The OzFoodNet Working Group

Introduction

The Australian Government Department of Health established the OzFoodNet network in 2000 to collaborate nationally to investigate foodborne disease. In each Australian state and territory, OzFoodNet epidemiologists investigate outbreaks of enteric infection. OzFoodNet conducts studies on the burden of illness and coordinates national investigations into outbreaks of foodborne disease. This quarterly report documents investigations of outbreaks of gastrointestinal illness and clusters of disease potentially related to food, which occurred in Australia between 1 July and 30 September 2012.

Data were received from OzFoodNet epidemiologists in all Australian states and territories. The data in this report are provisional and subject to change, as additional outbreak data may have been collected following the collation of data for this report.

During the 3rd quarter of 2012, OzFoodNet sites reported 614 outbreaks and clusters of enteric illness, including those transmitted by contaminated food. In total, these outbreaks affected 11,560 people, of whom 268 were hospitalised. There were 49 deaths reported during these outbreaks. This compares with a 5 year mean between 2007 and 2011 of 610 outbreaks affecting 11,825 people, 275 hospitalisations and 39 deaths for the 3rd quarter. Outbreaks of gastroenteritis are often not reported to health agencies or the reports may be delayed, meaning that these figures under-represent the true burden of enteric disease outbreaks. The majority of outbreaks during the 3rd quarter of 2012 (86%, n=531) were due to person-to-person transmission (Table 1), with 57% (304/531) of these occurring in residential aged care facilities.

Table 1: Outbreaks and clusters of gastrointestinal illness reported by OzFoodNet, 1 July to 30 September 2012, by mode of transmission

Transmission mode	Number of outbreaks and clusters	Per cent of total
* Percentages do not add up due to rounding.
Foodborne and suspected foodborne	31	5
Waterborne and suspected waterborne	1	<1
Person-to-person	531	86
Unknown (Salmonella cluster)	7	1
Unknown (other pathogen cluster)	4	1
Unknown	40	7
Total	614	100*

Top of page

Foodborne and suspected foodborne disease outbreaks

There were 31 outbreaks during this quarter where consumption of contaminated food was suspected or confirmed as being the primary mode of transmission. These outbreaks affected 241 people, and resulted in 16 hospitalisations and 1 death. This compares with 31 outbreaks in the 2nd quarter of 2012¹ and a 5 year mean of 30 outbreaks for the 3rd quarter between 2007 and 2011.

Salmonella Typhimurium was identified as the aetiological agent in 9 outbreaks (29%) during this quarter (Table 2). Of the remaining outbreaks, 3 (10%) were due to norovirus, 2 (6%) were due to ciguatera fish poisoning, and 1 each due to scombroid poisoning, Shiga toxin-producing Escherichia coli (STEC), Clostridium perfringens, Listeria monocytogenes, and Campylobacter. In 12 outbreaks (39%), the aetiological agent was unknown.

Table 2: Outbreaks of foodborne or suspected foodborne disease reported by OzFoodNet sites,* 1 July to 30 September 2012 (n=31)

