---

A national survey of the incidence of gastroenteritis in Australia and associated health-seeking behaviour, and to collect other information, was conducted for 12 months over 2001–02. The study was a representative, retrospective, cross-sectional survey across all states.

Data were collected by computer-assisted telephone interviews. The sample frame was all people living in residential households with a land telephone line. Households were selected by random-digit dialling.

The person in the household with the most recent birthday was interviewed. If the selected respondent was not at home, nine further attempts were made to contact the person at different times of the day before moving on to the next randomly selected respondent. The response rate was 67% of contacted households.

All respondents were asked about vomiting and diarrhoea, chronic illness, food safety perceptions, demographics and socioeconomic status. Respondents reporting diarrhoea or vomiting were asked for more details on symptoms and timing, health care use, investigation and treatment practices, and the effect of their illness on work and activities.

Case definition of gastroenteritis

The definition of gastroenteritis used in this study includes cases of moderate to severe illness (at least three loose stools or two vomits in 24 hours). Although there is no internationally agreed standard definition, this is consistent with most definitions of gastroenteritis that are based on a severity of three or more loose stools in 24 hours. Cases where persons identified a non-infectious cause of their symptoms (such as pregnancy, alcohol or chronic illness) were excluded. To minimise the influence of respiratory infections that might have concomitant gastrointestinal symptoms, a higher level of gastrointestinal symptoms, of at least four loose stools or three vomits, was required if respiratory symptoms were reported.

The data from the gastroenteritis study were weighted to the Australian population by age, sex and household size, and extrapolated to the whole population. The results of the survey gave an estimated 17.2 million cases (95% CI: 14.5-19.9 million cases) of moderate to more serious gastroenteritis in one year in Australia.

It is important to recognise that a stricter case definition would include only more severe cases, while a less strict definition would include more cases with a lesser level of illness. The survey showed that health care behaviour and time off work are related to the severity of the illness, with the more severe cases seeking health care and experiencing interruptions to work and activity. This means that regardless of the strictness of the definition used, the total population costs associated with gastroenteritis remain basically steady. It is the average cost per case that will change according to the definition, as this influences the total number of cases. Data presented as ‘cost per case’ should be interpreted with this in mind.

Proportion of gastroenteritis that is foodborne

The proportion of gastroenteritis due to contaminated food was estimated at 32% (95% CrI: 24-40%). This gives an estimated 5.4 million cases of foodborne gastroenteritis, including moderate and more serious cases in Australia each year (95% CrI: 4.0-6.9 million cases). Hall and colleagues (2005) show how this estimate was derived.

Visits for health care and treatment

The community survey gave estimates for visits to GPs and emergency departments, and for medications taken. There were so few admissions to hospital in the survey that the estimates for hospital admissions were not robust (with a large standard error). This means that even one more or less person attending hospital in the community survey could have significantly influenced the estimate of the number of hospital admissions. Accordingly, the National Hospital Morbidity Database was used for estimating hospitalisations as being a more reliable data source for this item.

The number of stool tests reported in the survey was also low, with resultant wider standard errors. The estimate was validated by comparison with Health Insurance Commission data that indicate a similar number of stool tests for the same period, 2001-02.

Time lost

Time lost due to the illness, for both paid work and other activities, was estimated from the survey. This included time lost by either the person with gastroenteritis or another person looking after the ill person.

Top of page

Mortality

Death due to foodborne illness is a rare event in Australia. The National Hospital Morbidity Database 1993 to 1999 was used to estimate the number of deaths per year due to gastroenteritis. All hospital admissions where infectious gastroenteritis was either the main reason for hospital admission or a contributing reason (principal or one of nine additional diagnoses) were examined. Of those people who died, gastroenteritis was likely to have contributed to death, although it was not necessarily the only or main reason.

A summary of estimates for gastroenteritis is shown in Table 3.1. Details are in Appendix A.

Table 3.1 Estimates of illnesses due to foodborne transmission in Australia in one year. Estimated numbers of cases, deaths, health care visits, treatments and time lost

Foodborne illness	Percentage of population due to foodborne transmission (95% CrI)(a)	No. cases due to foodborne illness/year (95% CrI)(b)	Deaths/ year	Hospitalisations/ year	Mean days in hospital/ patient	Visits to GPs/ year	No. of days cases in lost paid work and activities/ year(c)
Gastroenteritis	32 (24–40)	5.4 million (4.0–6.9)	80	15,000	2	1.4 million	5.8 million
Listeriosis	98 (92–100)	120 (100–130)	26	120	23	180	3,000
Toxoplasmosis	35 (0–71)	5,900 (0–13,900)	0	21	8	1,200	9,020
Hepatitis A	10 (0–24)	150 (0–1,000)	1	24	4	540	6,020
Haemolytic uraemic syndrome	50 (0–100)	20 (0–40)	3	30	9	45	290
Irritable bowel syndrome	30 (10–60)	20,200 (6,400–35,800)	3	2,700	2	91,700	49,500(d)
Guillain-Barré syndrome	20 (15–25)	120 (90–160)	12	120	13	770	9,900
Reactive arthritis	30 (20–40)	21,000 (6,400–36,000)	0	20	5	17,100	61,050

(a) The proportion of cases due to foodborne transmission varies with different illnesses.
(b) The numbers in brackets indicate the credible interval—see text for details.
(c) Includes lost work days and household activity days.
(d) New and old cases of irritable bowel syndrome.
Sources: See Appendix A.

prev page| TOC |next page