Foodborne disease in Australia: incidence, notifications and outbreaks. Annual report of the OzFoodNet network, 2002

In 2002, OzFoodNet continued to enhance surveillance of foodborne diseases across Australia and has expanded its network to cover all Australian states and territories. This annual report was published in Communicable Diseases Intelligence Vol 27, No 2, June 2003, and can be viewed as 6 HTML documents and is also available in PDF format.

Page last updated: 30 June 2003

A print friendly PDF version is available from this Communicable Diseases Intelligence issue's table of contents.

The OzFoodNet Working Group1

Abstract | Introduction | Methods | Author affiliations


In 2002, OzFoodNet continued to enhance surveillance of foodborne diseases across Australia. The OzFoodNet network expanded to cover all Australian states and territories in 2002. The National Centre for Epidemiology and Population Health together with OzFoodNet concluded a national survey of gastroenteritis, which found that there were 17.2 (95% CI 14.5-19.9) million cases of gastroenteritis each year in Australia. The credible range of gastroenteritis that may be due to food each year is between 4.0-6.9 million cases with a mid-point of 5.4 million. During 2002, there were 23,434 notifications of eight bacterial diseases that may have been foodborne, which was a 7.7 per cent increase over the mean of the previous four years. There were 14,716 cases of campylobacteriosis, 7,917 cases of salmonellosis, 505 cases of shigellosis, 99 cases of yersiniosis, 64 cases of typhoid, 62 cases of listeriosis, 58 cases of shiga toxin producing E. coli and 13 cases of haemolytic uraemic syndrome. OzFoodNet sites reported 92 foodborne disease outbreaks affecting 1,819 persons, of whom 5.6 per cent (103/1,819) were hospitalised and two people died. There was a wide range of foods implicated in these outbreaks and the most common agent was Salmonella Typhimurium. Sites reported two outbreaks with potential for international spread involving contaminated tahini from Egypt resulting in an outbreak of Salmonella Montevideo infection and an outbreak of suspected norovirus infection associated with imported Japanese oysters. In addition, there were three outbreaks associated with animal petting zoos or poultry hatching programs and 318 outbreaks of suspected person-to-person transmission. Sites conducted 100 investigations into clusters of gastrointestinal illness where a source could not be identified, including three multi-state outbreaks of salmonellosis. OzFoodNet identified important risk factors for foodborne disease infection, including: Salmonella infections due to chicken and egg consumption, bakeries as a source of Salmonella infection, and problems associated with spit roast meals served by mobile caterers. There were marked improvements in surveillance during 2002, with all jurisdictions contributing to national cluster reports, increasing use of analytical studies to investigate outbreaks and 96.9 per cent of Salmonella notifications on state and territory surveillance databases recording complete information about serotype and phage type. During 2002, there were several investigations that showed the benefits of national collaboration to control foodborne disease. Sharing surveillance data from animals, humans and foods and rapid sharing of molecular typing information for human isolates of potentially foodborne organisms could further improve surveillance of foodborne disease in Australia. Commun Dis Intell 2003;27:209-243.

Top of page


The World Health Organization recently developed a strategy to address the global issue of food safety.1 The strategy highlighted that, 'surveillance is the basis for the formulation of national strategies to reduce food-related risks'. Many countries recognise the importance of improving foodborne disease surveillance due to high incidence and increasing spread of foodborne diseases, particularly in outbreaks.2 While outbreaks may attract media attention and cause community concern, sporadic cases of foodborne disease far outweigh the number associated with outbreaks.3 In addition, foodborne diseases have a major impact on communities and are increasingly affecting trade.4

In 2000, the Commonwealth Department of Health and Ageing (DoHA) established the OzFoodNet to enhance surveillance for foodborne disease.5 OzFoodNet built upon an 18-month trial of active surveillance in the Hunter region of New South Wales and was modelled on the Centers for Disease Control and Prevention's FoodNet surveillance system (see,7 The purpose of enhancing surveillance for foodborne disease in Australia was to investigate, describe and understand foodborne disease at the national level to provide better evidence of how to prevent foodborne illness.

The OzFoodNet network consists of epidemiologists specifically employed by each state and territory health department to conduct 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. The Commonwealth Department of Health and Ageing funds OzFoodNet and convenes a committee to manage the Network.

This is the second annual report of OzFoodNet and covers data and activities for 2002.

Top of page


Population under surveillance

In 2002, the coverage of the network included the entire Australian population, which was estimated to be 19,662,781 persons.8

During 2002, OzFoodNet coverage expanded to include the Northern Territory and all of New South Wales. Prior to this, New South Wales had enhanced surveillance only in the Hunter region.

In 2002, the Hunter site continued to operate as a sentinel for foodborne disease occurrence in New South Wales. The Hunter site conducts thorough local investigation and provides a baseline for foodborne disease incidence in New South Wales. In 2002, the population covered by the Hunter site was estimated to be 544,623 persons.

