International Journal of Food Microbiology, 13 (1991) 225-230
225
© 1991 Elsevier Science Publishers B.V. 0168-1605/91/$03.50 FOOD 00390
Practical monitoring of the chill chain Alex Tolstoy FDB (F~ellesforeningenfor Danmarks Brugsforeninger), Department of Product Quality and Consumer Service, Albertsluna[ Denmark
Monitoring of temperature of chilled products at fresh food terminals is integrated into the quality assurance program of a retail chain. Each batch of products is examined on the receiving dock and rejected if the temperature is found above 7 o C, 5 o C being the legal maximal temperature. Measurement of temperature of proces~d meat products delivered to our company over a period of 14 months indicates that temperature abuse of chilled foods is not uncommon. Consistent rejection of batches because of high temperature resulted in a drop in the number of violations. Key words: Quality assurance; Chill chain: Temperature recordings: Monitoring
Introduction Chilling of perishable foods is widely used, alone or in combination with other preservation techniques such as low a w, low p H or modified atmosphere packaging as a means of prolonging shelflife. The effects of low temperature on chemical and biological processes are well documented. Endogenous spoilage processes are delayed and for both raw and processed meats growth of spoilage and food poisoning bacteria is significantly retarded (Brown, 1982). The biochemical changes that occur in bacteria include changes in permeability of the m e m b r a n e as well as changes in metabolic activity (Brown, 1982). In chilled foods the lag phase of psychrophilic and mesophilic bacteria is prolonged, frequently from hours to days, depending on the bacterial species as well as the temperature at which the food is stored. However, once the lag phase is passed, psychrophilic bacteria will grow at near maximal growth rate upon entering the logarithmic phase. In contrast, generation times of mesophilic bacteria are significantly longer in chilled foods as compared to growth at optimal temperature (Jepsen and Skovgaard, 1975). Both phenomena are exploited by the m o d e m food industry to prolong shelflife of perishable foods. The ability of pathogenic bacteria to grow at low temperatures has been somewhat underestimated. More and more of the traditional food poisoning bacteria as well as emerging pathogens are being recognized as capable of growth within or Correspondence address: A. Tolstoy, FDB, Department of Product Quality and Consumer Service, 65
Roskildevej, 2620 Albertslund, Denmark.
226
near the 0-5 ° C range (Doyle. 1989). For practical reasons it is extremely important that investigators specify under what conditions growth is determined. Indeed, growth in pure culture and in laboratory media is not the same as in food, where nutrients may be less available and competing bacteria may be present in high numbers. Chilling foods after processing and maintaining a temperature below 5 ° C are not easy operations for food production on an industrial scale. Professional manufacturers will be familiar with the problems of chilling and know their causes, be they inadequate cold storage facilities, seasonal problems due to high external temperature, premature stacking following cooking, a heat-generating slicing or sealing process or purely negligence in production or during transportation. The trend among consumers for a greater variety of chilled foods often implies more centralised distribution systems in order to supply the many outlets of a retail chain. Therefore many different products will be in close contact in fresh food terminals and distribution vehicles. Danish national legislation prescribes an array of 'keep below' temperatures for various products, e.g., processed food (5°C), pasteurized milk (6 ° C), fermented milk products (8 ° C), butter (10 ° C), shell eggs (12°C). Such temperature limits are based on technological as well as hygienic considerations. Although good sense was used in setting these limits, differences in 'keep below' temperatures are unfortunate because the most sensitive products tend to suffer through contact with warmer products. Monitoring also becomes more difficult. Since May, 1989, systematic temperature monitoring of chilled processed foods has been integrated into the quality control program of our company. Limits have been set in accordance with national legislation. Systematic temperature monitoring of deliveries takes place at fresh food terminals. Distribution to outlets is done via refrigerated trucks. Deliveries outside opening hours are made into specially designed, locked cold stores. Outlets are submitted to regular hygiene inspections and staff are cautioned against overfilling of cabinets. Gunvig (1988b) actually reports minor influences of chill cabinet conditions on product quality due to high turnover rates of the vulnerable upper layers of stacked products. The greatest impact on overall product quality is therefore achieved at the supplier level. Indeed, chilling installations at fresh food terminals and at outlets are only designed to maintain low product temperatures. Temperature measurements are best done immediately upon arrival of the products, and a predetermined action plan must be ready when violations occur. Suppliers to our company are reprimanded if temperature limits are slightly exceeded. If, however, the legal limit of 5 ° C is exceeded by more than 2 . 0 ° C in the warmest place of the pallet, the pallet is rejected and will not be accepted, even after chilling is completed.
