Food Surveillance News  - Winter edition

Review of recent food survey reports conducted by ACT Health

Australian Quarantine and Inspection Service updates its border testing of imported seafood

Keeping an eye on food recalls

NZFSA Food Recall advertising survey

Implementation Sub Committee Coordinated Food Survey Planning Workshop in Brisbane

Snapshot survey on the microbiological quality of kebabs in New South Wales

New Zealand Food Safety Authority management of tutu toxin in honey

Research on Key Food Label Elements

 

Review of recent food survey reports conducted by ACT Health

An ACT Health Protection Service (HPS) review of five recent food survey reports shows food safety generally continues to improve.

The HPS has been conducting microbiological analysis of food and water samples for approximately 30 years as part of its annual food survey program.   HPS develops the program by considering proposals put forward by OzfoodNet and Communicable Disease Control and, where possible, by collaborating on surveys that are part of the ISC co-ordinated survey plan.

The HPS annual food program consists of a year round Ready to Eat (RTE) survey that is complemented by a number of short-term surveys.   The RTE survey has been running since 2000 and involves an analysis of RTE food samples collected from retail food establishments within the ACT.  

These samples are tested for compliance with the FSANZ Ready to Eat guidelines 2001.   The supplementary short-term food surveys operate to help the HPS maintain surveillance of the quality of food provided to the ACT community.

All the surveys target food services that may present a risk to the community so the HPS can develop corrective processes, if necessary, before food related risks develop. The surveys are an active food risk management activity and, where possible, are collaborative as part of the ISC co-ordinated survey plan, for which FSANZ is the sponsor.

The five surveys reviewed by the HPS are:

• Microbiological Quality of Ready to Eat Foods (RTE) 2006 – 7 (including a comparison with previous four years)
• Quality of Cooked Rice
• Retail Packaged Salads
• Incidence of Salmonella and Listeria Monocytogenes in Cooked Chicken
• Microbiological Quality of Sushi (including comparison with previous survey)

The full details of all published surveys can be found on the ACT Health web site at: http://www.health.act.gov.au/c/health?a=da&did=10054021

Conclusions derived from these surveys

The RTE survey tests the compliance for Standard Plate Count (SPC),E. coli, Coagulase positive Staphylococci,Bacillus cereus, Salmonella sp.andListeria monocytogeneswith FSANZ guidelines for microbiological quality. The 2006-7 survey report indicates that 2006-7 was the best year to date, with all tests forBacillus cereus, Salmonella spandE. colibeing in the satisfactory category.   Since the inception of this survey, a comparison of the previous four years’ test results indicates a continuing improvement in microbiological quality of the RTE foods in the ACT. The comparative findings can be found on the ACT Health website.

The Quality of Cooked Rice survey also tested rice for compliance to SPC (level 1),E. coli, Coagulase positive Staphylococci,Salmonella sp., Bacillus cereus and Listeria monocytogenes with FSANZ guidelines for microbiological quality. The overall results were very good with no Salmonella sp., Bacillus cereus or Listeria monocytogenes detected in any of the samples. Two samples were marginal for Coagulase positive Staphylococci and three were marginal for E. coli. The satisfactory standard for SPC (level 1) was met nearly 93% of the time, with two marginal cases, and three unsatisfactory cases.

HPS conducted the Retail Packaged Salads survey following a sudden rise in the number of ACTSalmonella sp.cases in early 2006. Samples of pre-packaged salads were bought from retail outlets and divided into four sub-samples. No Salmonella sp.was detected in any of the sub samples. A quarter of the sub samples were tested forE. coliand all were negative.

The incidence of Salmonella sp.and Listeria monocytogenesin Cooked Chicken survey was prompted by a quantitative risk assessment article in the Journal of Food Microbiology. The article indicated the possibility of a ‘significant number’ of chicken pieces being contaminated, or cross-contaminated, with Salmonella sp. during handling. However,Salmonella sp. was not isolated from any of the chicken pieces. Soon after the survey started, HPS decided to add Listeria monocytogenesto the testing regime.  Subsequently,Listeria monocytogenes was isolated from two ‘hot’ samples and one cold shredded sample.

The Microbiological Quality of Sushi survey was conducted in conjunction with the ISC coordinated survey on Sushi, led by NSW, and allows comparisons with the previous ACT Sushi survey held in 2003.   HPS detected no potentially hazardous samples in the tests which found:

• the number of unsatisfactory samples reduced by 66%
• the number of marginal samples reduced by 20%
• the percentage of satisfactory results for all tests was more than 90% and the majority of tests more than 95%.

The results of the survey indicate that the microbiological quality of sushi in the ACT is considerably better than it was four years ago.

Australian Quarantine and Inspection Service updates its border testing of imported seafood

Background

The Australian Quarantine and Inspection Service (AQIS) tests imported seafood for the presence of antimicrobial chemicals at the rate of 5% of consignments to monitor compliance with the Australia New Zealand Food Standards Code (the Code). This testing program is in addition to testing for food classified as “risk” by Food Standards Australia New Zealand (FSANZ). The rates of inspection of risk food and for compliance monitoring are prescribed by the Imported Food Control Act 1992.

AQIS has tested imported prawns for nitrofurans and chloramphenicol since 2003. In 2005, AQIS introduced testing for malachite green as part of the routine sampling of imported aquaculture fish. In the 2006 calendar year, imported seafood tested for these antimicrobial compounds showed above 95% compliance with the Code.

