Year 2 of a UK-wide survey of campylobacter contamination on fresh chickens at retail (July 2015 to March 2016)

Last updated:
14 March 2017
This survey investigated the prevalence and levels of campylobacter contamination on fresh whole chilled chickens and their packaging. The survey examined more than 3,000 samples of whole chickens bought from UK retail outlets and smaller independent stores and butchers. The sets of results from quarter 1, 2 and 3 (sampling period July to March 2016) are available.

Third quarter results (published May 2016)

The results for the third quarter of testing, from January to March 2016, continue to show a decrease both in the number of birds with campylobacter on them and those with the highest level of contamination from the equivalent quarter last year.

The latest data show 9.3% of chickens tested positive for the highest level of contamination in this quarter, down from 21.8% for the three months from December 2014 to February 2015*.

Campylobacter was present on 50% of chicken samples, down from 71% in the equivalent quarter of the previous year. We tested 1,009 samples of fresh whole chilled UK-produced chickens and packaging this quarter.

One of the reasons the survey results are lower this quarter is because of the decision taken by a number of retailers and their suppliers to remove neck skin from the bird before it goes on sale. The neck skin is the most contaminated part of the chicken. However, it is also the part of the bird that we have been testing in our survey and this means that comparisons with previous results are not as reliable as we would like.

Therefore, this quarter, we are giving an overall figure for the amount of campylobacter on chicken and not breaking the figures down by retailer as we normally do. We have also stopped this survey and will begin a new one in the summer, with a different method of testing campylobacter levels on chicken. First results from this survey, which will rank retailers, are due in January 2017.

*No data were available for March 2015.

Second quarter results (published February 2016)

The results for the second quarter of testing, from October to December 2015, continue to show a decrease in the number of birds with the highest level of contamination from the same months last year. These most heavily contaminated birds, carrying more than 1,000 colony forming units per gram (cfu/g), are the focus of the current target agreed by industry, which is equivalent to no more than 7% of chickens at retail having the highest levels of contamination. Research has shown that reducing the proportion of birds in this category will have the biggest positive impact on public health.

The latest data show 11% of chickens tested positive for the highest level of contamination, down from 19% in October to December 2014. Campylobacter was present on 59% of chicken samples, down from 74% in the same months of the previous year.

In this second quarter of the FSA’s second survey, 966 samples of fresh whole chilled UK-produced chickens and packaging have been tested. The chickens were bought from large UK retail outlets and smaller independent stores and butchers. The survey commenced sampling in July 2015.

The data continue to show an improvement from the previous year. Interventions, including improved biosecurity, SonoSteam, and the trimming of neck skins, introduced by some retailers to reduce levels of campylobacter, may be helping to deliver the improved results. The trimming of the neck skin, the most highly contaminated skin area, means chickens are carrying less campylobacter. The results of this intervention, while making chickens safer, mean comparisons to the first year’s survey may potentially be more difficult in future quarters as most samples from the previous year will have analysed more neck skin. The FSA will review the impact of this successful intervention to ensure the survey results remain robust.

All results below are taken from the Official Statistics report for the survey which gives a full explanation of the results and background to the methodology.

The FSA advises that the data for individual retailers have to be interpreted carefully. Confidence intervals are given in brackets for each retailer and the 'others' category. These show the likely range of the results allowing for the number of samples taken. The 95% confidence intervals means that we would expect the true prevalence to fall within the lower and upper confidence limits 95% of the time.

Retailer

No. of
samples

% skin samples positive for Campylobacter

% skin samples
over 1000 cfu/g Campylobacter

% packaging samples

positive for Campylobacter

Aldi

96

62.5

(52.0 - 72.2)

7.3

(3.0 - 14.4)

4.2

(1.1 - 10.3)

Asda

97

60.8

(50.4 - 70.6)

11.3

(5.8 - 19.4)

6.2

(2.3 - 13.0)

Co-op

91

53.8

(43.1 - 64.4)

9.9

(4.6 - 17.9)

8.8

(3.9 - 16.6)

Lidl

94

64.9

(54.4 - 74.5)

8.5

(3.7 - 16.1)

9.6

(4.5 - 17.4)

M&S

99

54.5

(44.2 - 64.6)

15.2

(8.7 - 23.8)

3.0

(0.6 - 8.6)

Morrisons

92

45.7

(35.2 - 56.4)

5.4

(1.8 - 12.2)

3.3

(0.7 - 9.2)

Sainsbury's

102

65.7

(55.6 - 74.8)

17.6

(10.8 - 26.4)

5.9

(2.2 - 12.4)

Tesco

99

61.6

(51.3 - 71.2)

10.1

(5.0 - 17.8)

6.1

(2.3 - 12.7)

Waitrose

95

60.0

(49.4 - 69.9)

9.5

(4.4 - 17.2)

6.3

(2.4 - 13.2)

Others

101

56.3

(46.4 - 66.1)

11.4

(5.5 - 18.1)

5.5

(1.2 - 10.6)

All

966

58.9

(55.4 - 62.5)

10.7

(8.6 - 13.1)

5.7

(4.1 - 7.4)

**The ‘Others’ category includes supermarkets where the market share was deemed small using 2015 Kantar data: e.g. Iceland, convenience stores, independents, butchers etc.

