in Perth in March and April resulted in 204 symptomatic cases, 112 of whom had
laboratory confirmation of infection with S.enteritidis. In July an outbreak in Borders
affected 73 people, all of whom had S.enteritidis PT4 infection confirmed. |
| |
| Campylobacteriosis |
| In 1997, reports to SCIEH of Campylobacter
identifications exceeded 5000 for only the second time, and the rate for Scotland exceeded
100 per 100,000 population for the first time ever. Rates varied between health boards
with Lothian having the highest rate of Campylobacter reports, 114 per 100,000
population. During the period 1 April to 30 September 1997, 1738 isolates of Campylobacter
species of human origin were confirmed and typed by the Reference Laboratory. During
November 1997, an increased in incidence of serotype 11 infection occurred, eventually
accounting for 18% of the Scottish total. This trend has been particularly marked in
Grampian, Lanarkshire, and Forth Valley. |
| |
| Only two outbreaks of Campylobacter
infection involving more than one family were identified in 1997, the same number as in
1996. |
| |
| Viral Gastroenteritis and Hepatitis A |
| The number of reports of rotavirus
identification rose by 11%, from 1608 in 1996 to 1758 in 1997; rotavirus is the commonest
cause of diarrhoea identified in those under the age of three years. Numbers of reports of
small round structured virus (SRSV) remained low, declining from 83 in 1996, to 80 in
1997. Even taking account of differences in population size, these numbers are much lower
than those reported in England and Wales, where the totals were 2437 and 2048
respectively. This may reflect under-ascertainment in Scotland. |
| |
| Of adenoviruses, only serotypes 40 and 41 of
subgenus F are recognised as gastrointestinal pathogens. In 1997, only eight of the 716
reports of adenovirus to SCIEH specified a type, and in five of these it was type 40/41.
Clinical details provided are often incomplete and so, while accepting that the figure is
an overestimate of the burden of adenovirus gastroenteritis, Table 8.1 includes all
reports of identification of adenovirus from faeces. Reports of astrovirus and calicivirus
are too infrequent for meaningful interpretation of trends. Reports of hepatitis A
specific IgM fell from 89 in 1996 to 74 in 1997. |
| |
| In 1997, 32 outbreaks of viral
gastroenteritis involving more than one family were identified, compared with 20 in 1996.
Five of these 32 outbreaks were confirmed virologically (four were due to small round
structured virus and one to rotavirus), compared with three in 1996. |
| |
| Other Gastrointestinal Infections |
| Most cases of amoebic and bacillary
dysentery, as well as enteric fevers, cholera, hepatitis A, helminths, and flukes are
acquired abroad. The prevalence of infection due to Shigella sonnei has continued to
decline since the national epidemic of 1991/1992, falling by 46% from 150 reports in 1996
to 81 in 1997. |
| |
| Reports of Aeromonas species in faeces
declined markedly between 1996 and 1997. The importance of this is unclear, as the role of
Aeromonas in the aetiology of infectious intestinal disease is disputed. Reports of Yersinia
species also fell sharply, most of the decline being due to a fall from 51 to 25 in
reports of Yersinia enterocolitica. |
| |
| Surveillance of Outbreaks of Infectious
Intestinal Disease, 1997 |
| SCIEH identified 66 outbreaks of infectious
intestinal disease involving more than one household in 1997, compared with 52 in 1996.