Testing imported seafood for the above chemicals was introduced following information from domestic findings, including a number of Australian surveys that detected antimicrobial chemicals, such as:                

• A 2005 FSANZ-led national survey of chemical residues in aquaculture fish which tested for a range of veterinary residues such as nitrofurans, chloramphenicol, sulphonamides, tetracyclines, penicillins, macrolides, and quinolones, and also for polychlorinated biphenyls and a number of heavy metals. This survey found that residues of these veterinary chemicals in aquaculture fish were generally compliant with the Code except for residues of malachite green and/or its metabolite leuco-malachite green in some domestically produced finfish and imported seafood.
  
• A 2005 survey in South Australia which detected low levels of chloramphenicol in some imported crab meat.

In 2006, there were concerns that other chemicals may be present in seafood, particularly in farmed seafood, and testing would need to be broadened to cover other chemicals. Of particular concern was the possible occurrence of veterinary compounds that are of critical importance in human medicine.

To investigate whether new chemicals needed to be added to the testing program, AQIS conducted a snapshot survey of antimicrobial and pesticide chemicals in imported seafood.

AQIS has reviewed its antimicrobial screen applied to imported seafood.

In September 2007, AQIS decided to add, for a trial period of six months, three additional antimicrobial chemicals (fluoroquinolones, quinolones and penicillin) to its existing screen for imported seafood, which tested for nitrofurans and malachite green. This testing occurs under the random surveillance program in which it randomly tests 5% of imported seafood.

AQIS introduced the extra three chemicals following its small survey of chemical residues in imported seafoods, which it carried out between 2006 and 2007 (see Food Surveillance News Spring 2007).

The purpose of the 2006/07 survey was to provide a snapshot of chemical residues in imported seafood to check if the testing program was up-to-date. It found one or more antimicrobial chemicals in some imported seafood, prompting AQIS to test for the additional antimicrobials over the next six months to gather more information.  

After the six months additional testing on imported seafoods, AQIS reviewed the test data. In those six months, AQIS has conducted 522 tests, with seven detections of an antimicrobial chemical (Table 1).

The seafood tested included fish (barramundi, basa, cod, eel, mackerel, marlin, red emperor, salmon, trout, tuna, whiting and other species), lobsters, crabs and prawns. These foods were sourced from Argentina, Chile, China, Denmark, India, Indonesia, Italy, Japan, Malaysia, Myanmar, Namibia, Netherlands, New Zealand, Niue, Norway, Papua New Guinea, Portugal, South Africa, South Korea, Sweden, Taiwan, Thailand, United States and Vietnam.

Table 1: Summary of failed tests

 

 

 

The Result

After six months of testing under the broadened antimicrobial screen, there had been no detections for penicillin or quinolones. As a result, AQIS removed these chemicals from the antimicrobial screen. It will continue testing imported seafood products for fluoroquinolones, nitrofurans and malachite green.

Keeping an eye on food recalls

Government agencies in Australia and New Zealand constantly monitor the food supply to ensure that it is safe, and that foods comply with standards including for microbiological contaminants, pesticide and veterinary medicine residue limits and chemical contamination.

Food identified as a risk to public health and safety is recalled. FSANZ is the coordinating agency for all food recalls in Australia and the New Zealand Food Safety Authority (NZFSA) is the responsible authority in New Zealand.

Over the 2008 autumn season the number of recalls in New Zealand has been less than average, while in Australia there have been more than usual.

Australian recalls over the months of April to June included:

• Woolworths recalled all its lamb products sold in supermarkets, except in Queensland where the only products affected were: Lamb and Cheese Rissoles, Lamb Mini Roast and Economy Burgers. The company said the reason for its consumer level recall was as a precaution only and followed the detection of a strange odour coming from a number of lamb products processed by Gregory M (GM) Scott Pty Ltd in Cootamundra. Lamb products with a use by date between 31/05/08 to 23/06/08 and 07/06/08 to 14/06/08 for QLD only may have been exposed to these vapours. There were no reports of consumer illness or injury.
  
• Unibic Australia Pty Ltd ordered a voluntary consumer level recall of its Unibic Anzac Biscuits – 375g Value packs due to the possibility the packs (with a best before date of 26/03/09) may have been contaminated with metal pieces which may pose a health risk if swallowed.
  
• Australian International Foods Pty Ltd advised customers of a voluntary consumer level recall on a number of its foods because it feared they may have been contaminated with metal wire which may have posed a health risk if swallowed.  The products were:
  
  
  • Darshan   Savoury Beef Rolls, 300 gms
  • Darshan Savoury Chicken Rolls, 300 gms
  • Darshan Savoury Vegetable Rolls, 300 gms
  • Darshan Assorted Mini Rolls, 450 gms
  • Darshan Entertainment Pack, 400 gms

All these foods had best before dates of 11/02/09 to 14/04/09.

• Nutricia Australia Pty Ltd voluntarily recalled 110g glass jars of baby food named Golden Circle Little Extras From 6+ months Orchard Fruit Gels as a precautionary health and safety measure, following a report of an unverified foreign object. The recall applied only to the product with the nominated size and best before dates 12/9/09 and 13/9/09 and bought from Woolworths or Safeway.
  
  
• Patak’s Foods Ltd (UK) through its Australian distributor (General Mills Australia Pty Ltd) recalled all 540g jars of Rogan Josh Simmer Sauce with a Best Before date of 20/05/2009. The sauces potentially contained glass fragments.
  