First quarter results (published November 2015)

The results for the first quarter of testing, from July to September 2015, show a decrease in the number of birds with the highest level of contamination from the same months last year. These most heavily contaminated birds are the focus of the current target agreed by industry, which is equivalent to no more than 7% of chickens at retail having the highest levels of contamination. Research has shown that reducing the proportion of birds in this category will have the biggest positive impact on public health.

The new data shows 15% of chickens tested positive for the highest level of contamination, down from 22% in July to September 2014. Campylobacter was present on 76% of chicken samples, down from 83% in the same months of last year.

The results for the first quarter show:

  • 15% of chickens tested positive for campylobacter within the highest band of contamination*
  • 76% of chickens tested positive for the presence of campylobacter
  • 0.3% of packaging tested positive at the highest band of contamination
  • 6% of packaging tested positive for the presence of campylobacter

*More than 1,000 colony forming units per gram (cfu/g). These units indicate the degree of contamination on each sample.

In this first quarter, 1,032 samples of fresh whole chilled UK-produced chickens and packaging have been tested. The chickens were bought from large UK retail outlets and smaller independent stores and butchers. The new survey commenced sampling in July 2015.

The FSA has been testing chickens for campylobacter since February 2014 and publishing the results as part of its campaign to bring together the whole food chain to tackle the problem. Campylobacter is the most common cause of food poisoning in the UK, making an estimated 280,000 people ill every year.

As with the previous survey, the data shows variations between the retailers. Testing of chickens from Co-op and Waitrose show both retailers have made the most significant reductions in the proportion of the chickens they sell that are most highly-contaminated.

All results below are taken from the Official Statistics report for the survey which can be found at the link below. This report gives a full explanation of the results and background to the methodology.

The FSA advises that the data for individual retailers have to be interpreted carefully. Confidence intervals are given for each retailer and the 'others' category. These show the likely range of the results allowing for the number of samples taken. The 95% confidence intervals means that we would expect the true prevalence to fall within the lower and upper confidence limits 95% of the time.

Retailer

Number of
samples

% skin samples positive for campylobacter

% skin samples
over 1000 cfu/g campylobacter

% packaging samples positive for campylobacter

Aldi

99

76.8 (67.2 - 84.7)

9.1 (4.2 - 16.6)

3.0 (0.6 - 8.6)

Asda

102

76.5 (67.3 - 84.5)

23.5 (15.7 - 33.0)

8.8 (4.1 - 16.1)

Co-op

106

72.6 (62.8 - 80.7)

4.7 (1.5 - 10.7)

4.7 (1.5 - 10.7)

Lidl

101

65.3 (56.0 - 75.1)

11.9 (6.3 - 19.8)

7.9 (3.5 - 15.0)

M&S

104

82.7 (74.0 - 89.4)

18.3 (11.4 - 27.1)

1.0 (0.0 - 5.2)

Morrisons

109

86.2 (78.3 - 92.1)

25.7 (17.8 - 34.9)

14.7 (8.6 - 22.7)

Sainsbury's

108

80.4 (71.8 - 87.5)

17.8 (11.0 - 26.3)

5.6 (2.1 - 11.7)

Tesco

110

76.4 (67.3 - 83.9)

10.0 (5.1 - 17.2)

1.8 (0.2 - 6.4)

Waitrose

98

59.2 (48.6 - 68.5)

4.1 (1.1 - 10.1)

9.2 (4.3 - 16.7)

Others**

95

71.6 (62.0 - 80.8)

18.0 (10.5 - 26.3)

9.6 (4.2 - 15.9)

Total

1,032

76.3 (73.3 - 79.2)

14.9 (12.5 - 17.4)

6.4 (4.9 - 8.0)

**The ‘Others’ category includes supermarkets where the market share was deemed small using the 2010 Kantar data: eg Iceland, plus convenience stores, independents, butchers etc.

Read the 19 November 2015 news story

More in this section