State	Month†	Setting prepared	Agent responsible	Number affected	Hospitalised	Evidence	Responsible vehicles
* No foodborne or suspected foodborne outbreaks were reported by the Australian Capital Territory † Month of outbreak is the month of onset of first case or month of notification/investigation of the outbreak A Analytical epidemiological association between illness and 1 or more foods BT Binary type D Descriptive evidence implicating the suspected vehicle or suggesting foodborne transmission M Microbiological confirmation of agent in the suspected vehicle and cases MLVA Multi-locus variable number tandem repeat analysis PFGE Pulsed-field gel electrophoresis PT Phage type ST Serotype
NSW/Vic.	July	Commercially manufactured	Listeria monocytogenes ST 1/2a,3a, BT 58, PFGE 18A:17A:10	3	3	A	Suspected smoked salmon
NSW	July	Restaurant	Unknown	9	0	D	Unknown
NSW	August	Restaurant	Unknown	10	0	D	Mushroom sauce
NSW	August	Commercial caterer	Salmonella Typhimurium PT 170 / MLVA profile 03-09-08-14-523	14	0	D	Raw egg mayonnaise
NSW	August	Restaurant	Unknown	3	0	D	Unknown
NSW	August	Unknown	S. Typhimurium MLVA profile 03-27-08-21-496	9	0	D	Unknown
NSW	August	Restaurant	S. Typhimurium PT 135 / MLVA profile 03-17-09-12-523	2	0	M	Unknown
NSW	August	Aged care	S. Typhimurium PT 170 / MLVA profile 03-09-07-13-523	3	2	D	Unknown
NSW	August	Restaurant	Unknown	5	0	D	Unknown
NSW	September	Restaurant	Unknown	10	0	D	Unknown
NSW	September	Restaurant	Unknown	5	0	D	Unknown
NT	September	Camp	Shiga toxin-producing Escherichia coli	5	1	D	Kangaroo meat
Qld	July	Restaurant	Norovirus genotype II	6	0	D	Oysters
Qld	July	Restaurant	Unknown	5	0	D	Unknown
Qld	July	Private residence	Scombroid	4	0	M	Fresh mullet fillets
Qld	July	Restaurant	Clostridium perfringens	7	0	A	Lamb curry
Qld	August	Restaurant	S. Typhimurium PT 16 / MLVA profile 03-13-11-11-524	3	3	D	Chicken Caesar salad with raw egg dressing
Qld	August	Private residence	Ciguatera fish poisoning	2	0	D	Coral trout
Qld	September	Private residence	Ciguatera fish poisoning	2	0	D	Coral trout
Qld	September	Aged care	S. Typhimurium PT 16 / MLVA profile 03-13-10-11-524	14	0	D	Unknown
SA	July	Restaurant	Campylobacter	15	1	A	Chicken liver pâté
SA	July	Bakery	S. Typhimurium PT 9	8	3	D	Unknown
SA	September	Restaurant	S. Typhimurium PT 9	11	1	M	Fried ice cream made using raw eggs
Tas.	July	Commercial caterer	Unknown	5	0	D	Unknown
Tas.	July	Commercial caterer	Unknown	10	0	D	Unknown
Vic.	July	Private residence	S. Typhimurium PT 135a	7	0	D	Chocolate mousse containing raw eggs
Vic.	August	Restaurant	Unknown	5	0	D	Unknown
Vic.	August	Unknown	Unknown	2	1	D	Unknown
Vic.	September	Restaurant	Norovirus	43	0	D	Unknown
WA	September	Bakery	Norovirus	4	1	D	Multiple foods
WA	September	Camp	Unknown	10	0	D	Unknown
Totals				241	16

Top of page

Fifteen outbreaks (48% of foodborne or suspected foodborne outbreaks) reported in this quarter were associated with food prepared in restaurants (Table 3).

Table 3: Outbreaks of foodborne or suspected foodborne disease reported by OzFoodNet, 1 July to 30 September 2012 by food preparation setting

Food preparation setting	Outbreaks
Restaurant	15
Private residence	4
Commercial caterer	3
Aged care	2
Bakery	2
Camp	2
Unknown	2
Commercially manufactured	1
Total	31

To investigate the 31 outbreaks, sites conducted 4 cohort studies, 1 case control study and collected descriptive case series data for 24 investigations. No individual patient data were collected for 2 outbreaks. The evidence used to implicate food included analytical evidence in 3 outbreaks and microbiological evidence in 3 outbreaks. Descriptive evidence alone was obtained in 25 outbreak investigations.

The following jurisdictional summaries describe key outbreaks and public health actions that occurred during the quarter.

Australian Capital Territory

There were no reported outbreaks of foodborne or suspected foodborne illness during the quarter.

New South Wales

There were 10 reported outbreaks of foodborne or suspected foodborne illness during the quarter. The aetiological agent was identified in 4 outbreaks; all were due to S. Typhimurium. The aetiology remains unknown for the other 6 outbreaks.