  
• Safcol Australia Pty Ltd voluntarily recalled its SAFCOL 95g Premium Selection Tuna Range with Gold Foil Top Packaging and SAFCOL 95g Sandwich Delite Tuna Range with Gold Foil Top Packaging. The company feared the foil lids might come loose causing the contents to spoil.   The recall applied to all products in the specified ranges except the Safcol Premium Selection Tuna Extra Virgin Olive Oil, which is sealed with a metal ring pull top.
  
  
• Alba Cheese Manufacturing Pty Ltd recalled its Alba Cheese Shredded Mozzarella, 3kg, 1kg and 500gm plastic bags with recall dates of 25/04/08 due to indications of Listeria monocytogenes contamination. No other Alba Cheese Manufacturing Pty Ltd foods were affected by this recall.
  
• Coles supermarkets voluntarily recalled two flavoured lamb mini roast products sold in Coles, BI-LO and Pick ‘n Pay stores throughout Australia. The affected products were:You’ll Love Coles branded lamb mini roast with savoury mint flavouring with Best Before dates up to and including 11th April 2008, and You’ll Love Coles branded lamb mini roast with herb and mustard flavouring with Best Before dates up to and including 11th April 2008. No other flavoured lamb products were affected. Coles recalled the foods due to the non-labelling of an allergen (soy).
  
• Ridders Fresh Smallgoods voluntarily recalled Farmwood Delicatessen Smoked Chicken Breast 500g sold through ALDI Stores in New South Wales, Victoria and Australian Capital Territory after tests revealed the possible presence of Listeria monocytogenes. The plastic wrapped chicken had a Use By date of 10/04/08.
  
  
• P & M Quality Smallgoods Pty Ltd recalled Courtway Double Smoked Ham in 300g plastic containers with a laminate top sold through ALDI Stores in NSW, Queensland, Victoria and the ACT, due to indications of Listeria monocytogenes contamination. The containers had a Use By date of 04/04/08.
  
  
• Gaybill’s Goat Dairy voluntarily recalled its Unpasteurised Frozen Goat Milk in 1 litre and 2 litres plastic bottles with a Use By date of 28/05/08. Gaybill’s recalled the milk in response to tests which indicated the presence of Salmonella Zanzibar. Symptoms of Salmonella poisoning are vomiting, diarrhoea and abdominal pain which persists for more than 24 hours.
   
  
• Goodman Fielder Limited voluntarily recalled its Praise Seafood Sauce in 370mL plastic squeezy bottles with a Best Before date of 15/11/08 because the bottles contained an amount of fish, malt and barley which was not clearly identified on some of their labels.

There were just five recalls in New Zealand over the same period:

• Quality Supply Ltd recalled its Bacchus Energy Drink, regular and sugar free varieties in 250ml cans, because they lacked a mandatory advisory statement. The drinks contained a small amount of royal jelly which has been reported to cause severe allergic reactions and in rare cases fatalities, especially in asthma and allergy sufferers. The drinks have a Best Before date of 28/12/09.
  
  
• The New Zealand Food Safety Authority (NZFSA) ordered a recall of Chinese herbal teas, warning people not to consume in any way seeds that may be found in such teas made from the plant Abrus Cantoniensis. The teas were sold in retail packs under various brand names such as Heng Ming through Asian supermarkets and also sold by Eastern medicine outlets. The teas may also be sold under the names Canton Love Pea or Herba Abri. NZFSA advised anyone with the product not to consume it and to return it to the place of purchase. This plant product is not permitted to be sold as a food or food ingredient under the Australia New Zealand Food Standards Code. It is believed that the seeds and possibly the seed pods are toxic.
  
  
• Comb honey contaminated with toxins from the native tutu plant. People who should be alerted to this recall are those that have purchased honey from the Coromandel Peninsula from early January 2008. Three outlets in Whangamata sold the honey: Quarry Orchard, Valley Orchards and Meat at the Beach. The number of suspected cases is now 22. The NZFSA is advising anyone buying honey to check with their supplier that the harvested comb honey meets all food safety requirements. The NZFSA is also advising people that if they feel they might have suffered an illness associated with honey, to contact the Waikato District Health Board’s helpline on 0800 276 216 for information and advice. The two brands of comb honey produced by Projen Apiaries are Wentworth Valley and Moana Point, as pictured below. There are a number of containers of this honey still unaccounted for and NZFSA is keen to locate them. The authority would like all consumers who may have already eaten these products or disposed of them to contact the 0800 helpline with details.
  

• Heller Tasty Ltd is recalling its lamb and mint sausages after small pieces of soft blue plastic were found in them. There have been no reports of illness; however any person concerned about their health should seek medical advice. The affected sausages have been sold through New World, Pak ‘n’ Save and Fresh Choice supermarkets in the South Island between 26 February and 6 March 2008.
  
  
• Ceres Enterprises Ltd, Venerdi Ltd, Organic Bakeworks Ltd and Paraoa Bakehouse are recalling a number of foods which may contain broken glass particles. These foods are:
  
  
  • Ceres Organics - Organic Sunflower Seeds 300g Best Before 22/1/09 to 20/2/09.
    
  • Venerdi - Spelt Seven Seed Sourdough Best Before 23/2/08 to 26/2/08, Full Flavour Six Seed G/F Best before 10/2/08 to 26/2/08.
    
