Appendix H: Case studies

Whilst a comprehensive study to calculate the total cost of food-borne illness in Australia is beyond the scope and the resourcing of this Review, the Review Committee has examined two case studies of illnesses caused by food in Australia, in order to provide an indication of the potential cost arising from an outbreak. The two case studies selected involve a Salmonella outbreak in Victoria and an E. coli outbreak in South Australia. Although the cause of these outbreaks was not from imported food, these foods are tested for pathogens that cause both these diseases.

As already mentioned in Section 5.2, the estimates derived here only relate to the medical/productivity cost of acute illness. The costs of any long term complications of these outbreaks cannot, at present, be estimated because of the short space of time since the incidents occurred. It is recognised that long term costs are likely. Other costs identified in Section 5.2.2 have been largely ignored due to difficulties in measurement. However, program management should attempt to derive a better information base from which to make necessary policy decisions. For these reasons, the estimates made in this appendix must be viewed as indicative of the minimum cost impact of these outbreaks.

H.1 Case study: Salmonella outbreak - Victoria 1997

Much of the following discussion is based on information provided by the Victorian Department of Human Services and on Lester et al. (1997).

On 23 March 1997 the Department of Human Services in Victoria was notified of large numbers of patients seeking treatment for gastroenteritis at two hospitals in the south eastern suburbs of Melbourne. The outbreak was identified to have been caused by the presence of Salmonella in some of the ingredients used in the production of pork rolls. The hot bread shop identified as the source was closed on the day in which the outbreak was notified and remains closed to date. Most of the suspect rolls were purchased from that shop, although a number of other retail establishments which sold rolls supplied by the hot bread shop had cases associated with them. The number of rolls consumed ranged from one bite to four rolls.

In total 862 persons were identified as being affected by the outbreak. Medical attendance was recorded for 859 cases, with 854 recorded as having attended a general practitioner or hospital Emergency Department at some time during their illness. Eighty cases required hospitalisation. No deaths were reported in relation to this outbreak. The age of the victims ranged from less than one year old to 85 years old. The median age was 29 years. Males and females were equally affected.

Of the 265 cases for whom the number of visits to a doctor was recorded, the number of visits ranged from 0 to 7 visits. The mean was 3 visits. The duration of illness was recorded in 274 cases and ranged from 1 to 21 days. The mean was 7.2 days.

Although data for the length of hospital stay for this outbreak are not available, data from the Victorian Department of Human Services show that in 1996-97 the average length of stay in hospital for a case of Salmonella was 4.5 days. The length of stay ranged from one day to 64 days. In 93 percent of the cases patients were released from hospital within the first 10 days, with 26 percent of the patients staying in hospital for only one day.

Table H.1 summarises the costs of the outbreak and provides an explanation for some of the calculations. A number of victims have since developed symptoms linked to reactive arthritis, a chronic condition that is associated with Salmonella infection.

Table H.1 Costing of Salmonella outbreak: Victoria 1997

Number

 No

GP visits or days in hospital

Unit cost

Cost ($)

Medical expenses

Number of cases
Medical attendance
Hospitalisation
Medication
Laboratory
Sub-total

Productivity losses

Number of cases
Hospitalised
Non hospitalised
Sub-total

Total cost

862
782
80
862
407




862
80
782 

 

3 visits
4 days





Number of days


11 days off work
5 days off work

 

 

$25 /visit
$618/day
$27/prescription
$50/test






$159
$159


 

 

58 650
197 760
23 274
20 350
300 034





139 920
621 690
761 610

1 061 644


Assumptions and methodology

A daily hospitalisation cost of $618 is used based on a national public hospital average cost for all conditions (Commonwealth Department of Health and Family Services 1998). This includes accommodation, medical care and medication whilst in hospital and any pathology tests done.

It is assumed that all 862 patients took medication once at a cost of $27 per prescription.

Laboratory tests are assumed to cost $50 per test.

For the purposes of estimating productivity losses:

  • For patients who were hospitalised, the length of illness was assumed to be 3 times the mean hospital stay. This estimate was adjusted for weekends (13 days less 2 days).
  • For patients who were not hospitalised, the mean length of illness was used. This estimate was adjusted for weekends (7 days less 2 days).

To calculate productivity losses, the average adult full time weekly earnings was used and divided by 5 to obtain a daily rate.

H.2 Case study: E. Coli outbreak South Australia 1995

The material contained in the section draws heavily from the Coroners Inquest concerning the death of Nikki Robinson (Chivell 1995), and from Henning et al. (1997).