Top of page

Description of key outbreaks

Over an 8 week period, a cluster of cases of S. Typhimurium multi-locus variable number tandem repeat analysis (MLVA) profile 03-09-08-14-523 [phage type (PT) 170] infection were notified to authorities. Nine cases reported eating a beef, egg and mustard sandwich from an event kiosk on the day before their onset of illness. Three other cases reported eating egg sandwiches or an egg salad with mayonnaise at a catered work function. The same catering company supplied food for both events. A further 2 cases reported working directly or indirectly for the catering company and had consumed the foods prepared. The New South Wales Food Authority (NSWFA) inspected the catering premises on 2 occasions and found that the sandwiches contained mayonnaise made by the caterer using raw egg and that it was possible that all the sandwiches consumed by cases contained mayonnaise from the same batch. The caterers are now using a commercial mayonnaise. No batch code details were available to enable trace-back to the farm.

Investigators identified a potential point source outbreak during a review of a cluster of salmonellosis (S. Typhimurium PT 135 MLVA profile 03-17-09-12-523). A group of 49 people attended a function at a restaurant where the menu was buffet style with a variety of poultry, other meat dishes and salads. Two of the 9 people interviewed became ill with nausea, vomiting, abdominal pain and diarrhoea 48 hours after consuming the meal. Both cases were confirmed as having S. Typhimurium (MLVA profile 03-17-09-12-523) but otherwise did not know each other. The NSWFA conducted an inspection of the premises and did not identify any problems. NSWFA attempted to trace back to the chicken suppliers; however this did not yield any extra information. See also Cluster investigations.

Northern Territory

There was one reported outbreak of foodborne or suspected foodborne illness during the quarter.

A group of 7 people from a remote Indigenous community all ate kangaroo that was killed, cooked and eaten in a bush setting.³ Five of the group experienced diarrhoea and vomiting, with three suffering from bloody diarrhoea and subsequently required hospital attention. One patient was admitted to hospital. Stool samples were taken from the 3 cases with bloody diarrhoea and 1 case tested positive for the stx₂ toxin gene produced by STEC. Multiplex polymerase chain reaction testing confirmed this result. No food samples were collected for testing.

Queensland

There were 8 reported outbreaks of foodborne or suspected foodborne illness during the quarter. The aetiological agent was identified in seven of these outbreaks: two each were due to S. Typhimurium and ciguatera fish poisoning; and one each due to C. perfringens, norovirus and scombroid fish poisoning.

Description of key outbreaks

Seven of a group of 12 people who shared a common meal at a restaurant developed diarrhoea and stomach cramps between 3 and 15 hours following a meal that included lamb curry, chicken, beef rendang, and pork. Food samples and environmental swabs from the restaurant kitchen tested negative for bacterial pathogens. A retrospective cohort study identified that persons who consumed lamb curry had an elevated risk of illness [relative risk (RR) 3.0; 95% confidence interval (CI) 1.0 to 9.3; P=0.06]. Four stool samples were positive for C. perfringens with vegetative cell counts ranging between 3.8 x 10⁵ and 5.7 x 10⁷ cells/g and faecal spore counts ranging between 2.1 x 10⁶ and 3.9 x 10⁷ spores/g. C. perfringens isolates were confirmed as Type A, but the C. perfringens enterotoxin gene was not detected. Control measures included thorough cleaning of the restaurant and advice to improve handling and temperature monitoring of cooked foods.

Fourteen residents of an aged care facility experienced illness with symptoms including diarrhoea, vomiting, stomach cramps and fever. S. Typhimurium PT 16 (MLVA profile 03-13-10-11-524) was isolated from the stools of 8 residents. No hospitalisations were reported. A wide variety of food items were consumed in the week prior to illness (including vitamised meals). Extensive environmental samples were collected for microbiological testing as well as food samples including eggs. All samples tested negative for Salmonella. Investigators were unable to identify a food vehicle or source of infection. The MLVA profile and phage types of case isolates suggests that this outbreak may have had a common source of infection with another outbreak investigated during the quarter in which 3 cases of salmonellosis (S. Typhimurium PT 16 MLVA profile 03-13-11-11-524) all consumed Caesar salad. This was prepared using a raw egg dressing at a sporting club restaurant. No common exposures between the 2 outbreaks were found.