  • Organic Bakeworks - Liberte Sunflower & Linseed G/F Bread 540g Best Before 4/2/08 to 25/2/08, Liberte Sour Seed G/F Bread 560g Best Before 4/2/08 to 25/2/08, Organic Carrot Sultana Cookies Best Before 28/2/08 to 10/3/08, Organic Coconut Cookies Best Before 28/2/08 to 10/3/08 or Gluten Free Bread Crumbs 300g Best Before 1/11/08 to 25/11/08.
    
  • Purebread - Crunchy Granola 600g Best Before 25/7/08 to 20/8/08.

Also, anyone buying any item containing sunflower seeds from any of the health food stores listed within the past month should check with the store to ensure that the product bought is not affected.

NZFSA Food Recall advertising survey

The New Zealand Food Safety Authority (NZFSA) commissioned a pilot survey to find out how efficient newspaper advertisements are in communicating information about food recalls to the public. The survey also explored other avenues that might be more efficient at reaching people.

Typically, food recalls are advertised in newspapers in a standard format. While the NZFSA could take other steps to inform the public (such as website notifications, email alerts and point of sale notices), it relies heavily on newspaper advertisements. However, newspaper advertising is expensive, one-off in nature, and may not deliver the confidence the authority seeks in managing the risk.

For the survey, the NZFSA contracted UMR Research Limited to phone 750 New Zealanders aged 18 years and over living in private households. UMR used a standard phone survey questionnaire. The survey was conducted in May 2007, 10 days after a national consumer level recall was made by a well-known brand of breakfast cereals.

The results of the survey indicated that:

• The current communication strategy for notifying the public of food recalls results in, at best, a third of consumers becoming aware of food recalls.
• Currently, newspaper advertisements by themselves appear effective in ensuring 20% of the population becomes aware of recalls.
•  Television was seen by 74% of respondents as the most effective means of notifying recalls.  
•  56% of respondents still believe that newspaper advertising is effective.
• 61% of respondents see radio as an effective notification tool if the case is urgent.
• Press releases do help to get the message across.
• Notices in shops seemed to score relatively well even though they are not a current requirement in New Zealand, although one supermarket chain uses them as policy (scored effective 36%).

With this knowledge from the survey, the NZFSA proposes to:

1. Continue with newspaper advertising as the minimum advertising media for food recalls.
2. Undertake a cost benefit analysis for the different options for notifying the public , including consideration of the cost differences between the different forms of advertising and the timeframes to get advertising underway in the different media options. Such a cost benefit analysis may be useful for considering a requirement to use radio advertising in urgent situations, or where risk is severe. For example, such an analysis will investigate how practical the different media options will be, how quickly ‘advertising’ on radio can be set up compared with a press release, and will the cost be prohibitive to small/medium businesses?
3. Investigate whether the food recall awareness of young, old, pregnant and immune compromised consumers matches those of the rest of the population.
4. Investigate making a press release a requirement (that carries no extra cost to the company) as there appears to be a better coverage across consumers than newspaper advertising coverage alone.
5. Promote use of shop or point of sale notices.
6. Promote the email alert system in place for recalls from the NZFSA website to consumers (at that time there were 270 subscribers to recall alerts, 8% of total subscribers to our website).

The survey provides some good baseline information. If additional requirements are placed on recalls (e.g. press release and/or shop notices), the NZFSA may consider running a similar survey in the future to see if there is greater recall awareness indicating any such extra measures are effective.

Implementation Sub Committee Coordinated Food Survey Planning Workshop in Brisbane

The first biannual Implementation Sub Committee (ISC) Coordinated Food Survey Planning Workshop for 2008 was held on 3 April in Brisbane.  

The aim of these workshops is to promote a consistent and coordinated approach to surveillance and monitoring activities across Australian jurisdictions and New Zealand.  

The main purpose of the Brisbane workshop was to complete the Coordinated Food Survey Plan (the Plan) for 2008-2011. The Plan is a three year forward plan for collaborative survey activities of national or bi-national significance. The workshop considered detailed proposals for new surveys to be included in the Plan for 2008-2011.

The workshop also discussed recent and emerging food related concerns in Australia and overseas. The issue of recycling of water by food businesses was flagged again for further consideration with a view to possible future survey work, as well as microbiological contamination of eggs and vitamin D levels in foods.

The workshop was co-chaired by Food Standards Australia New Zealand and the Queensland Department of Health and was well attended by representatives of food regulatory agencies in Australia and New Zealand.

Snapshot survey on the microbiological quality of kebabs in New South Wales

The NSW Food Authority (NSWFA) will continue to monitor kebabs for microbiological quality after a snapshot survey was conducted in early 2008.While no sample tested from the 25 stores contained potentially hazardous levels of bacteria and the majority of samples were within acceptable microbiological limits, the survey revealed a slight trend of more samples with borderline, though acceptable, results than in a previous survey.

In 2004, NSWFA did a survey of kebabs sold across New South Wales (Janssonet al., 2008). The survey methodology involved a combination of observations made during inspection and microbiological analyses. The results from the survey were assessed against guidelines developed by Food Standards Australia New Zealand (FSANZ, 2001) and less than 12% of samples were found to be unsatisfactory, with the majority failing due to slightly elevated levels of E. coli.  

The detection of E. coli in foods is not a direct indication that the food is unsafe but more an indication of potential problems involving the preparing and handling of foods. In all cases, follow-up work ensued to ensure any issues were rectified.