Data are provisional but, so far, summary report forms have been returned on behalf of
consultants in public health medicine documenting 48 (73%) of these outbreaks, compared to
36 (69%) in 1996. Rates of return of report forms varied widely between health boards. A
suspected vehicle of infection was reported in 12 of 19 outbreaks attributed to foodborne
spread; in four of these 12 the suspicion was supported by statistical or microbiological
evidence. |
| |
| Surveillance of Gastrointestinal Infection
in Primary Care in Scotland |
| Collection of data in the survey of
gastrointestinal infections presenting to primary care in Scotland began in April 1996 and
continued throughout 1997. By the end of June 1997, 23 general practices, serving a total
population of 152,000, had submitted reports of 3232 patient consultations for
gastrointestinal infections which met the study case definition; the crude monthly
consultation rate ranged from 125 to 194 per 100,000 patients. The effect of the central
Scotland outbreak of E. coli O157 infection was reflected by increased consultation
rates in November and December 1996 as some practices in areas affected by the outbreak
are included in the study. Only 5% of the 3232 infections reported by general
practitioners had been statutorily notified. Microbiological examination of specimens by
the local laboratory was requested in 36% of cases and a probable pathogen identified in
22% of these, the pathogens most commonly reported being Campylobacter sp and Salmonella
sp. The study will continue until 31 March 1998. |
| |
| Transmissible Spongiform Encephalopathies
(TSEs) |
| |
| On 20 March 1996, the Government announced
that the Spongiform Encephalopathy Advisory Committee had reported to them the
identification of a distinct new variant of Creutzfeldt-Jakob disease (nvCJD), which had
been diagnosed in ten people in the United Kingdom during the previous 14 months. Since
then further cases have been identified, giving a total of 23 by the end of 1997, three of
whom were resident in Scotland (Table 8.3).
|
| |
| The publication of a number of independent
research studies during 1997 provided further evidence in support of the hypothesis that
there is a link between bovine spongiform encephalopathy (BSE) and nvCJD. Studies indicate
that nvCJD is caused by a prion strain which is indistinguishable from the strain which
causes BSE in cattle. Researchers from the Institute of Animal Health in Edinburgh showed
that injecting infectious brain material from patients with nvCJD into inbred strains of
mice produced a disease indistinguishable from that produced by injecting material from
cows with BSE, but clearly distinct from the disease observed in mice exposed to sporadic
CJD material. Studies by the Prion Disease Group at Imperial School of Medicine, London,
in which disease specific prion protein was treated with a protein digesting protease
enzyme, showed that the resultant fragment size profile of the nvCJD agent was
indistinguishable from that of the BSE agent, and again these were distinct from agents
giving rise to other forms of CJD. This research group also carried out transmission
studies in ordinary and transgenic mice which further confirmed the similarity between
nvCJD and BSE. These studies, in combination with complementary surveillance data, provide
compelling evidence of a link between BSE and nvCJD. It is postulated that human exposure
resulted from the consumption of infected offal before it was banned from the human food
chain in late 1989. |
| |
| It remains impossible, however, to predict
accurately the total number of cases of nvCJD which will occur in the UK. The rate of new
cases is not increasing, which provides hope that the numbers will be relatively small,
but much depends on the average incubation period of nvCJD: the longer the incubation
period, the higher the final figure is likely to be. At present this incubation time is
not known, nor is it possible to estimate the infective dose of the prion agent required
to produce disease in humans. In cattle, the number of cases of BSE continues to decline,
in line with the predicted course of this epidemic (Table 8.4). |
| |
| A comprehensive research programme addressing
both human and animal TSEs is currently underway in the UK. A large number of Government
departments and research organisations are involved and mechanisms have been put in place
to ensure effective co-ordination of research priorities and commissioning activities.