Materials and Methods
Deliveries of processed foods are manually~ scanned for the location of the highest temperature, usually one or two layers from the top of the pallet. Non-de-
22?
structive measurements are made by piling several packages onto the sensor probe of a digital thermometer. If the temperature is borderline or too high, destructive measurements are made by inserting the probe into the product. Efforts are focussed on the most perishable items, including raw and processed meats, dairy products and mayonnaise salads. Registration is limited to batches that are rejected due to temperature breaches or underweight.
Results
Fig. 1 shows the number of batches of processed meats rejected from June, 1989, through July, 1990. The batches monitored ranged in weight from approx. 50 kg to a maximum of 10 t. Only batches of processed meats with temperatures exceeding 7 ° C are included in the figure, since monitoring of this product group has been consistent over the entire period. Monthly variation in numbers of rejected batches from June, 1989, through July, 1990, is shown in Fig. 1. The influence of the external temperature is clearly reflected in the drop in number of rejections during the winter months. The number of rejections during the summer of 1989 is probably extraordinarily high, since the monitoring program was only started in May, 1989. Following many productive discussions regarding procedures for random sampling, thermometer calibration and chilling technology, suppliers quickly became well acquainted with the monitoring scheme. The decreases in number of rejections observed over the summer periods of 1989 and 1990 are most likely due to increasing efforts of suppliers to minimize Number of rejected batches
25 2O 15
10 5
0 1981) June
aug
oct
1~)O___ m r
.,my
I~K)
month Fig. I. Diagram showing a number of processed meat products rejected at fresh food terzmrsalsdue to temperature violations over the period of June, 1989 through July 1990.
228 TABLE 1 Serious breaches of the chill chain. Processed meat products found to have a temperature above 10 ° C over the period June. 1989 through July. 1990 Item
Month
Product temp.
Pat6 Hotfilled ready meals a Bockwurst Beef salami (sliced) Smoked ring sausage Emulsion sausage Liver paste Chick Vienna sausage Emulsion sausage (sliced) Sardell sausage Frankfurters Bratwurst Cooked salami
June June July August September October April May June June June July July
11.4 10.3-13.0 17.1 10.0 12.0 16.6 10.4 10.9 11.9 12.3 11.7 10.2 11.2
°
C
All products were packaged under modified atmosphere or vacuum packed and were prescribed to be stored below 5 . C. " F o u r varieties.
temperature abuse, realising that routine practices used in the cold season are not suitable for the high temperature season. Listed in Table I are types of processed meat products that were rejected during the period of May, 1989 through May, 1990 because the temperature exceeded
lO°C. No correlation was observed between the size of packing company and the number of rejected batches over the period (data not included).
Discussion and Conclusion Systematic rejection of products exceeding well-defined temperature limits has caused a significant reduction in the number of shipments violating temperature limits. At the time of rejection, products are generally not judged to be hazardous or even spoiled; however, shelf life was undoubtedly reduced. In Denmark, shelf life is declared by the supplier based on his own experience, frequently with no consideration of safety margins for temperature violations. At any rate, destruction of the products is generally not considered necessary. Relabelling with a reduced shelf life or reworking are the manufacturers only options, however cumbersome, by which the consumers are not supplied with inferior products relative to the price paid. Manufacturers authorized to export products are required to obtain permission by the veterinary authorities before relabelling or reworking. It remains uncertain what actually happens to rejected batches. Are they simply delivered to another retailer or are they actually relabelled or reprocessed?
