Communicable Diseases Intelligence Volume 28 Issue No 3 - September 2004
Foodborne disease investigation across Australia: Annual report of the OzFoodNet network, 2003
This report published in Communicable Diseases Intelligence Volume 28 Issue, Number 3, contains the annual report of the OzFoodNet network for 2003.
Communicable Diseases Surveillance
This issue - Vol 28 No 3, September 2004
Communicable Diseases Intelligence
A print friendly PDF version is available from this Communicable Diseases Intelligence issue's table of contents.
The OzFoodNet Working Group1
Introduction | Methods | Results | Discussion | Acknowledgements | References
Abstract
In 2003, OzFoodNet conducted enhanced surveillance of foodborne diseases across Australia, which covered all states and territories. During 2003, there were 23,250 notifications of eight potentially foodborne diseases, of which 67 per cent and 30 per cent were due to Campylobacter and Salmonella infections respectively. The most common Salmonella serotype was Typhimurium, as in previous years. Most S. Enteritidis were acquired overseas, except for Queensland where 52 per cent of infections were acquired locally. Locally acquired S. Enteritidis infections in Australia were predominantly due to phage type 26 or reaction does not conform. The most common serotype of shiga toxin producing E. coli was O157, although for 49 per cent of notified infections serotype was unknown due to the use of polymerase chain reaction based screening tests. There were 12 materno-foetal listeriosis infections in 2003, which was an increase compared to recent years. During 2003, there were 444 outbreaks of gastroenteritis and foodborne disease recorded. Ninety-nine of these were of foodborne origin affecting 1,686 persons, hospitalising 105 and causing six deaths. A wide range of agents and foods caused these outbreaks, with Salmonella Typhimurium being the most common pathogen. Outbreaks associated with fish and seafood dishes, poultry meat, and Asian style and imported foods were common. Four outbreaks with international implications were reported: an outbreak of Salmonella in Montevideo involving contaminated tahini from the Middle East and three outbreaks of norovirus infection associated with imported Japanese oysters. Outbreak data indicated a need to monitor food safety in aged care settings, restaurants and catering. Eighty-nine investigations into clusters of gastrointestinal illness where a source could not be identified were conducted, including multi-state outbreaks of salmonellosis. One multistate investigation of antibiotic resistant Salmonella Paratyphi b Java identified 18 cases who had recent exposure to tropical fish aquariums. Ninety-seven per cent of Salmonella notifications on state and territory surveillance databases have complete information on serotype and phage type. In 2003, OzFoodNet demonstrated the benefits of national collaboration to control food borne disease. Commun Dis Intell 2004;28:359–389.
Introduction
Foodborne disease surveillance is a fundamental part of ensuring a safe food supply.1 Many countries have conducted passive surveillance of foodborne diseases through statistics of patient encounters with health systems.2 These systems have several limitations, particularly where the data are based on syndromic diagnoses rather than isolation of microorganisms. To improve the capacity to interpret surveillance data, some countries have collected extra data on patients infected with foodborne illness, or have collected complementary data from animals and foods.3,4
The Centers for Disease Control and Prevention in the United States of America (USA) established the FoodNet active surveillance system in 1995.3 FoodNet consists of 10 sentinel sites across the USA where laboratories report weekly cases of infections that may be transmitted by food. The system has helped to quantify the burden of foodborne illness and understand its causes. FoodNet has provided a major platform for special research into foodborne diseases.
In 2000, the Australian Government Department of Health and Ageing established the OzFoodNet network to enhance surveillance for foodborne disease.5,6 This built upon an 18-month trial of active surveillance in the Hunter region of New South Wales.
OzFoodNet was modelled on the FoodNet surveillance system, although it differs in some important respects.3OzFoodNet:
- does not actively contact laboratories for reports of individual infections (active surveillance), but relies upon Australia’s robust laboratory-based notification system;
- coordinates investigations into outbreaks of national significance;
- collects data on outbreaks of foodborne and gastrointestinal illness due to all modes of transmission;
- covers the whole Australian population; and
- conducts studies of locally important pathogens in different jurisdictions.
OzFoodNet is primarily a national network of epidemiologists that conducts investigations and applied research into foodborne disease. The network involves many different organisations, including the National Centre for Epidemiology and Population Health, and the Public Health Laboratory Network. OzFoodNet is a member of the Communicable Diseases Network Australia (CDNA), which is Australia’s peak body for communicable disease control.7 The Australian Government Department of Health and Ageing funds OzFoodNet and convenes committees to manage the network and oversee the scientific quality of its work.
This is the third annual report of OzFoodNet and covers data and activities for 2003.
Methods
Population under surveillance
In 2003, the coverage of the network included the entire Australian population, which was estimated at mid-year to be 19,662,781 persons (Australian Bureau of Statistics (ABS), June 2004).
OzFoodNet sites were located in every state in 2003, there was an OzFoodNet site in the Hunter Area Health Service of New South Wales, which complemented foodborne disease surveillance across New South Wales. The Hunter site conducts thorough local investigation and provides a baseline for foodborne disease incidence in New South Wales. In 2003, the population covered by the Hunter site was estimated to be 544,623 persons.
Data sources
Rates of notified infections
All Australian states and territories require doctors and/or pathology laboratories to notify patients with infectious diseases that are important to public health. Western Australia is the only jurisdiction where laboratory notification is not mandatory under legislation, although most laboratories still notify the health department. OzFoodNet aggregated and analysed data on patients notified with the following diseases or conditions, a proportion of which may be acquired from food:
- Campylobacter infections;
- Salmonella infections;
- Listeria infections;
- Yersinia infections;
- shiga toxin producing E. coli infections and haemolytic uraemic syndrome;
- typhoid; and
- Shigella infections.
To compare the current rates of disease with previous levels, OzFoodNet compared crude numbers and rates of notification to the means of the previous five years. Where available, numbers and rates of notifications for specific sub-types of infecting organisms were compared to notifications for the previous year.
To calculate rates of notification the estimated resident populations for each state or territory for June 2003, or the specified year, were used (ABS, June 2004). Age specific rates for notified infections in each state or territory were also calculated.
The date that notifications were received was used to analyse notification data. These data are similar to those reported to the National Notifiable Diseases Surveillance System (NNDSS), but may differ for methodological reasons.
Gastrointestinal and foodborne disease outbreaks
OzFoodNet collected information on gastrointestinal and foodborne disease outbreaks that occurred in Australia during 2003. The reports collate summary information about the outbreak setting, the date, the aetiological agent, the number of persons affected, the type of investigation, the level of evidence and the food vehicle. Data on outbreaks due to transmission from animals and cluster investigations were also summarised.
Risk factors for infection
To identify risk factors for foodborne infection in Australia, OzFoodNet reviewed summary data from outbreaks that occurred in 2003 and compared them to previous years. Data from several complementary OzFoodNet studies of foodborne illness in Australia were also examined.
Surveillance evaluation and enhancement
OzFoodNet compared the results of surveillance across different sites, including rates of reporting outbreaks, and investigation of clusters of Salmonella. To measure the quality of national surveillance data, OzFoodNet examined the completeness of information on state and territory databases in 2003. The proportions of Salmonella notifications with serotype and phage type information were compared with results for the previous three years.