As a follow on from the 2004 survey, the recent 2008 survey investigated kebab businesses with the aim to provide a second snapshot on the microbiological quality of kebabs sold in NSW.  The survey showed a decrease in the level of unacceptable results compared with the results of the 2004 survey. Details on the survey methodology and results are as follows

Method and Material

20% of premises visited during the 2004 survey were revisited and kebabs purchased from each outlet. Where businesses had closed down since the 2004 survey, kebabs were purchased from another shop in the same area. Similarly, if the business had changed names, kebabs were purchased from the new business. In total, 48 kebabs from 25 premises across metropolitan Sydney where tested as part of the 2008 snapshot survey.

Samples were transported to the laboratory on ice and tested for:

• E. coli using AS 1766.2.4
• Coagulase positive staphylococci using AS 1766.2.4
• Clostridium perfringens using AS 1766.2.8
• Salmonella species using AS 1766.2.5.

Results and Discussion

Comparison of the microbiological results to the FSANZ microbiological guidelines revealed that 93.7% of samples were within acceptable microbiological limits (Table 1) categories. Where samples were considered microbiologically unacceptable (falling in the unsatisfactory category within the FSANZ guidelines), this was due to the levels ofE. coliexceeding 10² cfu per gram. No sample contained potentially hazardous levels of bacteria.

Table 1: Assessment of results using the FSANZ microbiological criteria for ready-to-eat foods (FSANZ, 2001).

	Microbiologically acceptable				Microbiologically unacceptable		Potentially hazardous
Organism	Satisfactory		Marginal		Unsatisfactory
	No.	%	No.	%	No.	%	No.	%
E. coli	29	60.4	16	33.3	3	6.2	0	0
Coagulase positive Staphylococci	47	97.4	1	2.1	0	0	0	0
C. perfringens	46	95.8	2	4.2	0	0	0	0

A summary of the results forE. coli, coagulase positive staphylococci andC. perfringensis presented in Table 2.Salmonella was not detected in any of the samples tested.

Table 2: Summary of microbiological results for whole doner kebabs

Test	Mean	Median	Minimum	Maximum
E. coli (log cfu/g)	0.99	0.48	0.48	5.0
Coagulase positive staphylococci (log cfu/g)	2	2	2	2.3
C. perfringens(log cfu/g)	2	2	2	2.3

A frequency distribution was calculated for E. coli results (Figure 1). From the figure it can be seen that the majority of samples contained low levels (< 10 cfu/g or < log 1 cfu/g) of E. coli, with levels greater that log 3 cfu/g (1000 cfu/g) only detected in 3 samples. Frequency distributions were not calculated for Coagulase positive staphylococci andC. perfringens as the vast majority of results were below the level of detection (100 cfu/g). A low level of Coagulase positive staphylococci was detected in one sample (100 cfu/g) and two samples contained low levels (100 and 200 cfu/g) of C. perfringens.

Figure 1: Frequency distribution ofE. colifor doner kebabs collected from retail outlets

Levels of E. coli exceeding 100 cfu per gram are unacceptable and might indicate:

• ·         Poor temperature during the display of raw or potentially hazardous foods (e.g. shredded lettuce, sauces) included in kebabs
• ·         Poor handling during the preparation of kebab components prior to display with or without poor temperature control
• ·         Poor personnel hygiene (Hocking, 2003).

The three unacceptable samples were sourced from two different businesses. One sample contained only slightly elevated levels ofE. coliand when resampled, results were within acceptable levels. The other business with two unacceptable results was inspected and further samples taken for testing with appropriate enforcement action taken.  

Caution must be taken when comparing these results to the previous survey because, for both surveys, knowledge of the number of kebabs produced in NSW, what is representative of a batch and the throughput of individual NSW businesses that enable a sound statistical design and comparison are not available. Taking this into account, when compared to the results obtained from the 2004 survey, the current snapshot showed a slight improvement in the number of samples falling within acceptable microbiological limits; 93.7% in 2008, 88.6% in 2004.  

In the snapshot survey, the presence of low levels of E. coli, Coagulase positive staphylococci or C. perfringens may be due to a variety of factors including:

• Low level contamination on vegetables used raw in kebabs (e.g. tomato, lettuce, onion)
• Incidental contamination during the handling and preparation of the kebab
• The presence of low levels of bacteria in ingredients such as herbs and species used in the sauce served on the kebab (e.g. hommus, garlic sauce).

The presence of bacteria due to the above factors is not unexpected. It would not normally constitute a safety risk, provided proper temperature control is maintained after preparing kebab ingredients and the final kebab is not kept for a long period (> 2 hours) at room temperature prior to eating.

Conclusion

The majority (93.7%) of kebabs surveyed were within acceptable microbiological limits. Where results were unacceptable, appropriate action was taken. The survey revealed a slight trend of more samples with borderline, although microbiologically acceptable, results than in the previous survey. This highlights the need for industry to remain rigorous in applying appropriate food handling controls.

The full details of the survey can be found on the NSWFA web site at: http://www.foodauthority.nsw.gov.au/_Documents/corporate_pdf/survey-report-kebabs-2008-final.pdf

References

Food Standards Australia New Zealand (2001).Guidelines for the microbiological examination of ready-to-eat foods. Retrieved March 25, 2008, from FSANZ Web site: http://www.foodstandards.gov.au/_srcfiles/Guidelines%20for%20Micro%20exam.pdf .

Hocking, A. (2003). Foodborne microorganisms of public health significance (6^th ed.). Marrickville: Southwood Press.

Jansson, E., Bird, P., Saputra, T. and Arnold, G. (2008).Food safety survey of retail doner kebabs in NSW. Food Australia. 60, (3), 95-98.