Funding in excess of £50m had been committed to the TSE programme by the end of the
fiscal year 1996/97 and investment in excess of £24m from the UK Government funding
agencies is planned during 1997/98. |
| |
| By the end of 1996 a number of significant
measures had been put in place to ensure the safety of the human food supply. Cattle over
30 months of age were excluded from the human food chain, as were specified bovine risk
materials, including the head (with the exception of the tongue), spinal cord, tonsils,
thymus, spleen and intestines of all cattle over six months of age. The heads of sheep and
goats were banned for use in human food, irrespective of age. |
| |
| During 1997, abnormal prion protein was, for
the first time, identified in nervous tissue (the dorsal root ganglia) lying outwith the
brain or spinal cord of affected cattle. Following a detailed risk assessment, in December
1997 the Government announced, as a precautionary measure, a ban on the sale for human
consumption of all beef on the bone from cattle over six months of age. From 1 January
1998, the spinal cord should be removed from the carcasses of sheep and goats more than
one year old before sale, the vertebral column of sheep should not be used in the
production of mechanically recovered meat and the spleen should be removed from all sheep
and goats. |
| |
| At the end of 1997 the Government announced
the start of a Public Inquiry into BSE; this is expected to report during 1999. |
| |
| Respiratory Infections |
| |
| Influenza |
| Influenza A virus has the potential to
undergo major antigenic shifts at unpredictable intervals, resulting in the emergence of
new strains of virus. Worldwide, the population would have had no opportunity to acquire
immunity to such new strains and infection could spread rapidly. A pandemic exists once
human infection with a new virus has been confirmed, and a high incidence of disease, with
rapid international spread, has been observed. |
| |
| The ever-present threat of pandemic influenza
is one of the major reasons for the extensive international influenza monitoring network
co-ordinated by the World Health Organization (WHO). Previous pandemics, in 1918, 1957 and
1968, have resulted in major mortality and morbidity world-wide; the 1918 pandemic was
held to be responsible for over 20 million deaths, more than the total number resulting
from the Great War. |
| |
| In 1997 the United Kingdom was one of the few
countries in the world to have a fully developed national plan for use in the event of a
potential influenza pandemic. The Scottish version of the UK Health Departments’
Multiphase Contingency Plan for Pandemic Influenza lays out the responsibilities of
the various relevant organisations and addresses reduction of morbidity and mortality
through vaccination, drug treatments and minimisation of spread; healthcare for large
numbers of ill and dying patients; maintenance of essential services; and communication
and dissemination of information. In addition, the main roles of the organisations
involved (Scottish Office Department of Health, health boards, SCIEH, NHS laboratories,
GPs) are laid out for each phased stage of Plan implementation. |
| |
| Chinese Avian Influenza |
| In June 1997 a fatal case of influenza with
Reye’s Syndrome was described in a child in Hong Kong; the virus responsible was
identified as an avian influenza A H5N1 strain. Although this was the first description of
human infection with this virus, similar isolated cases of human infection with avian
viruses are known to occur from time to time. With the ascertainment of subsequent cases
in November 1997, the possibility of a pandemic was raised. By the end of 1997, 16 cases
and four deaths had been confirmed, all in Hong Kong residents. The virus strain was found
to be indistinguishable from H5N1 virus isolated from chickens in Hong Kong in the wake of
an ongoing local epidemic of avian influenza. |
| |
| In December 1997 the UK moved to Phase 1 of
the Contingency Plan ie ‘Emergence of a new influenza virus outside the UK’;
Phase 2 applies when there is evidence of person-to-person spread of a new virus outside
the UK. Arrangements for precautionary handling of specimens from patients with
influenza-like illness arriving in the UK from Hong Kong were put in place at the end of
1997. |
| |
| Subsequent studies have suggested that this
H5N1 virus is of low infectivity to humans, but the possibility of genetic reassortment
with other influenza A viruses, leading to higher infectivity, will sustain enhanced
investigation and surveillance of the situation well into 1998. |
| |
| The 1996/97 Influenza Season |
| The Scottish ‘flu spotter’ system
collates reports of ‘flu-like illness’ each year, and currently involves around
90 general practices in 12 health board areas, covering around 10% of the Scottish
population. It is part of a group of early warning systems operating throughout Europe
which contribute data from clinical consultations and from laboratory tests to the
European Influenza Surveillance Scheme. A major change introduced throughout the UK prior
to the 1996/97 season was the implementation of threshold values for classifying the level
of influenza activity (Table 8.5);
England, Wales and Scotland all have their own threshold values because of differences in
the way that clinical ‘flu spotting’ is carried out. |
| |
| Compared with 1995/96, the 1996/97 season was
notable for a higher level of activity and a later seasonal peak (Figure 8.1). At its peak in January 1997, the 1996/97 season
generated ‘higher than expected’ rates for Scotland overall (758 per 100,000
population). Some individual health board areas (Forth Valley, Lothian and Orkney) showed
brief ‘epidemic’ peak rates of >1000 per 100,000 population, but this may
have been due in part to the relatively small number of practices involved. Overall, the
new threshold levels were found to be very useful, particularly in communicating with the
Press. Extrapolation of ‘spotter practice’ data allows us to estimate that some
232,000 people (one in 22 of the Scottish population) consulted their general practitioner
with ‘flu-like illness’ during the 1996/97 season; this was 30,000 more than the
1995/96 estimate. |
| |
| Initially the great majority of laboratory
diagnoses in 1996/97 were of infl}enza A, which characteristically appears earlier in the
season than influenza B. A second wave of infection, this time with influenza B, followed
at the end of January 1997. Typing of virus isolates at the Virus Reference Laboratory at
Colindale showed that all isolates were of strains contained in the 1996-97 influenza
vaccine formulation (influenza A/Wuhan/H3N2 and influenza B/Beijing). The total number of
laboratory confirmed infections for the season (influenza A and B) was the highest on
record. |
| |
| Whooping Cough |
| Whooping cough in Scotland has a four-yearly
epidemic cycle; an increase in cases from baseline level was noted from late 1996 (Figure 8.2). The increase continued in
1997, and by the end of the year there was little to distinguish the current outbreak from
the previous one in terms of cumulative number of cases. From historical data the peak of
the outbreak is predicted at the end of 1998. |
| |
| Whooping cough was in decline until the
mid-1970s when anxieties about possible side effects of immunisation led to a decline in
vaccine coverage to under 50% (Figure 8.3).