In January 1995 an outbreak of illness caused by E. coli O111:H was linked to the consumption of sausage mettwurst produced by a smallgoods manufacturer in South Australia. The outbreak was declared after three children developed Haemorrhagic Uraemic Syndrome (HUS), and were reported to the Communicable Diseases Unit of South Australia. According to the Coroners Report on the death of one of the affected children, the outbreak involved 23 cases of paediatric HUS, 4 cases of thrombotic thrombocytopaenic pupura and reports of some 200 cases of haemorrhagic colitis and diarrhoea. One child that developed HUS died, and five of the other children continued to suffer impaired renal function one year after infection. Nine children suffered major non-renal complications. These included colonic necrosis, cerebral haemorrhage/infarction, convulsions and glucose intolerance.

Although E. coli O111:H has been associated with HUS before, this was the first large outbreak reported in Australia.

Nineteen children had a prodromal illness characterised by abdominal pain and bloody diarrhoea. The median duration was of four days. All children had evidence of haematological and renal disease on admission to hospital. Most of the children required substantial supportive treatment, renal dialysis in 18 cases and repeated blood transfusions in all cases. The children were hospitalised for an average of 20 days and underwent dialysis for an average of 14 days.

Two years after the outbreak, the outcome is excellent for 17 children without renal dysfunction. However, five children have been left with impaired kidney function and they must be considered to be at risk for a progressive deterioration in the renal function in the long term.

In addition to the 23 HUS cases, there were a number of other children and adults who were affected by the outbreak and suffered from haemorrhagic diarrhoea. The Review Committee could not obtain official data on these cases although it appears that this figure was based on the number of telephone calls received by the South Australian Health Commission from people reporting symptoms that can be caused by E. coli. Nevertheless, from the Coroners Report it can be established that seven people were hospitalised as a result of this outbreak with at least two of them suffering renal failure. In the absence of more accurate data the Review Committee has assumed that of the suspected 150 cases of E. coli, only half developed symptoms severe enough to warrant medical attention and use of sick leave.

Table H.2 provides a costing of the acute phase of this outbreak in relation to the paediatric HUS and the other 150 cases.

Table H.2 Costing of E. coli outbreak: South Australia 1995

Number GP visits or days in hospital Unit cost Cost ($) Total

Medical expenses
 

Paediatric HUS cases
Number of cases
Medical attendance
Hospitalisation
Dialysis
Medication
Laboratory
Sub-total

Other cases

Number of cases
Medical attendance
Hospitalisation
Dialysis
Medication
Laboratory
Sub-total

Productivity losses

Paediatric HUS cases
Other cases

Sub-total
Total cost

23
23
23
18
23
23



200
100
6
2
100
100




23
75
 20 visits
20 days
14 days
5
2




2 visits
19 days
19 days
1 prescription
1 test


Number of days

44
4
 $25 /visit
$1 000/day
$200/day
$27/prescription
$50/test


$25 /visit
$1 000/day
$200/day
$27/prescription
$50/test





$159
$159
 11 500
460 000
50 400
3 105
1 150
526 155



5 000
114 000
7 600
2 700
5 000
134 300



160 908
47 700

224 508 884 963

Note: Factors such as ongoing and long term health problems, ultimately give a minimum cost in excess of $1.17 million.

Assumptions and methodology

A daily hospitalisation cost of $1000 is used. The higher rate used here, compared to the Salmonella case, reflects the severity of the illness and the fact that intensive care had to be provided to most of the patients.

Laboratory tests are assumed to cost $50 per test.

It was assumed that there were on average two visits to general practitioners prior to hospitalisation and that visits continued on a monthly and bi-monthly basis, 12 and 24 months after discharge, respectively.

Of the 200 or so cases reported to have been affected by the outbreak, it was assumed that only half developed symptoms which required medical attention and resulted in days of work lost.

For the purposes of estimating productivity losses, the length of illness was estimated to be 3 times the mean hospital stay. This estimate was adjusted for weekends (60 days less 16 days).

To calculate productivity losses, the average adult full time weekly earnings was used and divided by 5 to obtain a daily rate.

Besides the tragic effect of the outbreak on the victims and their families, the identification of the companys product and its linkage with the death and severe illness of the children involved had a catastrophic effect upon the companys business, such that it ceased operations on Monday 6 February 1995. This resulted in the downfall of one of the biggest smallgoods manufacturers in South Australia and the loss of more than 100 jobs. The outbreak also had a deleterious effect upon several other producers of smallgoods in South Australia. According to trade data in the aftermath of this outbreak, the entire smallgoods sector suffered a major drop in turnover, with sales still considerably lower than pre-outbreak levels a year after the event.

Last reviewed:
23 Apr 2007