Results
Rates of notified infections
In 2003, OzFoodNet sites reported 23,250 notifications of eight diseases that were potentially foodborne. This was a 5.5 per cent increase from the mean of 22,035 notifications for the previous five years. Reports for these eight diseases make up almost a quarter of notifications to the NNDSS Diseases Surveillance System.14 A summary of the number and rates of notifications by OzFoodNet sites is shown in Appendix 1.
Salmonella infections
In 2003, OzFoodNet sites reported 7,032 cases of Salmonella infection, which indicated a rate of 35.4 cases per 100,000 population and a decrease of 4.8 per cent from the mean for the previous five years (Figure 1). The rates ranged from 20.4 cases per 100,000 population in the Hunter region to 187.1 cases per 100,000 population in the Northern Territory.
Figure 1. Notification rates of Salmonella infections for 2003 compared to mean rates for 1998 to 2002, by OzFoodNet site
Overall, notification rates of salmonellosis for 2003 were increased in New South Wales (11.2%) and Victoria (9.2%) compared to historical means. There were moderate declines in the notification rate of Salmonella in all other states and territories, with more significant declines in South Australia (–29.3%) and Western Australia (–24.0%).
The overall ratio of male to female cases was approximately 1:1, ranging from 1.2:1 in Tasmania to 0.8:1 in South Australia and Western Australia. The median age of cases ranged between 17 and 23 years at all OzFoodNet sites, except for the Northern Territory and Queensland where the median ages were three and 12 years respectively. There were no major changes in the median ages of salmonellosis cases from 2002 to 2003.
The highest rate of Salmonella infection was 189.5 cases per 100,000 population in males aged 0 to 4 years of age. The rate was highest in this age group for all sites and ranged from 76.8 cases per 100,000 population in the Australian Capital Territory to 1,432.8 cases per 100,000 population in the Northern Territory. Notification rates were also high in the 5–9 year age group in all jurisdictions. In most jurisdictions there was also a secondary peak in notification rates in the 20–29 year age range for males and females, which was most noticeable in the Northern Territory.
Rates of salmonellosis were highest in northern areas of Australia. The highest rate is consistently reported in the Kimberley region of Western Australia.8,9 Western Australia reported that the Kimberley region had a rate of 314 cases per 100,000 population, with the majority of infections in Indigenous people. In Western Australia, rates of salmonellosis were higher in Indigenous people in all age groups, particularly in children aged 0–4 years of age. Thirty-nine per cent (128/330) of Salmonella notifications in the Northern Territory were in persons of Aboriginal or Torres Strait Island origin. As in previous years, OzFoodNet sites reported that notification rates of salmonellosis increased from south to north along the eastern seaboard of Australia. The rate of notification increased from 25.8 per 100,000 population in Victoria to 106 per 100,000 population in far north Queensland.
During 2003, the most commonly reported Salmonella serotype was S. Typhimurium. There were 714 notifications of Salmonella Typhimurium 135 (including 135a) to OzFoodNet sites making it the most common infection (Table 1). Eighteen per cent (125/714) of these related to a single outbreak in Victoria. There were 678 notifications of this phage type last year. There were 405 notifications of S. Typhimurium 9 in 2003 compared to 583 for the previous year, which represents a 31 per cent drop in this common phage type. S. Typhimurium 170 and S. Typhimurium 108 continued to emerge as a significant phage type around Australia. In 2002, NSW investigators recognised that these two phage types were in fact the same organism after human specimens went to one laboratory for typing and food samples went to another. This explains why certain states and territories never reported cases of S. Typhimurium 170 and others never reported cases of S. Typhimurium 108. In the remainder of this report infections due to this organism are referred to as S. Typhimurium 170/108. There were 382 cases of S. Saintpaul, making it the most common Salmonella serovar following S. Typhimurium. The highest specific rates for single serotypes reported in OzFoodNet sites were S. Ball and S. Saintpaul in the Northern Territory and S. Mississippi in Tasmania with rates of 22.2, 14.1 and 14.7 per 100,000 population respectively.
Salmonella Enteritidis
S. Enteritidis is a serotype that can infect the internal contents of eggs through the oviducts of infected chickens, predominantly with S. Enteritidis phage type 4. People may become infected with this serotype after eating raw or undercooked eggs. This phage type has caused major problems in the northern hemisphere where it has become established in commercial egg laying flocks, although the incidence has declined in many countries.10 Australia is largely free of S. Enteritidis phage type 4 except in people infected overseas. There are other phage types of S. Enteritidis that are acquired locally in Australia, although the causes of these local infections are largely unknown.
OzFoodNet has been conducting a case control study of all locally acquired S. Enteritidis infections in Australia to determine the risk factors for infection. The case control study was established in 2001 and assesses food-based and zoonotic risk factors for infection. These are compared to the exposure histories for up to three age-matched controls per case. Cases infected while overseas are not enrolled in the study, but they are asked about the countries they visited.
During 2003, OzFoodNet sites recorded 227 cases of S. Enteritidis, of which 63 per cent (142/227) had travelled overseas (Table 2). Travel history was unknown for 15% (33/227) of cases, while 23% (52/227) reported no travel out of Australia. Relevant travel histories were difficult to obtain, as people have often travelled to several countries before visiting Australia. Asian countries were commonly mentioned, reflecting that they are common travel destinations for Australians. In the Asian region, cases of S. Enteritidis infection reported travelling to Indonesia and Bali (46%), Singapore (11%), Malaysia (8%), Thailand (7%), and other Asian destinations (6%). Approximately 20 per cent of people acquiring their infection overseas reported travelling to Europe.
The most common phage types isolated varied with the region that the person travelled to. For people returning from Bali and Indonesia, the most common phage types were 6a, 4, and 4b. In Malaysia and Singapore the most common infecting phage types were 1 and 6a, with no phage type 4 reported at all. Travellers to Thailand were also infected with phage types1 and 6a, along with phage type 4. Where cases had returned from Pacific countries, phage type 26 predominated. For travellers returning from Europe, phage types 1, 4 and 6 were most common.
OzFoodNet Sites reported a decrease in the total number of overseas-acquired S. Enteritidis infections, particularly as fewer travellers visited Bali, where this serotype is endemic. There was also a shift in the numbers of different phage types being notified, with Salmonella Enteritidis 4 declining and other phage types such as 6a increasing. Phage type 4b was recognised for the first time in 2002 after reference laboratories commenced testing for this particular phage in September 2001. Isolates of Phage type 4 typed prior to September 2001 have not been retested to determine whether a proportion of them are 4b. It is not possible to say whether there has been a real shift in phage types, from 4 to 4b or whether it is just a result of changed typing methods or changes in travel patterns.
Overall, 23 per cent (52/227) of patients infected with S. Enteritidis acquired their infection in Australia (Figure 2). The median age of cases was 24 years old (age range 0–76 years) and 50 per cent were male. Locally-acquired S. Enteritidis infections were much more common in Queensland than in Victoria, with 52 per cent (39/75) versus four per cent (3/49) locally acquired. Most locally acquired infections in Queensland were due to phage type 26 (Table 3). There was a temporal clustering of cases of S. Enteritidis Reaction Does Not Conform (RDNC) in December 2003, although no common sources were identified (Figure 2). There were no locally acquired cases of S. Enteritidis in Tasmania or the Northern Territory.