229 In complete accordance with investigations performed by 3uul Sorensen (1985). G u n v i g (1987-89, 1988a, b,c) and Gunvig and Bogh-Sorensen (1987), practical temperature monitoring demonstrates that serious breaches of the chill chain are not u n c o m m o n occurrences a t the supplier level. Provided the products are not consumed at a late stage of shelf life, the consumer will not be aware of such abuse. Should the c o n s u m e r choose to eat the p r o d u c t shortly before expiration of shelf life, serious dissapointment will occur due to premature spoilage of the p r o d u c t or, at worst, food poisoning. Strengthening of and strict adherence to the chill chain is needed. Realisation of temperature abuse can easily be improved by simply recording temperature, e.g. with probe thermometers. It is recommended that each step along the chill chain should institute temperature monitoring of perishable foods on a routine basis.
References Brown, M,H. (1982) Meat Microbiology. Applied Science Publishers Ltd. Essex, U.K. Doyle, M.R. (1989) Foodbome Bacterial Pathogens, Marcel Dekker Inc., New York. Gunvig, A. and Begh-Serensen, L. (1987) Temperature conditions during chilled distribution. Danish Meat Products Laboratory, Roskilde, Denmark (unpublished). Gunv/g, A. (198"7-89) Chill Chain. Danish Meat Products Laboratory, Roskilde, Denmark (unpublished). Gunvig, A. (1988a) Shelflife of pork chops at different temperatures l&ll. Danish Meat Products Laboratory, Roakilde, Denmark (unpublished). Gunvig, A. (1988b) Temperature conditions during transportation of fresh pork I. Danish Meat Products Laboratory, Ro~kilde, Denmark (unpublished). Gunvig, A. (1988c) Time temperature relationship in chill cabinets. Danish Meat Research Laboratory, Roskilde, Denmark (unpublished). ICMSF. (1980) Microbial Ecology of Foods. Vol. I, Academic Press, New York. Jepsen, A. and Skovgaard, N. (19"/5) Levnedsmiddelm/krobiologi og Hyglejne. 3rd. edn., Carl Ft. Monensen, Copenhagen. 3uul Szre~sen, I. (1985) Storage results II of chilled meat products. Danish Meat Products Laboratory. Roskilde, Denmark (unpublished).
225
© 1991 Elsevier Science Publishers B.V. 0168-1605/91/$03.50 FOOD 00390
Practical monitoring of the chill chain Alex Tolstoy FDB (F~ellesforeningenfor Danmarks Brugsforeninger), Department of Product Quality and Consumer Service, Albertsluna[ Denmark
Monitoring of temperature of chilled products at fresh food terminals is integrated into the quality assurance program of a retail chain. Each batch of products is examined on the receiving dock and rejected if the temperature is found above 7 o C, 5 o C being the legal maximal temperature. Measurement of temperature of proces~d meat products delivered to our company over a period of 14 months indicates that temperature abuse of chilled foods is not uncommon. Consistent rejection of batches because of high temperature resulted in a drop in the number of violations. Key words: Quality assurance; Chill chain: Temperature recordings: Monitoring
Introduction Chilling of perishable foods is widely used, alone or in combination with other preservation techniques such as low a w, low p H or modified atmosphere packaging as a means of prolonging shelflife. The effects of low temperature on chemical and biological processes are well documented. Endogenous spoilage processes are delayed and for both raw and processed meats growth of spoilage and food poisoning bacteria is significantly retarded (Brown, 1982). The biochemical changes that occur in bacteria include changes in permeability of the m e m b r a n e as well as changes in metabolic activity (Brown, 1982). In chilled foods the lag phase of psychrophilic and mesophilic bacteria is prolonged, frequently from hours to days, depending on the bacterial species as well as the temperature at which the food is stored. However, once the lag phase is passed, psychrophilic bacteria will grow at near maximal growth rate upon entering the logarithmic phase. In contrast, generation times of mesophilic bacteria are significantly longer in chilled foods as compared to growth at optimal temperature (Jepsen and Skovgaard, 1975). Both phenomena are exploited by the m o d e m food industry to prolong shelflife of perishable foods. The ability of pathogenic bacteria to grow at low temperatures has been somewhat underestimated. More and more of the traditional food poisoning bacteria as well as emerging pathogens are being recognized as capable of growth within or Correspondence address: A. Tolstoy, FDB, Department of Product Quality and Consumer Service, 65
Roskildevej, 2620 Albertslund, Denmark.