This resulted in an immediate increase in whooping cough notifications and, although the
decline in cases has since resumed, around 15 years were lost in what would otherwise have
been a consistent fall in the incidence of whooping cough. The cohort with low vaccination
uptake has contributed a disproportionately large proportion of whooping cough
notifications since, and has now reached childbearing age, late teens to early 20s. It is
a matter of concern that some of these young adults may be more susceptible to infection,
with the possibility of passing infection to their own children who are too young to have
been vaccinated themselves. The advice from the Joint Committee on Vaccination and
Immunisation (JCVI) is that there is no upper age limit for primary vaccination. |
| |
| Pneumococcal Infection |
| 1997 saw the largest number of reports to
SCIEH of septicaemia due to Streptococcus pneumoniae in many years (Figure 8.4). While the incidence of meningitis due to
pneumococcal infection was unusually low (17 cases in 1997 compared with an annual average
of 39 cases over the previous ten years), the reported incidence of invasive pneumococcal
infection (septicaemia plus meningitis) has shown a consistent increase in recent years.
Pneumococcal infection is a relatively common cause of complications following influenza,
and there is good evidence that invasive pneumococcal disease in particular is preventable
by use of pneumococcal vaccine. Splenectomised patients are especially vulnerable to
invasive pneumococcal infection, and there is evidence that a substantial proportion of
this group have not been vaccinated. |
| |
| The Chief Medical Officer’s letter
relating to influenza immunisation, issued in early October 1997, drew attention to the
merits of also offering pneumococcal immunisation to those in high risk groups, including
patients with chronic respiratory, cardiac and renal disease. |
| |
| Legionellosis |
| During 1997, 27 cases of infection with Legionella
were reported to SCIEH by the Scottish Legionella Reference Laboratory, three more than in
1996. Of these, 23 were due to infection with Legionella pneumophila serogroup 1. A
particularly striking feature of the 1997 cases was the high proportion (70%) imported
from abroad; this compares with 62%, 47% and 28% in 1996, 1995 and 1994 respectively. All
but one of these cases were Scottish tourists travelling abroad; the implicated countries
are shown in Table 8.6. |
| |
| One imported case was a Spanish fisherman who
arrived severely ill in a port in north-west Scotland. Laboratory investigation of an
isolate from the patient showed it to be indistinguishable from L. pneumophila
serogroup 1 isolated from the water supply on board his fishing boat. The patient
recovered and measures were instigated to clean and disinfect the water tank on board the
vessel; this latter process was complicated by the design and installation of the tank. A
second case, in a Scottish resident, was linked to a cruise ship in Germany; this proved
to be one of at least six cases associated with this vessel, the other five being English
residents. The Scottish case was the second to be ascertained, and the cluster was
identified by the European Working Group for Legionella Infections (EWGLI) Centre at
Colindale. In all, six of the 19 imported Scottish cases contributed to clusters of
infection identified by EWGLI. |
| |
| Three of the Scottish legionellosis cases in
1997 were nosocomial infections; of the five patients in the ‘community/unknown’
infection category, three had a history of recent travel within the UK. |