Figure 2. Salmonella Enteritidis infections acquired in Australia by phage type and month of notification, 2003
Salmonella Clustering
In total, state and territory health departments conducted 73 investigations into clusters and point source outbreaks of salmonellosis during 2002. A source of infection was identified for 47 per cent (34/73) of these investigations. Approximately 55 per cent (40/73) of these investigations were due to various phage types of S. Typhimurium.
Table 1. Numbers, rates and proportions of the top 10 Salmonella infections, 2002 to 2003, by OzFoodNet site*
| OzFoodNet site | Salmonella type (sero/phage type) | Top 10 infections | |||||
|---|---|---|---|---|---|---|---|
| 2003 n | 2003 Rate† | Proportion (%)‡ | 2002 n | 2003 rate | Ratio§ | ||
| Australian Capital Territory | Typhimurium 135 | 25 |
7.7 |
31.3 |
9 |
2.9 |
2.8 |
| Typhimurium 170 | 4 |
1.2 |
5.0 |
4 |
1.3 |
1.0 |
|
| Typhimurium 6var1 | 4 |
1.2 |
5.0 |
0 |
0.0 |
- |
|
| Typhimurium 9 | 4 |
1.2 |
5.0 |
17 |
5.4 |
0.2 |
|
| Typhimurium U290 | 4 |
1.2 |
5.0 |
3 |
1.0 |
1.3 |
|
| Infantis | 3 |
0.9 |
3.8 |
1 |
0.3 |
3.0 |
|
| Bovismorbificans 14 | 2 |
0.6 |
2.5 |
0 |
0.0 |
- |
|
| Mississippi | 2 |
0.6 |
2.5 |
0 |
0.0 |
- |
|
| Saint Paul | 2 |
0.6 |
2.5 |
0 |
0.0 |
- |
|
| Typhimurium 104L | 2 |
0.6 |
2.5 |
0 |
0.0 |
- |
|
| Hunter | Typhimurium 170 | 10 |
1.8 |
8.9 |
7 |
1.3 |
1.4 |
| Typhimurium 4 | 10 |
1.8 |
8.9 |
2 |
0.4 |
5.0 |
|
| Infantis | 9 |
1.6 |
8.0 |
0 |
0.0 |
- |
|
| Typhimurium U290 | 8 |
1.5 |
7.1 |
8 |
1.5 |
1.0 |
|
| Bovismorbificans 14 | 7 |
1.3 |
6.3 |
2 |
0.4 |
3.5 |
|
| Typhimurium 9 | 7 |
1.3 |
6.3 |
14 |
2.6 |
0.5 |
|
| Montevideo | 6 |
1.1 |
5.4 |
21 |
3.9 |
0.3 |
|
| Birkenhead | 3 |
0.5 |
2.7 |
2 |
0.4 |
1.5 |
|
| Chester | 3 |
0.5 |
2.7 |
4 |
0.7 |
0.8 |
|
| Typhimurium 135a | 3 |
0.5 |
2.7 |
2 |
0.4 |
1.5 |
|
| Typhimurium 197 | 3 |
0.5 |
2.7 |
4 |
0.7 |
0.8 |
|
| New South Wales | Typhimurium 170 | 233 |
3.5 |
12.5 |
148 |
2.3 |
1.6 |
| Typhimurium 135 | 134 |
2.0 |
7.2 |
189 |
2.9 |
0.7 |
|
| Typhimurium 9 | 133 |
2.0 |
7.1 |
255 |
3.9 |
0.5 |
|
| Infantis | 87 |
1.3 |
4.7 |
38 |
0.6 |
2.3 |
|
| Birkenhead | 68 |
1.0 |
3.6 |
94 |
1.4 |
0.7 |
|
| Typhimurium 197 | 68 |
1.0 |
3.6 |
61 |
0.9 |
1.1 |
|
| Virchow | 60 |
0.9 |
3.2 |
74 |
1.1 |
0.8 |
|
| Chester | 40 |
0.6 |
2.1 |
28 |
0.4 |
1.4 |
|
| Typhimurium 12 | 38 |
0.6 |
2.0 |
19 |
0.3 |
2.0 |
|
| Typhimurium 135a | 37 |
0.6 |
2.0 |
48 |
0.7 |
0.8 |
|
| Northern Territory | Ball | 44 |
22.2 |
11.9 |
49 |
24.8 |
0.9 |
| Saintpaul | 28 |
14.1 |
7.5 |
19 |
9.6 |
1.5 |
|
| Anatum | 22 |
11.1 |
5.9 |
15 |
7.6 |
1.5 |
|
| Typhimurium 135 | 17 |
8.6 |
4.6 |
9 |
4.6 |
1.9 |
|
| Chester | 16 |
8.1 |
4.3 |
18 |
9.1 |
0.9 |
|
| Muenchen | 14 |
7.1 |
3.8 |
14 |
7.1 |
1.0 |
|
| Havana | 11 |
5.5 |
3.0 |
3 |
1.5 |
3.7 |
|
| Subsp 1 ser 16:1,v | 11 |
5.5 |
3.0 |
6 |
3.0 |
1.8 |
|
| Adelaide | 10 |
5.0 |
2.7 |
5 |
2.5 |
2.0 |
|
| Weltevreden | 10 |
5.0 |
2.7 |
5 |
2.5 |
2.0 |
|
| Queensland | Saintpaul | 167 |
4.4 |
7.4 |
227 |
6.3 |
0.7 |
| Virchow 8 | 165 |
4.3 |
7.3 |
279 |
7.7 |
0.6 |
|
| Typhimurium 135 | 155 |
4.1 |
6.9 |
110 |
3.0 |
1.4 |
|
| Birkenhead | 109 |
2.9 |
4.8 |
136 |
3.7 |
0.8 |
|
| Chester | 98 |
2.6 |
4.3 |
84 |
2.3 |
1.2 |
|
| Typhimurium 197 | 90 |
2.4 |
4.0 |
31 |
0.9 |
2.9 |
|
| Aberdeen | 75 |
2.0 |
3.3 |
112 |
3.1 |
0.7 |
|
| Hvittingfoss | 72 |
1.9 |
3.2 |
114 |
3.1 |
0.6 |
|
| Typhimurium 170 | 70 |
1.8 |
3.1 |
138 |
3.8 |
0.5 |
|
| Muenchen | 55 |
1.4 |
2.4 |
60 |
1.7 |
0.9 |
|
| South Australia | Typhimurium 108 | 32 |
2.1 |
7.3 |
25 |
1.7 |
1.3 |
| Typhimurium 9 | 28 |
1.8 |
6.3 |
24 |
1.6 |
1.2 |
|
| Chester | 24 |
1.6 |
5.4 |
11 |
0.7 |
2.2 |
|
| Typhimurium 4 | 23 |
1.5 |
5.2 |
7 |
0.5 |
3.3 |
|
| Infantis | 20 |
1.3 |
4.5 |
9 |
0.6 |
2.2 |
|
| Typhimurium 135a | 18 |
1.2 |
4.1 |
15 |
1.0 |
1.2 |
|
| Typhimurium 135 | 17 |
1.1 |
3.9 |
13 |
0.9 |
1.3 |
|
| Typhimurium 12 | 15 |
1.0 |
3.4 |
17 |
1.1 |
0.9 |
|
| Typhimurium 12a | 15 |
1.0 |
3.4 |
15 |
1.0 |
1.0 |
|
| Saintpaul | 13 |
0.9 |
2.9 |
11 |
0.7 |
1.2 |
|
| Anatum | 13 |
0.9 |
2.9 |
1 |
0.1 |
13.0 |
|
| Tasmania | Mississippi | 70 |
14.7 |
50.7 |
78 |
16.6 |
0.9 |
| Typhimurium 9 | 7 |
1.5 |
5.1 |
11 |
2.3 |
0.6 |
|
| Typhiumurium 135 | 6 |
1.3 |
4.3 |
15 |
3.2 |
0.4 |
|
| Saintpaul | 5 |
1.0 |
3.6 |
3 |
0.6 |
1.7 |
|
| Typhimurium 170 | 5 |
1.0 |
3.6 |
0 |
0.0 |
- |
|
| Typhimurium U290 | 5 |
1.0 |
3.6 |
2 |
0.4 |
2.5 |
|
| Typhimurium 4 | 4 |
0.8 |
2.9 |
1 |
0.2 |
4.0 |
|
| Infantis | 3 |
0.6 |
2.2 |
1 |
0.2 |
3.0 |
|
| Typhimurium 126 | 3 |
0.6 |
2.2 |
4 |
0.9 |
0.8 |
|
| Typhimurium 12a | 3 |