Top of page

South Australia

There were 3 outbreaks of foodborne or suspected foodborne illness investigated during the quarter. The aetiological agents were identified as S. Typhimurium PT 9 for 2 outbreaks and Campylobacter for the remaining outbreak.

An outbreak of Campylobacter occurred following a private function in a commercial restaurant. A retrospective cohort study was conducted amongst the 57 party guests.⁴ All of the guests completed the questionnaire, 15 of whom met the case definition. Analysis of the data showed a significant association between illness and the consumption of a chicken liver pâté (RR 16.7; 95% CI 2.4 to 118.6; P<0.001). No other food or beverage served at the party was associated with illness. Three guests submitted stool samples, all tested positive for Campylobacter and one was further identified as Campylobacter jejuni. The environmental investigation noted that the cooking process used in the preparation of chicken liver pâté may have resulted in some portions not being cooked adequately to inactivate potential Campylobacter contamination.

Investigators identified an outbreak of S. Typhimurium PT 9 after 11 cases of gastrointestinal illness were found to be associated with dining at or receiving takeaway food from a particular restaurant. All 11 cases had consumed fried ice cream at the restaurant. Nine of 11 cases submitted faecal samples and all 9 were confirmed as having S. Typhimurium PT 9 infection. An environmental investigation was conducted and S. Typhimurium PT 9 was identified from the breadcrumb plate used to prepare the fried ice cream. Ice cream with raw egg coating was rolled in these breadcrumbs.

Tasmania

There were 2 reported outbreaks of foodborne or suspected foodborne illness during the quarter affecting a total of 15 people. The outbreaks were clustered and were linked to the same catering company but no pathogen could be identified.

Victoria

There were 4 reported outbreaks of foodborne or suspected foodborne illness during the quarter. The aetiological agent was identified for two of these outbreaks with one each due to S. Typhimurium and norovirus.

Investigators were notified of an outbreak of salmonellosis in a family who became ill after sharing a meal that consisted of roast beef, vegetables and chocolate mousse containing raw eggs. All 7 attendees ate the mousse and all became ill approximately 24 hours later. The mousse contained free range eggs. There were no leftover eggs for testing and no batch code details were available to enable trace back to the farm.

An outbreak of gastroenteritis affecting 6 people from a group of 12 who dined at a hotel was reported to a council in regional Victoria. Authorities received a second complaint of illness affecting a large group of 38 people who had dined at the same hotel restaurant a week later. In total, 7 separate groups reported illness after dining at the hotel. Sixty-two interviews were conducted with 3 affected groups and food handlers, and 43 people were identified as cases. Illness was not associated with the consumption of any specific food or drink items. Eight food handlers were interviewed and 6 reported having a gastrointestinal illness but stated that they remained away from work for the recommended time. However, 1 food handler became ill whilst at work and is known to have prepared some food just prior to his illness onset. This food was then served over subsequent days. One faecal sample was submitted by a case and was positive for norovirus. Despite being unable to definitively identify the mode of transmission for this cluster of related outbreaks, the incubation period for each of the affected groups and the high attack rate (74%) for the large group of 38 suggests that cases were exposed to a common source such as food that may have been intermittently contaminated by an infectious food handler or a contaminated environment. Findings that support this hypothesis include the absence of adequate handwashing facilities in the kitchen, and no documented clean up after the first outbreak.

Western Australia

There were 2 reported outbreaks of foodborne or suspected foodborne illness during the quarter. Norovirus was identified as the aetiological agent for one of these outbreaks.

An increase in gastroenteritis in a rural town was reported by hospital staff. Of the 17 people investigated, 12 were from the community and 5 were relatives of a food business owner. Of the 12 community cases, 9 were interviewed and all reported diarrhoea and/or vomiting. Four of these 9 cases had reported eating a range of food items from a food business prior to illness, including bread rolls containing salad (n=3) and a meat pie (n=1). None of the 4 community cases reported contact with an ill person prior to their illness. The median incubation period was 30 hours (range 20–40 hours) and the median duration of diarrhoea was 24 hours. Two cases had faecal specimens positive for norovirus, including one who had eaten food from the food business. Of the 5 ill relatives of the food business proprietor, three presented at hospital with diarrhoea and/or vomiting and at least one of the five was an employee at the implicated food business. The food business owner was given advice on cleaning and sanitising the food preparation areas and exclusion of ill workers. The evidence suggested that illness was due to norovirus and the 4 community cases had suspected foodborne illness due to an ill food handler preparing food. However, as norovirus was also circulating in the community at the time, it is possible that the 4 cases who ate food from the implicated food business may have acquired their illness via person-to-person transmission.