New Zealand Food Safety Authority management of tutu toxin in honey

The New Zealand Food Safety Authority (NZFSA) hopes to establish regulatory limits for tutin and its derivative in honey later in 2008.

The authority has been working towards implementing regulatory limits since tracing the cause of illness of 22 people over Easter 2008 to comb honey contaminated with tutin and its derivative hyenanchin.  Tutin has not been identified in any other sources, however its derivative, hyenanchin, has been detected at low levels in some of NZFSA’s residue monitoring programme samples. This presence of hyenanchin is not known to be associated with illness.  The presence of tutin in honey was identified as part of the NZFSA’s routine residue monitoring program.   There have been no further illnesses over the past three months from contaminated honey, indicating that the issue has been contained.  

Tutin, a picrotoxin, is a naturally occurring toxin found in honey.   The toxin is produced as a result of bees gathering honeydew secreted by bugs feeding on sap of the poisonous ‘tutu’ plant (Coriaria arborea). The toxin has no effect on bees and honeydew honey is chemically very similar to floral honey and cannot be distinguished by taste, sight or smell from other non-toxic honeys. The toxin cannot be degraded by any heating or processing of honey. The toxins are believed to be very stable, and poisoning cases have resulted from people eating honey that was several years old.

Both comb honey and extracted honey from the tutu plant can be poisonous. Comb honey poses a greater risk because it is eaten directly off the comb, increasing the chance of consuming honey with a high concentration of tutin. Extracted honey is often bulked or blended with other honey thereby reducing the concentration of toxin.

To produce honey that may result in adverse effects, all of the following conditions are required:

• Concentrations of numerous tutu bushes

• High numbers of vine hoppers

• Hot dry weather to allow the honeydew to build up on the tutu (rain can wash it off).

• An absence of more attractive food sources for bees, usually caused by drought.

• Presence of honey bees (Apis mellifera) being managed for honey production.

Tutu is widely distributed throughout New Zealand, particularly along stream banks and in regenerating native scrub.   It historically occurs primarily in the Coromandel Peninsula, Eastern Bay of Plenty (EBOP) and the Marlborough Sounds, although it is possible it could occur elsewhere. Test results from the honey incident in March 2008 confirmed that the suspected tutu toxins and its derivative hyenanchin, were present in comb honey from the Coromandel Peninsula.   The test results combined with the symptoms reported were consistent with acute poisoning from tutin and hyenanchin. Some products potentially containing toxic comb honey of the Wentworth Valley and Moana Point brands are still unaccounted for.   Consumers should check if they have any of these brands of Projen Apiaries comb honey and, if so, to hand them in to their nearest public health unit.

Exposure to honey that may result in adverse effects is rare.  Prior to the March 2008 incident, the last recorded case from commercial honey was in 1974 involving 13 patients. There have been nine cases since 1974 with the last known poisoning occurring in 1991 in the EBOP area. Two poisonings have been caused by comb honey produced by hobby beekeepers in the Marlborough Sounds in 1982 and 1983 and the highest levels of tutin ever measured in honey were produced in this area.  

Exposure to honey containing this naturally occurring toxin can produce symptoms such as vomiting, delirium, giddiness, increased excitability, or in more severe cases stupor, coma and violent convulsions. Despite exposure to honey containing this toxin being rare, it is generally accepted that as little as one teaspoon (approximately 10 ml) of unblended honey can have a severe effect on the human nervous system.

NZFSA is compiling data obtained from the medical examinations of affected people and is trying to determine the level of tutin toxin that can be present in honey without causing symptoms or illness in the consumer. In addition, NZFSA is attempting to purify the toxins to understand and determine the level at which acute toxicity symptoms occur. When purified sources are available, the option for commercial testing of honey as part of the ongoing management of these substances will be possible.   Until then, NZFSA will continue using the available animal toxicity data. These data will help NZFSA to establish regulatory levels later this year. NZFSA will also continue to work with bee keepers to expand general knowledge of honey contaminated with tutin toxin.

Research on Key Food Label Elements

Background

Food Standards Australia New Zealand (FSANZ) has just published a report on how manufacturers present key information on their food labels.

Label monitoring surveys have been commissioned by FSANZ since 2002 to assess how food manufacturers manage key labelling requirements such as date marking, directions for use and storage and nutrition information requirements, as set out in the Australia New Zealand Food Standards Code (the Code). In addition, since 2003, surveys have looked at how food manufacturers present nutrition, health and related claims on packaged food labels.

Correct labelling is a key objective of the FSANZ Act 1991 to ensure consumers have adequate information to help them make informed choices.

The report on food labels collected in Australia and New Zealand in 2005 was jointly written by FSANZ and quality assurance agency AsureQuality Limited, which it commissioned to conduct the survey. This survey assessed the 1399 labels collected to determine the degree of consistency with the labelling requirements of the Code for certain key label elements and, where labels were inconsistent with the Code, the nature of the inconsistency with labelling provisions for those label elements assessed. The survey also enabled the gathering of data on other label information of interest. The report presents the results from the assessment of labels collected in 2005 and, where possible, compares them with results from an earlier label collection in 2003 (labels collected by Silliker Microtech Pty Ltd and findings presented in a report dated August 2004).

Key findings

The survey found that for nine of the twelve label elements, consistency with the Code was 90% or greater. Excluding labels that had only minor formatting or moderate inconsistencies in the nutrition information panel (NIP) as their only area of inconsistency, 63% of labels were consistent for all label elements.