0.6 |
2.2 |
0 |
0.0 |
- |
|
| Victoria | Typhimurium 135|| | 233 |
4.7 |
18.4 |
177 |
3.7 |
1.3 |
| Typhimurium 9 | 160 |
3.3 |
12.6 |
151 |
3.1 |
1.1 |
|
| Typhimurium 170 | 125 |
2.5 |
9.9 |
162 |
3.4 |
0.8 |
|
| Typhimurium U290 | 88 |
1.8 |
6.9 |
39 |
0.8 |
2.3 |
|
| Infantis | 54 |
1.1 |
4.3 |
22 |
0.5 |
2.5 |
|
| Typhimurium 197 | 21 |
0.4 |
1.7 |
10 |
0.2 |
2.1 |
|
| Stanley | 19 |
0.4 |
1.5 |
12 |
0.2 |
1.6 |
|
| Typhimurium 12 | 19 |
0.4 |
1.5 |
8 |
0.2 |
2.4 |
|
| Typhimurium 126 | 18 |
0.4 |
1.4 |
61 |
1.3 |
0.3 |
|
| Saintpaul | 17 |
0.3 |
1.3 |
43 |
0.9 |
0.4 |
|
| Western Australia | Typhimurium 135a | 41 |
2.1 |
6.7 |
63 |
3.3 |
0.7 |
| Chester | 36 |
1.8 |
5.9 |
34 |
1.8 |
1.1 |
|
| Saintpaul | 30 |
1.5 |
4.9 |
42 |
2.2 |
0.7 |
|
| Typhimurium 135 | 30 |
1.5 |
4.9 |
30 |
1.6 |
1.0 |
|
| Muenchen | 28 |
1.4 |
4.6 |
27 |
1.4 |
1.0 |
|
| Oranienburg | 21 |
1.1 |
3.4 |
6 |
0.3 |
3.5 |
|
| Mbandaka | 20 |
1.0 |
3.3 |
5 |
0.3 |
4.0 |
|
| Typhimurium 9 | 20 |
1.0 |
3.3 |
45 |
2.4 |
0.4 |
|
| Typhimurium 126 | 17 |
0.9 |
2.8 |
5 |
0.3 |
3.4 |
|
| Senftenberg | 15 |
0.8 |
2.5 |
8 |
0.4 |
1.9 |
|
| Anatum | 12 |
0.6 |
2.0 |
14 |
0.7 |
0.9 |
|
| Infantis | 12 |
0.6 |
2.0 |
18 |
0.9 |
0.7 |
|
* Where there were multiple tenth ranking Salmonella types all data have been shown, giving more than 10 categories for some sites.
† Rate per 100,000 population.
‡ Proportion of total Salmonella notified for this jurisdiction in 2003.
§ Ratio of the number of reported cases in 2003 compared to the number reported in 2002.
|| S. Typhimurium 135 also includes cases of S. Typhimurium 135a
Table 2. Numbers of Salmonella Enteritidis infections acquired overseas and in Australia in 2003, by OzFoodNet site
| OzFoodNet site | History of travel overseas | |||
|---|---|---|---|---|
| Yes | No | Unknown | Total | |
| Australian Capital Territory | 2 |
1 |
0 |
3 |
| New South Wales | 23 |
2 |
15 |
40 |
| Northern Territory | 1 |
0 |
0 |
1 |
| Queensland | 19 |
39 |
17 |
75 |
| South Australia | 17 |
5 |
1 |
23 |
| Tasmania | 3 |
0 |
0 |
3 |
| Victoria | 47 |
2 |
0 |
49 |
| Western Australia | 30 |
3 |
0 |
33 |
| Total | 142 |
52 |
33 |
227 |
Table 3. Number of locally acquired Salmonella Enteritidis infections in 2003, by phage type and state or territory
| Phage type | ACT | NSW | Qld | SA | Vic | WA | Total |
|---|---|---|---|---|---|---|---|
| 26 | 1 |
24 |
25 |
||||
| RDNC* | 5 |
4 |
1 |
10 |
|||
| RDNC/12 | 7 |
1 |
8 |
||||
| 6a | 1 |
1 |
|||||
| Untypable | 1 |
1 |
2 |
||||
| 13 | 1 |
1 |
|||||
| 21 | 1 |
1 |
|||||
| 11b | 1 |
1 |
|||||
| 1b | 1 |
1 |
|||||
| 21b var | 1 |
1 |
|||||
| 4b | 1 |
1 |
|||||
| Total | 1 |
2 |
39 |
5 |
2 |
3 |
52 |
* ‘Reaction Does Not Conform’ (RDNC) represents phage type patterns that are not yet assigned.
No cases were reported from the Northern Territory or Tasmania
Campylobacter infections
Data for campylobacteriosis were not available for New South Wales, including the Hunter Health Area. With this exception, in 2003 OzFoodNet sites reported 15,464 cases of Campylobacter infection, which indicated a rate of 112 cases per 100,000 population. This rate represented a 1.2 per cent increase over the mean for the previous five years (Figure 3). Tasmania, Northern Territory, Victoria and South Australia all recorded a greater than 10 per cent increase in rates of infections compared to the mean of the previous five years. Queensland and Western Australia reported slightly lower rates than for previous years. The highest rates of Campylobacter notification were in South Australia (174.2 per 100,000 population) and the lowest rates were in Western Australia (100.3 per 100,000 population).
Figure 3. Notification rates of Campylobacter infections for 2003 compared to mean rates for 1998 to 2002, by OzFoodNet site excluding New South Wales
The ratio of male to females cases ranged from 1.1:1 in the Australian Capital Territory and Tasmania to 1.3:1 in the Northern Territory. The median ages of cases ranged from 17 to 30 years, except in the Northern Territory where it was five years of age. The highest age specific rates were in male children in the 0–4 year age group, with a secondary peak in the 20–29 year age range for males and females. The highest age specific rates were in males in the 0–4 year age group in the Northern Territory (958 cases per 100,000 population) and South Australia (433 cases per 100,000 population). There were four identified outbreaks of Campylobacter during 2003, two of which occurred in association with visits to farms where school students drank unpasteurised milk and had close contact with animals.