Top of page

Cross-jurisdictional investigation

Through its National Enhanced Listeriosis Surveillance System (NELSS), OzFoodNet identified an outbreak of suspected foodborne illness involving 3 cases of Listeria monocytogenes serotype (ST) 1/2a,3a, binary type (BT) 58, pulsed-field gel electrophoresis (PFGE) pattern 18A:17A:10. NELSS commenced in January 2010 and contained no record of this strain being previously isolated from humans. Cases occurred in January (New South Wales), late May (Victoria) and early June (Victoria). One case died. A case–case analysis using data from NELSS involved 3 cases and 56 controls. In univariate analysis, only smoked salmon had a statistically significant association between consumption and illness; however, as all 3 cases consumed smoked salmon in the 28 days prior to illness onset, an odds ratio (OR) could not be calculated (OR undefined; 95% CI 2.2 to undefined; P=0.02). All 3 cases consumed the same brand of smoked salmon but 2 of the cases also consumed other brands of smoked salmon. The Microbiological Diagnostic Unit at the University of Melbourne has maintained PFGE typing data of Listeria isolates collected from environmental and food samples since 1995. While the dataset is not complete and does not contain a systematic record of food testing, there was no historical laboratory evidence of L. monocytogenes ST 1/2a,3a, BT 58, PFGE 18A:17A:10 in smoked salmon.

Health authorities met with the company of interest and concluded that it had an extensive program for Listeria identification and control in place. The company’s test and hold criteria for fish products exceeded the current microbiological requirements of the Australia New Zealand Food Standards Code.⁵ Twenty samples of ready-to-eat cold smoked salmon of varying brands, batches and date coding sampled from retail sites were all negative for Listeria.

Cluster investigations

During the quarter, OzFoodNet sites investigated multiple clusters of illness due to a range of pathogens, with five being due to S. Typhimurium, two due to other Salmonella serotypes, and one each due to Campylobacter, Cryptosporidium, Giardia and Yersinia enterocolitica. No particular source or transmission mode could be identified in the clusters.

An investigation into a S. Typhimurium cluster with a novel MLVA profile of 03-17-09-12-523 was commenced in New South Wales in July. Phage typing of clinical samples from the cluster identified the organism as S. Typhimurium PT 135. Sixty-nine cases were reported in the quarter, with a total of 41 cases interviewed using a hypothesis generating questionnaire. A trawling questionnaire was completed for 35 cases. Place of residence for cases varied across New South Wales. Frequently consumed foods for cases included fresh pre-cut chicken (88%), fresh beef cuts (60%), carrots (71%), cooked onions (63%), apples (63%) and bananas (57%). Fresh chicken purchased from large supermarket retailers was a feature of the cluster. The NSWFA conducted a trace back investigation based on place of purchase information provided by cases, and identified 3 predominant chicken suppliers. NSWFA approached the 3 suppliers, and one supplier noted that they had seen the same phage type in samples collected during the year and supplied the isolates for MLVA typing. These isolates were found to have the same MLVA pattern as the clinical isolates (MLVA profile 03-17-09-12-523). The supplier subsequently introduced a series of changes in the production process and the number of new cases decreased. One outbreak was identified as a result of this cluster investigation (reported above and in Table 2).

Comments

The majority of reported outbreaks of gastrointestinal illness in Australia are due to person-to-person transmission, and in this quarter, 86% of outbreaks (531/614) were transmitted via this route. The number of foodborne outbreaks this quarter (n=31) was the same as the previous quarter and consistent with the 5-year mean (2007–2011) of 30 outbreaks. S. Typhimurium was identified as the aetiological agent in 9 (29%) of the foodborne or suspected foodborne outbreaks during this quarter (Table 2). In fact, all (9/9) outbreaks with known Salmonella aetiology were due to this serotype. Of the 9 outbreaks where Salmonella was implicated as the responsible agent, 44% (4/9) were associated with raw or undercooked egg products (including raw egg dressings, raw/undercooked egg dessert and fried ice cream).