226
near the 0-5 ° C range (Doyle. 1989). For practical reasons it is extremely important that investigators specify under what conditions growth is determined. Indeed, growth in pure culture and in laboratory media is not the same as in food, where nutrients may be less available and competing bacteria may be present in high numbers. Chilling foods after processing and maintaining a temperature below 5 ° C are not easy operations for food production on an industrial scale. Professional manufacturers will be familiar with the problems of chilling and know their causes, be they inadequate cold storage facilities, seasonal problems due to high external temperature, premature stacking following cooking, a heat-generating slicing or sealing process or purely negligence in production or during transportation. The trend among consumers for a greater variety of chilled foods often implies more centralised distribution systems in order to supply the many outlets of a retail chain. Therefore many different products will be in close contact in fresh food terminals and distribution vehicles. Danish national legislation prescribes an array of 'keep below' temperatures for various products, e.g., processed food (5°C), pasteurized milk (6 ° C), fermented milk products (8 ° C), butter (10 ° C), shell eggs (12°C). Such temperature limits are based on technological as well as hygienic considerations. Although good sense was used in setting these limits, differences in 'keep below' temperatures are unfortunate because the most sensitive products tend to suffer through contact with warmer products. Monitoring also becomes more difficult. Since May, 1989, systematic temperature monitoring of chilled processed foods has been integrated into the quality control program of our company. Limits have been set in accordance with national legislation. Systematic temperature monitoring of deliveries takes place at fresh food terminals. Distribution to outlets is done via refrigerated trucks. Deliveries outside opening hours are made into specially designed, locked cold stores. Outlets are submitted to regular hygiene inspections and staff are cautioned against overfilling of cabinets. Gunvig (1988b) actually reports minor influences of chill cabinet conditions on product quality due to high turnover rates of the vulnerable upper layers of stacked products. The greatest impact on overall product quality is therefore achieved at the supplier level. Indeed, chilling installations at fresh food terminals and at outlets are only designed to maintain low product temperatures. Temperature measurements are best done immediately upon arrival of the products, and a predetermined action plan must be ready when violations occur. Suppliers to our company are reprimanded if temperature limits are slightly exceeded. If, however, the legal limit of 5 ° C is exceeded by more than 2 . 0 ° C in the warmest place of the pallet, the pallet is rejected and will not be accepted, even after chilling is completed.
Materials and Methods
Deliveries of processed foods are manually~ scanned for the location of the highest temperature, usually one or two layers from the top of the pallet. Non-de-
22?
structive measurements are made by piling several packages onto the sensor probe of a digital thermometer. If the temperature is borderline or too high, destructive measurements are made by inserting the probe into the product. Efforts are focussed on the most perishable items, including raw and processed meats, dairy products and mayonnaise salads. Registration is limited to batches that are rejected due to temperature breaches or underweight.
Results
Fig. 1 shows the number of batches of processed meats rejected from June, 1989, through July, 1990. The batches monitored ranged in weight from approx. 50 kg to a maximum of 10 t. Only batches of processed meats with temperatures exceeding 7 ° C are included in the figure, since monitoring of this product group has been consistent over the entire period. Monthly variation in numbers of rejected batches from June, 1989, through July, 1990, is shown in Fig. 1. The influence of the external temperature is clearly reflected in the drop in number of rejections during the winter months. The number of rejections during the summer of 1989 is probably extraordinarily high, since the monitoring program was only started in May, 1989. Following many productive discussions regarding procedures for random sampling, thermometer calibration and chilling technology, suppliers quickly became well acquainted with the monitoring scheme. The decreases in number of rejections observed over the summer periods of 1989 and 1990 are most likely due to increasing efforts of suppliers to minimize Number of rejected batches
25 2O 15
10 5
0 1981) June
aug
oct
1~)O___ m r
.,my
I~K)
month Fig. I. Diagram showing a number of processed meat products rejected at fresh food terzmrsalsdue to temperature violations over the period of June, 1989 through July 1990.