Listeria infections
OzFoodNet sites reported 72 cases of listeriosis in 2003, which represents a notification rate of 0.4 cases per 100,000 population (Figure 4). This was a 17 per cent increase in the number of notifications compared to the historical mean. There were no common source outbreaks of listeriosis detected during the period, although sites investigated instances of temporal clustering of cases identified using Pulsed Field Gel Electrophoresis on isolates.
Figure 4. Notification rates of Listeria infections for 2003 compared to mean rates for 1998 to 2002, by OzFoodNet site
Twelve infections in 2003 (17%) were materno-foetal infections, giving a rate of 4.7 cases per 100,000 live births.11 This represents a considerable increase from two materno-foetal infections in the previous year. Victoria reported five materno-foetal infections during 2003, compared to a total of three cases in the previous three years. Amongst the Victorian cases, 80 per cent (4/5) of the mothers had previously received information about Listeria infection from their doctors. Western Australia had the highest rate with three notifications, although small numbers make rates unstable. The case fatality rate of eight per cent (1/12) for materno-foetal infections was considerably lower than for previous years (Figure 5).
Amongst non-pregnancy related cases, the male to female ratio was 1.1:1. OzFoodNet sites reported that the median ages of non-pregnancy associated cases were between 57–76 years old. The highest age specific rate of 2.1 cases per 100,000 population was in males over the age of 60 years. Twenty seven per cent (16/60) of non-pregnancy associated cases died.
Figure 5. Notifications of Listeria infections showing non-pregnancy related infections and deaths and materno-foetal infections and deaths in Australia, 2000 to 2003
Yersinia infections
In January 2001, the CDNA agreed to stop reporting notifications of Yersinia infections to the NNDSS, due to declines in incidence and lack of identified outbreaks. Victoria has revised regulations to remove yersiniosis from the list of reportable conditions and Yersinia is also not notifiable in New South Wales.
In 2002, OzFoodNet sites reported 117 cases of yersiniosis, which equated to a rate of 1.7 notifications per 100,000 population (Figure 6). The overall rate declined 20 per cent from previous years, when adjusted for the absence of reporting from Victoria and New South Wales. Queensland and South Australia recorded the highest rates of infection, with 2.5 and 1.2 notifications per 100,000 population respectively. Queensland reported 80 per cent (94/117) of all cases and the rates of yersiniosis were similar in all three Queensland health zones. The male to female ratio was approximately 1:1 and the highest age specific rate was 16.6 per 100,000 in 0–4 year olds in Queensland infants.
Figure 6. Notification rates of Yersinia infections for 2003 compared to mean rates for 1998 to 2002, Australia excluding Victoria and New South Wales, by OzFoodNet site
There was one cluster investigation into four cases of Yersinia pseudotuberculosis in South Australia in November 2003. All four children affected were from metropolitan Adelaide and presented with severe abdominal pain. Three underwent surgical intervention resulting in appendectomies. An investigation did not identify any common food or environmental exposure among the infected patients.
While Yersinia notifications have decreased in recent years, continued surveillance for yersiniosis is important to monitor for foodborne outbreaks and the effect of zoonotic control programs. In Queensland, the incidence of yersiniosis has increased each year since 2001 when the lowest rate of 1.5 per 100,000 population was reported.
Shigella
OzFoodNet sites reported 443 cases of shigellosis during 2003, which was a notification rate of 2.2 cases per 100,000 population (Figure 7). This was a 43 per cent decrease in the rate of notification compared with historical averages, after adjusting for the introduction of notifications from New South Wales in January 2001.
Figure 7. Notification rates of Shigella infections for 2003 compared to mean rates for 1998 to 2002,* by OzFoodNet site
* Shigellosis became notifiable in New South Wales from 2001 onwards.
The highest rate of notification was in the Northern Territory (67 cases per 100,000 population), which was 30 times the overall Australian rate. Rates of shigellosis are considerably higher in Indigenous communities. In Western Australia, the rates of shigellosis approached 300 cases per 100,000 population in Indigenous children aged 0–4 years of age.
Overall the notification rate for shigellosis was 43 per cent lower than the mean of the previous five years, and this observed consistently across jurisdictions. The male to female ratio of shigellosis cases was 1:1. The highest age specific rates were in males (12 cases per 100,000 population) and females (11 cases per 100,000 population) in the 0–4 year old age group.
There was an outbreak of Shigella flexneri 2a reported at a school in Victoria in August 2003. There were also community increases of Shigella sonnei biotype A in central Australia during February and March 2003. These increases were noted in Western Australia, South Australia and the Northern Territory. There were no confirmed links with food in any of the outbreaks. In Australia, the majority of shigellosis infections probably were acquired by person-to-person transmission or overseas.
Typhoid
OzFoodNet sites reported 54 cases of typhoid infection during 2003, representing an overall notification rate of 0.3 cases per 100,000 population (Figure 8). The number of notifications was similar to previous years. The highest rates were reported in Western Australia and Victoria with rates of 0.5 and 0.4 cases per 100,000 population respectively. Tasmania, the Northern Territory and the Hunter sites did not report any cases.
Figure 8. Notification rates of typhoid infections for 2003 compared to mean rates for 1998 to 2002, by OzFoodNet site
Travel status was unknown for six cases. Information on phage type was reported for 78 per cent (42/54) of isolates. Where travel status was known, sites reported that 78 per cent (42/54) of cases of typhoid had recently travelled overseas (Table 4). Twenty-two per cent (12/54) of these cases had recently travelled from Indonesia or Bali and the predominant phage types was D2 (6 cases). Fifteen cases had travelled to the Indian subcontinent and the predominant phage type of S. Typhi was E1a (1 case).
There were several cases of typhoid infection that were locally acquired in Australia during 2003. These occurred in Western Australia (5), Victoria (5) and New South Wales (2). This included four cases of S. Typhi C4 infection in Victoria that were contracted from an asymptomatic carrier who prepared food. Three infections in Western Australia were long-term carriers, while one was locally acquired and another case was suspected to have been infected in a laboratory.
Table 4. Travel status for typhoid cases, Australia, 2003
| Country | Number of cases | Phage types |
|---|---|---|
| Afghanistan | 1 |
E1a (1)* |
| Bali | 1 |
D2 (1) |
| Bangladesh | 1 |
O (1) |
| India | 11 |
E1a (1), B1var (1), Degraded (5), 1a (1), O (1), Untyped (1), Untypable (2) |
| Indonesia | 10 |
D2 (5) Untypable (3), Untyped (2) |
| Indonesia/Singapore | 1 |
Untyped (1) |
| Kenya | 1 |
Untyped (1) |
| Lebanon | 2 |
Untyped (2) |
| Nigeria | 1 |
Degraded (1) |
| Pakistan | 2 |
M1 (1), Untypable (1) |
| Philippines | 3 |
B degraded (3) |
| Thailand | 1 |
M1 (1) |
| United States of America | 1 |
Degraded (1) |
| Asymptomatic carrier | 4 |
C4 (1), A (1), A degraded (1), E1a (1), |
| Locally acquired | 4 |
E9 (1), 40 (1), Untypable (2) |
| Infected by a carrier | 4 |
C4 (4) |
| Unknown | 6 |
Untyped (5) E1a (1) |
| Total | 54 |
* Numbers in parentheses represent number of cases infected by the phage type.