Data sources

Incidence of gastroenteritis

To determine the burden of gastroenteritis in Australia, the National Centre for Epidemiology and Population Health (NCEPH) conducted a cross-sectional survey between September 2001 and August 2002 on behalf of OzFoodNet. A research company used Computer Assisted Telephone Interviews to interview randomly selected individuals from each state and the Northern Territory. The Australian Capital Territory was included in the sample for New South Wales and there was an over sample in the Hunter region. Respondents were asked whether they had diarrhoea or vomiting in the past four weeks, and about the symptoms related to that episode. Interviewers asked people reporting gastroenteritis in the previous month whether they sought medical care, provided a specimen of faeces for testing, were unable to carry out normal daily activities, or missed paid work.

People were considered to have had 'infectious gastroenteritis' if they:
  • experienced three or more loose stools and/or two or more vomits in a 24 hour period;
  • experienced four or more loose stools and/or three or more vomits in a 24 hour period where they had concomitant respiratory symptoms of respiratory illness; and
  • did not have any non-infectious causes, such as pregnancy, medications, chronic illness, or alcohol consumption as a cause for their illness.
The results were analysed using a generalised regression estimator method and jackknife approach to estimation of standard errors (P Bell, Household Surveys Facilities, Australian Bureau of Statistics). Data were weighted by state, age, sex, the number of phone lines in the house and household size.

Estimating the burden of foodborne disease

To estimate the burden of foodborne disease we used Australian data from various sources and adopted the approach taken by Mead, et al.3 OzFoodNet considered 28 'known' bacterial, viral and parasitic pathogens that can cause infectious gastroenteritis. To estimate the community incidence of these pathogens in Australia, data from the National Notifiable Diseases Surveillance System and state surveillance systems, from outbreak investigations in Victoria (Joy Gregory, personal communication, November 2002), from laboratories and from published results of a longitudinal study of gastroenteritis in Australia were used.9,10,11

Using these data, the literature and a Delphi assessment of Australian foodborne disease specialists, OzFoodNet estimated the proportion of gastroenteritis that was foodborne for each pathogen.12 It was assumed that the proportion of gastroenteritis due to foodborne transmission among the 'unknown' agents was the same as for 'known' agents. The estimate of the proportion of foodborne among all these known pathogens was then used as proxy for estimating the proportion of all infectious gastroenteritis that was foodborne.

To account for inherent uncertainty in the data the potential distribution of the estimates were simulated to give credible intervals, similar to Bayesian inferential techniques.13 OzFoodNet calculated the credible interval of foodborne disease for a 'typical year in Australia-2000'.

Top of page

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 disease to historical totals, OzFoodNet compared crude numbers and rates of notification to the mean of the previous four 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 jurisdiction for June 2002, or the specified year, were used.8 Age specific rates for notified infections in each jurisdiction were calculated.

The date that notifications were received was used throughout this report to analyse notification data. These data are similar to those reported to the National Notifiable Diseases Surveillance System, but individual totals may vary with time and due to different approaches to analysis.

Gastrointestinal and foodborne disease outbreaks

OzFoodNet collected information on gastrointestinal and foodborne disease outbreaks that occurred in Australia during 2002. The reports collate summary information about the setting where the outbreak occurred, the month the outbreak occurred, the aetiological agent, the number of persons affected, the type of investigation conducted, the level of evidence obtained and the food vehicle responsible. To summarise the data, OzFoodNet categorised the outbreaks by aetiological agents, food vehicles and settings where the outbreak occurred. 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 2002 and compared them to previous years. Data from several complementary OzFoodNet studies of foodborne illness in Australia were also examined.

Surveillance evaluation and enhancement

To identify areas where improvements to surveillance are critical, OzFoodNet compared the results of surveillance across different sites, including rates of reporting outbreaks, and investigation of clusters of Salmonella. To measure how well jurisdictions conducted surveillance for Salmonella OzFoodNet examined the completeness of information contained on state and territory databases in 2002. The proportion of notifications with serotype and phage type information were compared with results for the previous two years.

Top of page

Author affiliations

The OzFoodNet Working Group is: (in alphabetical order): Rosie Ashbolt (Tas), Robert Bell (Qld), Barry Combs (SA), Scott Crerar (FSANZ), Craig B Dalton (Hunter PHU), Karen Dempsey (NT), Rod Givney (SA), Joy Gregory (Vic), Gillian V Hall (NCEPH), Brigid Hardy (AFFA), Geoff Hogg (MDU), Rebecca Hundy (NCEPH), Martyn D Kirk (FSANZ), Karin Lalor (Vic), Janet Li (FSANZ), Peter Markey (NT), Tony Merritt (Hunter PHU), Ian McKay (DoHA), Geoff Millard (ACT), Lillian Mwanri (SA), Jennie Musto (NSW), Leonie Neville (NSW), Jane Raupach (SA), Paul Roche (DoHA), Mohinder Sarna (WA), Craig Shadbolt (DoHA), Russell Stafford (Qld), Nola Tomaska (NCEPH), Leanne Unicomb (Hunter PHU), Kefle Yohannes (DoHA), Craig Williams (FSANZ), Jenny Williams (FSANZ)

Corresponding author: Mr Martyn Kirk, Coordinating Epidemiologist, OzFoodNet, Food Safety and Surveillance, Department of Health and Ageing, Canberra ACT 2601. Telephone: +61 2 6289 1555. Facsimile: +61 2 6289 5100. Email:

This article was published in Communicable Diseases Intelligence Volume 27, No 2, June 2003.

Communicable Diseases Intelligence subscriptions

Sign-up to email updates: Subscribe Now