Consistency with the Code was 97% or greater for the following label elements:

• label legibility (99%)
• product identification (98%)
• mandatory warning /advisory statements (99%)
• allergen labelling (97%)
• altered label (new label placed over incorrect one) (99%)
• product specific labelling (99%)
• country of origin statements (Australia only) (99%)

Consistency with the Code was 90% or greater for the following label elements:

• ingredient Declaration (92%)
• date marking (90%)

Label elements responsible for higher proportions of inconsistencies (as a percentage of the total number of labels assessed for that particular element) were:

• nutrition information panels (16% inconsistent, excluding labels that had only minor formatting or moderate inconsistencies; 85% inconsistent including labels that had minor formatting, moderate and significant inconsistencies)
• directions for use and storage (14%)
• percent characterising ingredients (11%)

Why is FSANZ conducting ongoing label monitoring surveys?

FSANZ needs to know how food manufacturers are interpreting the Code. Following an extensive review of the Australian Food Standards Code and New Zealand regulations, a joint Australia New Zealand Food Standards Code (the Code) was gazetted in December 2000 with a two year transition period to December 2002.

FSANZ began a pilot label monitoring survey in mid 2002 to assess how manufacturers were implementing the new food labelling provisions in the Code.

Through ongoing surveys, FSANZ can determine the effectiveness of current labelling regulatory measures and use the data to make better informed decisions about labelling laws in future.

What was surveyed?

This survey assessed 1399 labels on packaged food products available for retail sale in Australia and New Zealand during 2005 (746 from Australia and 653 from New Zealand). The foods represented 14 food categories (approximately 2% of the product lines available) and were sampled from a range of outlets including supermarkets and small retailers. The food categories were:

• dairy products
• ·edible oils and emulsions
• ice cream and edible ices
• fruit and vegetables
• confectionery
• cereal and cereal products
•  bread and bakery products
• meat and meat products
• fish and fish products
• egg and egg products
• sugar, honey and related products
• food intended for particular dietary uses
• non-alcoholic beverages
• mixed foods (e.g. sauces, dressings, desserts)

The survey looked at samples from as many different manufacturers as possible. It did not consider the market share of brands because the aim was to sample a wide range of label styles.

What key label elements were assessed?

The survey assessed a total of 12 key label elements, based on the core information required by the Code for a label on a package of food for retail sale. These 12 label elements were: [

1. Legibility of print
2. Product identification
3. Mandatory warning /advisory statements
4. Allergen labelling
5. Ingredient declaration
6. Date marking
7. Directions for use and storage
8. Nutrition information requirements
9. Percent characterising ingredients
10. Altered label (new label placed over incorrect one)
11. Product specific labelling [2]
12. Country of Origin (Australia only)

Labels were also assessed for false, misleading or deceptive representations according to the Australian Trade Practices Act and the New Zealand Fair Trading Act.

What did the survey find?

Overview of the consistency status of labels collected in 2005, by label element

For the 1399 labels collected in 2005 in Australia and New Zealand, consistency with the Code was 90% or greater for nine of the twelve label elements assessed. It should be noted that if any one of the twelve label elements was deemed inconsistent, the label was assessed as inconsistent overall. Excluding labels that had only minor formatting or moderate inconsistencies in the nutrition information panel (NIP) as their only area of inconsistency, 63% of labels were consistent for all label elements.   The consistency of labels for individual label elements is given in Figure 1.

  * n= total number of labels assessed for that particular element.

  ** Excluding labels that had only minor formatting or moderate inconsistencies in the NIP.

Figure 1: Overview of the consistency status of each label element assessed

All labels collected were assessed for legibility and product identification, with 99% and 98% of these assessed as consistent with the labelling provisions respectively. Failure to declare the name and address of the supplier according to the labelling provisions was the greatest cause of labels being assessed as inconsistent with the label element product identification.

Three percent of all labels (42 labels) were assessed for mandatory warnings and advisory statements with only one label assessed as inconsistent with the labelling provisions.

Seventy-one percent of all labels were assessed for allergen labelling with 97% of these assessed as consistent with the labelling provisions. Those labels assessed as having consistent allergen labelling were further assessed to determine the placement and prominence of the declaration. The majority (94%) declared the presence of the allergen in the ingredient list, with 25% featuring an allergen declaration elsewhere on the packet. This included using the name of the allergen in the name of the food, as well as summary statements and voluntary precautionary statements. Increased prominence of allergen declarations (i.e. bold type) was featured on 24% of the labels assessed as consistent for this label element. Gluten containing cereals were declared on 45% of labels with consistent allergen labelling, while milk was declared on 37% of these labels and soybeans on 33% of these labels.

Ninety-one percent of all labels were assessed against the labelling provisions for ingredient declaration with 92% of these labels assessed as consistent. The main reason for inconsistent ingredient labelling was the additive class name being absent or incorrect.

Date marking was assessed on 97% of all labels, and was assessed as consistent on 90% of these. The inconsistencies were due to three reasons: date marking absent, use of an incorrect format and illegible date marking.

Directions for use and storage was also assessed on 97% of all labels, and 86% of these were assessed as consistent with the labelling provisions. The main reason for inconsistency was that directions for use and storage were not provided where required.