Shiga toxin producing E. coli infections
OzFoodNet sites reported 53 cases of shiga toxin producing E. coli (STEC) infection during 2003, compared to 59 for 2002 (Figure 9). This number does not include cases of haemolytic uraemic syndrome where a shiga toxin producing E. coli was isolated. The notification rate of 0.3 cases per 100,000 population was a 13 per cent increase over the mean rate for previous years. South Australia (38 cases) reported the majority of cases and had the highest rate of notification of 2.5 per 100,000 population. All sites reporting cases had an increase in the number of cases notified, except for Victoria and Queensland where there were 38 per cent and 12 per cent declines respectively. There were no cases reported from Tasmania, the Australian Capital Territory or the Northern Territory during 2003. The male to female ratio of cases was 1.1:1, contrasting with a male:female ratio of 0.3:1 in 2002. In 2003, the highest rates of reported infection were in children aged 4–9 years of age.
Figure 9. Notification rates of shiga toxin producing E. coli infections for 2003 compared to mean rates for 1998 to 2002, by OzFoodNet site
Figure does not include Tasmania, the Australian Capital Territory and the Northern Territory, as they did not report cases of shiga toxin producing E. coli infections during 2003.
E. coli O157 was the most common (25%) serotype isolated in 2003, compared to 34 per cent of isolates in 2002. E.coli O111 was the second most common and was responsible for 15 per cent of reports in 2003 (Tables 5 and 6). Almost all cases of E. coli O111 occurred in females in South Australia, which were notified between February and May 2003 (Figure 10). This temporal clustering of cases was related to infections occurring in an aged care facility, which was suspected to be person-to-person spread. Another cluster of E. coli O157 occurred around the same time in metropolitan Adelaide, although no links with any specific foods were identified. Cases of E. coli O157 infection were notified throughout the year. Male cases were significantly more likely than females to be recorded as ‘toxin producing E. coli untyped’ (Table 5, Odds ratio 5.3, 95% C. I.1.4–22.9).
Figure 10. Numbers of notification of shiga toxin producing E. coli infections, by month of onset and serotype, Australia, 2003
* Does not include two cases, one of which was asymptomatic and another with onset of illness in late 2002.
The majority of these untyped infections were the result of positive polymerase chain reaction (PCR) tests for the presence of toxin producing genes, but no culture of E. coli was obtained or specific serotype identified. In South Australia all stools containing macroscopic blood are screened for genes encoding for the production of shiga toxins 1 and 2. Positive specimens are then tested using a multiplex PCR test for the O111, O157 and O28 serotypes, along with various virulence factors. This PCR method is highly sensitive and consequently cultures are often not obtained for further typing. Sixty three per cent (24/38) of cases in South Australia were detected by PCR and no typing details were available.(Table 6)
H typing information was available for only 12 per cent (6/50) of all cases in 2003. There were three E. coli O157:H- infections, one each of serotypes O28:H-, O5:H- and O130:H11.
Surveillance for STEC is strongly influenced by screening practices at laboratories. South Australia has the highest rates of infection with STEC because it screens far more bloody diarrhoea specimens using sensitive PCR tests. The proportion of faecal specimens that is positive for STEC is remarkably similar between state public health laboratories in Australia regardless of the method of detection used.12 In Australia, it is likely that many pathology laboratories do not routinely screen faeces for STEC using Sorbitol MacKonkey agar. Where this agar is used, only E. coli that do not ferment sorbitol are detected. In many other countries the predominant E. coli serotype—O157:H7—is routinely detected on this agar, although it is less common in Australia. E.coli that do not ferment sorbitol only represent a small proportion of this species that produce toxin-based infections.
Table 5. Number of notified cases of shiga toxin producing E. coli, by sex and serotype in 2003, Australia
| Serotype | Number of cases | Total | |
|---|---|---|---|
| Female | Male | ||
| O157 | 8 |
5 |
13 |
| O111 | 7 |
1 |
8 |
| O130 | 0 |
1 |
1 |
| O28 | 1 |
0 |
1 |
| O5 | 0 |
1 |
1 |
| Untypeable | 1 |
1 |
2 |
| Unknown | 3 |
3 |
6 |
| Not Typed | 5 |
16 |
21 |
| Total | 25 |
28 |
53 |
Table 6. Number of notified cases of shiga toxin producing E. coli, by State and serotype in 2003, Australia
| Serotype | Number of cases | Total | ||||
|---|---|---|---|---|---|---|
| NSW | Qld | SA | Vic | WA | ||
| O157 | 0 |
2 |
8 |
1 |
2 |
13 |
| O111 | 1 |
0 |
7 |
0 |
0 |
8 |
| O130 | 0 |
1 |
0 |
0 |
0 |
1 |
| O28 | 0 |
0 |
0 |
1 |
0 |
1 |
| O5 | 0 |
0 |
0 |
1 |
0 |
1 |
| Untypeable | 0 |
0 |
2 |
0 |
0 |
2 |
| Unknown | 3 |
3 |
0 |
0 |
0 |
6 |
| Untyped | 0 |
0 |
21 |
0 |
0 |
21 |
| Total | 4 |
6 |
38 |
3 |
2 |
53 |
Haemolytic uraemic syndrome
There were 15 cases of haemolytic uraemic syndrome (HUS) reported during 2003, corresponding to an overall rate of 0.1 case per 100,000 population. This compared to 13 cases of HUS in 2002. New South Wales reported five of these cases, three of which were notified in the Hunter OzFoodNet Site. Victoria reported four cases, South Australia three cases, and Western Australia, Queensland and the Northern Territory each reported a single case (Figure 11).
The male to female ratio of cases was 1:2. The highest rate of infection was in females aged 5–9 years old and males aged 0–5 years old, which were both 0.5 cases per 100,000 population. Sites reported that STEC were isolated from faeces in 20 per cent (3/15) of cases. One case was due to the O157 serotype, while two others were STEC unspecified. There was no obvious clustering of cases in 2003.
Figure 11. Numbers of notifications of haemolytic uraemic syndrome, by month of notification and jurisdiction, Australia, 1998 to 2003
Gastrointestinal and foodborne disease outbreaks
During 2003, OzFoodNet sites reported 444 outbreaks of gastrointestinal illness affecting 10,368 persons. Five-hundred and one people were hospitalised and 10 people died as a result of these outbreaks. Fifty seven per cent (255/444) of outbreaks were suspected to be spread from infected persons to other people (Figure 12).
Outbreaks of gastroenteritis spread by person to person contact were responsible for 71 per cent (7,388/10,368) of all persons affected by gastro outbreaks, and three deaths. Fifty three per cent (135/255) of person to person gastro outbreaks were reported in aged care facilities, while 16 per cent (41/255) and 11 per cent (27/255) of outbreaks were reported in hospitals and childcare settings. Forty three per cent (109/255) of person-to-person outbreaks were due to norovirus, while 48 per cent (123/255) were of unknown aetiology many of which would have been viral.