NELSS was established in 2010 in response to a L. monocytogenes outbreak affecting both an airline and catering company in 2009.^6,7 The surveillance scheme involves serotyping, binary typing and further molecular characterisation, including PFGE, of all clinical isolates of L. monocytogenes. An exposure history for each case is also recorded where available. In this quarter, NELSS allowed the identification of a small Listeria outbreak with just 3 cases in 2 jurisdictions with consumption of a smoked salmon brand common to all cases, indicating that the surveillance system is sensitive.

A limitation of the outbreak data provided by OzFoodNet sites for this report was the potential for differences in how investigators interpreted circumstances and classified and categorised features of the outbreaks. Changes in the number of foodborne outbreaks should be interpreted with caution due to the small number each quarter.

Top of page

Acknowledgements

OzFoodNet thanks the investigators in the public health units and state and territory departments of health, as well as public health laboratories, local government environmental health officers and food safety agencies who provided data used in this report. We would particularly like to thank reference laboratories for conducting sub-typing of Salmonella, Listeria monocytogenes and other enteric pathogens and for their continuing work and advice during the quarter.

OzFoodNet contributors to this report include (in alphabetical order): Robert Bell (Qld), Barry Combs (WA), Rachael Corvisy (Health), Anthony Draper (NT), Emily Fearnley (SA), Tove Fitzgerald (NSW), Gerard Fitzsimmons (Health), Neil Franklin (NSW), Catriona Furlong (NSW), Robyn Gibbs (WA), Joy Gregory (Vic.), Michelle Green (Tas.), Karin Lalor (Vic.), Megge Miller (SA), Cameron Moffatt (ACT), Sally Munnoch (NSW), Jennie Musto (NSW), Amy Parry (SA), Nevada Pingault (WA), Russell Stafford (Qld) and Mark Veitch (Tas.).

Correspondence

Ms Robyn Leader, OzFoodNet, Office of Health Protection, Australian Government Department of Health, GPO Box 9848, MDP 14, Canberra, ACT 2601. Telephone: +61 2 6289 2750. Facsimile: +61 2 6289 1070. Email: ozfoodnet@health.gov.au

Top of page

References

1. OzFoodNet Working Group. OzFoodNet quarterly report, 1 April to 30 June 2012 . Commun Dis Intell 2013;37(1):E73–E78.
2. Wang Q. National harmonization of MLVA typing scheme for Salmonella Typhimurium between Enteric Reference Laboratories in Australia. The Broad Street Pump 2012;27:9–10. Accessed on 18 October 2013. Available from: http://sydney.edu.au/mbi/PDFs/BSP-Feb12.pdf
3. Draper A, Yip T-W. An outbreak of Shiga toxin-producing E. coli (STEC) gastroenteritis associated with eating kangaroo – a case study from the Northern Territory. The Northern Territory Disease Control Bulletin 2013;20(1):14–15.
4. Parry A, Fearnley E and Denehy E, ‘Surprise’: Outbreak of Campylobacter infection associated with chicken liver pâté at a surprise birthday party, Adelaide, Australia, 2012. Western Pac Surveill Response J 2012;3(4):16–19.
5. Australian New Zealand Food Standards Code – Standard 1.6.1 – Microbiological limits for Food. Accessed on 27 June 2013. Available from: http://www.legislation.gov.au/Details/F2012C00862
6. OzFoodNet Working Group. Monitoring the incidence and causes of diseases potentially transmitted by food in Australia: Annual report of the OzFoodNet, 2009. Commun Dis Intell 2010;34(4):396–426.
7. OzFoodNet Working Group. Monitoring the incidence and causes of diseases potentially transmitted by food in Australia: Annual report of the OzFoodNet Network, 2010. Commun Dis Intell 2012;36(3):E213–E241.