Ninety-two percent of the labels collected were assessed as requiring or voluntarily providing a NIP. The level of severity of the NIP inconsistency was assessed; inconsistencies could be categorised as minor formatting inconsistencies, moderate inconsistencies or significant omissions and inconsistencies, according to their potential impact on consumer understanding of the NIP information. Where a NIP had more than one inconsistency, it was rated against the most severe level. Excluding labels that had only minor formatting or moderate inconsistencies, 84% of labels assessed under this element were consistent with the labelling provisions, assuming the NIP was in the prescribed format given in the Code. If minor formatting and moderate inconsistencies are included, 15% of labels assessed under this element were consistent with the labelling provisions.  

Nineteen percent of labels assessed as inconsistent were assessed as having significant omissions and inconsistencies, 5% of labels were assessed as having moderate inconsistencies and 76% of labels had only minor formatting inconsistencies as their most severe level of inconsistency.  

The majority of inconsistencies for the NIP were due to incorrect internal or external borders (51% of all NIP inconsistencies) when compared with the prescribed format in the Code.   This type of inconsistency was assessed as a minor formatting inconsistency as it is less likely to impact on consumer understanding of the information presented on the label. Sixteen percent of NIP inconsistencies related to serving size information.   Only a small percentage of these were due to the absence of serving size information, a significant omission. Most of the remaining inconsistencies for serving size information related to incorrect text case and alignment, these being minor formatting inconsistencies. A further sixteen percent of NIP inconsistencies related to the presentation of nutrients. The most common reason was the use of incorrect text case, a minor formatting inconsistency. Two percent of NIP inconsistencies were due to the absence of a NIP, a significant omission.

Over half of the labels collected (57%) were assessed as requiring or voluntarily providing percent characterising ingredient information. Of these, 89% were assessed as consistent with the labelling provisions. All inconsistencies were due to the absence of percent characterising ingredient information for ingredients emphasised in the name of the food or in the product description.

One label was assessed as inconsistent against the labelling provisions for an altered label, and another label was assessed as inconsistent against the labelling provisions for product specific labelling.

Country of origin statements were assessed on all 746 labels collected in Australia only, as the Country of origin labelling provisions of the Code do not apply in New Zealand. Thirty-two percent of labels were assessed as having ‘Product of’ type statements and 44% featured ‘Made in’ type statements. Of all labels assessed for country of origin statements, nine labels were assessed as having statements that were not in agreement with the labelling provisions.

More than 99% of all labels collected in Australia and New Zealand were assessed as consistent with the provisions of the trade practices legislation covering false, misleading or deceptive representations.

Overview of the consistency status of labels collected in 2005, by food category

Notwithstanding large differences in the total number of foods assessed in each food category (as a result of the agreed product sampling plan), the food categories with the highest proportion of consistency were foods intended for particular dietary use (30% of the total number of labels assessed in this food category), ice cream and edible ices (25%), fruit and vegetables (24%) and non-alcoholic beverages (24%). All labels collected in the food category sugars, honey and related products were inconsistent for one or more label elements. Ninety-five percent of labels in the food category edible oils and oil emulsions and 91% of labels in the food category dairy were inconsistent with at least one label element. Note that these results, as presented in Figure 2, include labels that had NIP inconsistencies at any of the three levels of severity.

* n= total number of labels assessed in each particular category.

  Figure 2: Overview of the consistency status of each food category with the labelling provisions assessed

Overview of the comparison of label assessments on labels collected in 2003 and 2005

Survey methodology differed between the 2003 and 2005 survey. Consequently, there were limitations with respect to comparing the data of these surveys.  

However, the following comparisons were made:

• There were higher proportions of labels consistent with the provisions of the Code between 2003 and 2005 for legibility (91% and 98%, respectively); date marking (82% and 88%, respectively); and percent characterising ingredients (75% and 92%, respectively); and
  
  
• The greatest proportion of inconsistencies in both 2003 and 2005 were identified on NIPs (98% and 94% of labels that were inconsistent, respectively, including labels that had NIP inconsistencies at any of the three levels of severity).  

What will happen now?

The results of this ongoing survey will assist FSANZ assess the effectiveness of labelling regulatory measures and also provide evidence to inform future decisions on labelling laws, as part of the standards development process. Future surveys may also include an assessment of contemporary labelling provisions, such as the assessment of country of origin labelling undertaken for the first time in 2005.

A further report is being prepared regarding labels collected and assessed in 2006. This report is expected to be published late 2008.

It is anticipated that FSANZ will not conduct another label monitoring survey until 2010. This will allow time for the introduction of new labelling requirements (such as the Standard for Nutrition, Health and Related Claims). Future label monitoring surveys may then undertake an assessment of how food manufacturers have managed changes to these new labelling requirements.

It should be noted that ongoing label monitoring surveys are not being conducted for enforcement purposes as this is the responsibility of the governments of Australia and New Zealand and the Australian States and Territories, and not FSANZ. However, FSANZ is continually working with all major stakeholders, including jurisdictions and food industry organisations, to provide manufacturers with helpful information on labelling requirements. Anecdotal evidence from the industry suggests that ongoing label monitoring surveys are one means of achieving this objective.

Copies of the executive summary and full report are now available on the FSANZ website  here  . For further information please contact Janis Baines by phone (+61 2 6271 2234) or email ( janis.baines@foodstandards.gov.au ).

ISSN 1445-8713

© Food Standards Australia New Zealand 2008

You may reproduce articles from Food Surveillance ANZ in unaltered form for your personal, non-commercial use or use within your organisation, with an appropriate acknowledgment. Apart from any other use as permitted under the Copyright Act 1968, all other rights are reserved. Requests and inquiries concerning reproduction and rights should be addressed to Information Officer, FSANZ, PO Box 7186, Canberra BC ACT 2610, Australia