Sites conducted investigations into 89 different clusters where the mode of transmission was not determined, or a foodborne source was not identified.
Figure 12. Outbreaks of gastrointestinal and foodborne disease, Australia, 2003
Foodborne disease outbreaks
In 2003, 99 foodborne disease outbreaks affected 1,686 persons, hospitalised 105 persons and caused six deaths (Table 7). This equates to an overall rate of 5.0 outbreaks of foodborne disease per million population. Appendix 2 shows a summary description of each outbreak.
Queensland reported the largest number of outbreaks, which represented 30 per cent (30/99) of all outbreaks reported (Table 7). The reporting rates of foodborne outbreaks for different OzFoodNet sites ranged from 0.7 per million population in Western Australia to 35.3 per million population in the Northern Territory. The majority of outbreaks occurred in summer and autumn (Figure 13).
Table 7. Outbreaks of foodborne disease in Australia, 2003, by OzFoodNet site
| State | Number of outbreak | Number affected | Hospitalised | Deaths | Mean number cases per outbreak | Outbreaks per million population |
|---|---|---|---|---|---|---|
| Australian Capital Territory | 3 |
35 |
7 |
1 |
11.7 |
9.3 |
| New South Wales | 29 |
521 |
29 |
1 |
18.0 |
4.3 |
| Northern Territory | 7 |
110 |
4 |
0 |
15.7 |
35.3 |
| Queensland | 30 |
311 |
28 |
2 |
10.4 |
7.9 |
| South Australia | 1 |
6 |
1 |
0 |
6.0 |
0.7 |
| Tasmania | 1 |
22 |
2 |
0 |
22.0 |
2.1 |
| Victoria | 20 |
499 |
27 |
1 |
25.0 |
4.1 |
| Western Australia | 8 |
182 |
7 |
1 |
22.8 |
4.1 |
| Total | 99 |
1686 |
105 |
6 |
17.0 |
5.0 |
Figure 13. Outbreaks of foodborne disease, by selected aetiological agents, Australia, 2001 to 2003
Aetiological agents
The most common agent responsible for foodborne disease outbreaks was Salmonella, which caused 31 per cent (31/99) of outbreaks (Table 8). These outbreaks affected a total of 710 persons with a hospitalisation rate of 12 per cent (82/710). S. Typhimurium was responsible for 81 per cent (25/31) of foodborne Salmonella outbreaks. Five of 6 fatalities were reported from four separate outbreaks of S. Typhimurium.
Table 8. Aetiological agents responsible for foodborne disease outbreaks showing number of outbreaks and numbers of persons affected in Australia, 2003
| Agent | Number of outbreaks | Number affected | Average size of outbreak | Hospitalised | Deaths |
|---|---|---|---|---|---|
| Campylobacter sp. | 3 |
34 |
11.3 |
1 |
0 |
| C. perfringens | 5 |
116 |
23.2 |
0 |
1 |
| Ciguatera | 10 |
47 |
4.7 |
5 |
0 |
| Escolar | 2 |
23 |
11.5 |
0 |
0 |
| Hepatitis A | 2 |
24 |
12.0 |
2 |
0 |
| Histamine poisoning | 4 |
29 |
7.3 |
2 |
0 |
| Norovirus | 9 |
258 |
28.7 |
0 |
0 |
| S. aureus | 2 |
21 |
10.5 |
4 |
0 |
| S. Other | 6 |
38 |
6.3 |
4 |
0 |
| S. Typhimurium | 25 |
672 |
26.9 |
78 |
5 |
| Sorbic acid | 1 |
23 |
23.0 |
0 |
0 |
| Unknown | 30 |
401 |
13.4 |
9 |
0 |
| Total | 99 |
1686 |
17.0 |
105 |
6 |
Many of the 21 outbreaks of illness due to toxins in 2003 were associated with contaminated fish. While ciguatera poisoning caused 48 per cent (10/21) of these outbreaks, all were small with a mean of 4.7 persons affected and a hospitalisation rate of 11 per cent. In comparison, four outbreaks of ciguatera poisoning in 2002 resulted in 50 per cent (7/14) of cases requiring admission to hospital. Histamine poisoning from contaminated fish caused four outbreaks affecting 29 people. There were two outbreaks of kerriorhea due to consumption of Escolar fish, which affected 23 people. Escolar fish contain high levels of indigestible wax esters, which result in excessively oily stools. They are not considered toxins per se, but may result in outbreaks of oily diarrhoea.
There were five outbreaks Clostridium perfringens intoxication and two of Staphylococcus aureus intoxication. One outbreak of C.perfringens resulted in a death in a nursing home resident. There were three outbreaks of Campylobacter that affected 34 people.
There were eleven outbreaks of known viral aetiology, nine of which were due to norovirus. These outbreaks of norovirus affected 258 persons, but no one was hospitalised. The other outbreaks of viral illness were due to hepatitis A, which affected 24 persons. Thirty per cent (30/99) of outbreaks were of unknown aetiology, which affected 401 persons including nine cases who were hospitalised.
Food vehicles
There was a wide variety of foods implicated in outbreaks of foodborne disease during 2003 (Table 9), although investigators could not identify a vehicle for 34 per cent (34/99) of outbreaks. Contaminated fish was the most common food vehicle and was responsible for 17 per cent (17/99) of outbreaks. Poultry were responsible for or suspected as the cause of eight outbreaks, while pork was responsible for four outbreaks. Egg dishes, oysters, sandwiches, rice dishes and mixed foods were implicated in three outbreaks each. There was one outbreak associated with Vietnamese pork rolls and one associated with contaminated tahini from the Middle East. Outbreaks involving egg dishes had a hospitalisation rate of 20 per cent (23/117) and resulted in two deaths.
Outbreak settings
The most common setting for the occurrence of outbreaks was at restaurants (34%), followed by the home (20%), events catered for by professional companies (14%) and aged care facilities (9%) (Table 10). There were six outbreaks in school camps or excursions and three outbreaks each community settings and take away food stores.
Table 9. Categories of food vehicles implicated in foodborne disease outbreaks in Australia, 2003
| Vehicle category | Number of outbreaks | Number affected | Hospitalised | Deaths |
|---|---|---|---|---|
| Beverage | 1 |
19 |
0 |
0 |
| Cakes | 2 |
73 |
1 |
0 |
| Cheese | 1 |
23 |
0 |
0 |
| Dessert | 1 |
31 |
0 |
0 |
| Egg dishes | 1 |
52 |
4 |
0 |
| Escolar fish | 3 |
45 |
0 |
0 |
| Fish (other) | 141 |
57 |
7 |
0 |
| Mixed vehicle | 3 |
62 |
10 |
0 |
| Oysters | 3 |
100 |
0 |
0 |
| Pasta | 2 |
29 |
0 |
0 |
| Pizza | 1 |
18 |
0 |
0 |
| Pork | 4 |
57 |
3 |
0 |
| Pork rolls | 1 |
213 |
22 |
1 |
| Poultry | 5 |
98 |
8 |
0 |
| Red meat/meat products | 1 |
7 |
0 |
0 |
| Rice dish | 3 |
47 |
5 |
1 |
| Sandwiches | 3 |
38 |
0 |
0 |
| Seafood | 2 |
21 |
0 |
0 |
| Sesame seed products | 1 |
3 |
0 |
0 |
| Unknown | 34 |
483 |
18 |
2 |
| Unpasteurised milk | 1 |
13 |
0 |
0 |
| Asian foods | 2 |
40 |
3 |
0 |
| Salad | 2 |
26 |
1 |
0 |
| Sliced meats | 1 |
1 |
0 |
0 |
| Suspected poultry | 3 |
37 |
2 |
0 |
| Suspected egg dishes | 2 |
65 |
19 |
2 |
| Raw vegetables | 2 |
28 |
2 |
0 |
| Total | 99 |
1686 |
105 |
6 |
Table 10. Categories of settings where food was prepared or consumed for foodborne disease outbreaks, Australia, 2003
| Setting category | Number of outbreaks | Number affected | Hospitalised | Deaths |
|---|---|---|---|---|
| Aged care facility | 9 |
167 |
17 |
3 |
| Bakery | 1 |
213 |
22 |
1 |
| Community | 4 |
35 |
3 |
0 |
| Home | 20 |
115 |
12 |
0 |
| Hospital | 2 |
22 |
10 |
1 |
| Institution | 2 |
49 |
7 |
0 |
| Restaurant | 34 |
619 |
25 |
1 |
| School | 1 |
19 |
0 |
0 |
| Take away | 4 |
67 |
0 |
0 |
| Camp/Excursion | 6 |
80 |
6 |
0 |
| Caterer | 14 |
264 |
3 |
0 |
| Childcare | 2 |
36 |
0 |
0 |
| Total | 99 |
1686 |
105 |
6 |
Investigative methods and levels of evidence
States and territories investigated 26 outbreaks using retrospective cohort studies and nine outbreaks using case control studies. Cohort studies were conducted for only 31 per cent (8/26) outbreaks of unknown aetiology, compared to 50 per cent in 2002. Twenty seven percent of investigations using cohort studies were for norovirus outbreaks.
To attribute the cause of the outbreak to a specific food vehicle, investigators obtained analytical evidence from epidemiological studies for nine outbreaks. Microbiological evidence of contaminated food was found in nine outbreaks, with a further seven outbreaks investigations obtaining both microbiological and analytical evidence. Investigators obtained analytical and/or microbiological evidence for 39 per cent (12/31) of Salmonella outbreaks, compared with 52 per cent for 2002. Seventy-four per cent (74/99) of outbreaks relied on descriptive evidence to implicate a food or foodborne transmission.
Significant outbreaks
There were five outbreaks affecting 50 or more persons in 2003, compared to six in 2002. Three were due to Salmonella Typhimurium, one to norovirus and one was of unknown aetiology. Three of the outbreaks occurred at restaurants, one was associated with a bakery and one with a commercial caterer. One of the outbreaks of S. Typhimurium occurred at a restaurant and was associated with dishes containing eggs, while another was associated with Vietnamese rolls from a bakery. The third outbreak was due to pigeon meat contaminated with with S. Typhimurium 99. Apple strudel served at a restaurant was responsible for a large outbreak of norovirus. The catering associated outbreak did not identify a food vehicle or aetiological agent.
There were 23 outbreaks affecting between 20 and 50 persons. Three of these outbreaks occurred in aged care facilities and were due to C.perfringens or Salmonella. Seafood was implicated in four of these outbreaks, including two due to oysters from Japan contaminated with norovirus. Nine outbreaks were due to Salmonella Typhimurium, of which phage types 135 (4 outbreaks) and 170 (2 outbreaks) were the most common causes. Two of these S. Typhimurium outbreaks were due to roast pork, while three were related to Asian foods.
The Tasmanian Department of Health and Human Services investigated an outbreak of hepatitis A following a four-day festival in the Northern Territory in April 2003. Four notifications of hepatitis A infection triggered an investigation involving OzFoodNet Sites in Queensland, New South Wales, Victoria, Western Australia and Tasmania. A retrospective cohort study of 213 out of 350 people attending the event identified 21 cases of hepatitis A. People who consumed cordial or coleslaw were at higher risk of developing hepatitis A. All food handlers tested were found to be negative for hepatitis A IgM and environmental investigations did not reveal any cause of the outbreak.
There were four outbreaks involving foods imported into Australia, highlighting the international implications of foodborne disease. One outbreak of S. Montevideo in Victoria affected 3 persons and was linked to tahini from Lebanon. This followed a previous outbreak in New South Wales in 2002, with cases of S. Montevideo being reported well into 2003. In total, there were 58 cases of S. Montevideo from these two separate outbreaks associated with sesame seed products in New South Wales and Victoria. As a result there were several local recalls of contaminated tahini and helva along with an international alert to investigators. The international alert identified a further ten human infections in New Zealand and assisted food safety agencies in Canada and the United Kingdom to identify contaminated food products, with subsequent recalls of contaminated tahini and helva in these countries.
Salmonella contamination of sesame seed based products continues to be a problem worldwide. Australian and New Zealand food safety authorities have implemented routine microbiological testing for imported foods containing crushed sesame seeds. However, since detecting low concentrations of sporadic contamination with Salmonella from random testing is very difficult, human health surveillance of Salmonella infections plays a vital role in ensuring the safety of these types of products.
There were three other outbreaks with international implications in Western Australia (2) and Northern Territory (1) in November 2003 that were associated with oysters. The oysters were Individually Quick Frozen (IQF) meat imported from Japan, although they were different importers and brands. The labelling on some of these oyster products indicated the need to ‘cook before consumption’. Despite this, the two outbreaks in Western Australia were both due to caterers using the oyster meat uncooked in ‘oyster shooter’ cocktails. The third outbreak in the Northern Territory occurred in a popular restaurant where the oysters were cooked for 8–10 minutes. Norovirus was detected in patients’ faeces in two of these outbreaks and suspected as the cause of the third. Norovirus was also detected in the oyster meat in the outbreak that occurred in the Northern Territory. Traceback investigations identified that all oyster products were supplied by a single company in Japan and the two batches were harvested at similar times.
Cluster investigations
A cluster is defined as an increase in infections that are epidemiologically related in time, place or person where investigators are unable to implicate a food vehicle or determine a mode of transmission. An example is a temporal or geographic increase in the number of cases of a certain type of Salmonella serovar or phage type. In this category, some outbreaks where the mode of transmission was indeterminate have been included.
During 2003, states and territories conducted 89 investigations of clusters of enteric diseases that affected 1,298 people, hospitalising 88 people and causing one fatality. Investigators were unable to determine the mode of transmission or source of infections for these clusters, which were due to organisms such as Salmonella, Campylobacter and hepatitis A. These clusters do not include all investigations conducted at the State, Territory or public health unit level, but the number is indicative of the effort to investigate enteric diseases in Australia. Forty-four per cent (39/89) of these investigations related to clusters of Salmonella, which affected 427 persons with 33 cases hospitalised. S. Typhimurium was responsible for 36 per cent (14/39) these Salmonella cluster investigations. Of the remaining 25 investigations, 17 other different Salmonella serovars were involved.
Many of the cluster investigations were suspected to be related to animal or food-based exposures, which could not be confirmed. An example was an investigation into four cases of S. Reading in Queensland in February 2003. It was likely that most of t