228 TABLE 1 Serious breaches of the chill chain. Processed meat products found to have a temperature above 10 ° C over the period June. 1989 through July. 1990 Item
Month
Product temp.
Pat6 Hotfilled ready meals a Bockwurst Beef salami (sliced) Smoked ring sausage Emulsion sausage Liver paste Chick Vienna sausage Emulsion sausage (sliced) Sardell sausage Frankfurters Bratwurst Cooked salami
June June July August September October April May June June June July July
11.4 10.3-13.0 17.1 10.0 12.0 16.6 10.4 10.9 11.9 12.3 11.7 10.2 11.2
°
C
All products were packaged under modified atmosphere or vacuum packed and were prescribed to be stored below 5 . C. " F o u r varieties.
temperature abuse, realising that routine practices used in the cold season are not suitable for the high temperature season. Listed in Table I are types of processed meat products that were rejected during the period of May, 1989 through May, 1990 because the temperature exceeded
lO°C. No correlation was observed between the size of packing company and the number of rejected batches over the period (data not included).
Discussion and Conclusion Systematic rejection of products exceeding well-defined temperature limits has caused a significant reduction in the number of shipments violating temperature limits. At the time of rejection, products are generally not judged to be hazardous or even spoiled; however, shelf life was undoubtedly reduced. In Denmark, shelf life is declared by the supplier based on his own experience, frequently with no consideration of safety margins for temperature violations. At any rate, destruction of the products is generally not considered necessary. Relabelling with a reduced shelf life or reworking are the manufacturers only options, however cumbersome, by which the consumers are not supplied with inferior products relative to the price paid. Manufacturers authorized to export products are required to obtain permission by the veterinary authorities before relabelling or reworking. It remains uncertain what actually happens to rejected batches. Are they simply delivered to another retailer or are they actually relabelled or reprocessed?
229 In complete accordance with investigations performed by 3uul Sorensen (1985). G u n v i g (1987-89, 1988a, b,c) and Gunvig and Bogh-Sorensen (1987), practical temperature monitoring demonstrates that serious breaches of the chill chain are not u n c o m m o n occurrences a t the supplier level. Provided the products are not consumed at a late stage of shelf life, the consumer will not be aware of such abuse. Should the c o n s u m e r choose to eat the p r o d u c t shortly before expiration of shelf life, serious dissapointment will occur due to premature spoilage of the p r o d u c t or, at worst, food poisoning. Strengthening of and strict adherence to the chill chain is needed. Realisation of temperature abuse can easily be improved by simply recording temperature, e.g. with probe thermometers. It is recommended that each step along the chill chain should institute temperature monitoring of perishable foods on a routine basis.
References Brown, M,H. (1982) Meat Microbiology. Applied Science Publishers Ltd. Essex, U.K. Doyle, M.R. (1989) Foodbome Bacterial Pathogens, Marcel Dekker Inc., New York. Gunvig, A. and Begh-Serensen, L. (1987) Temperature conditions during chilled distribution. Danish Meat Products Laboratory, Roskilde, Denmark (unpublished). Gunv/g, A. (198"7-89) Chill Chain. Danish Meat Products Laboratory, Roskilde, Denmark (unpublished). Gunvig, A. (1988a) Shelflife of pork chops at different temperatures l&ll. Danish Meat Products Laboratory, Roakilde, Denmark (unpublished). Gunvig, A. (1988b) Temperature conditions during transportation of fresh pork I. Danish Meat Products Laboratory, Ro~kilde, Denmark (unpublished). Gunvig, A. (1988c) Time temperature relationship in chill cabinets. Danish Meat Research Laboratory, Roskilde, Denmark (unpublished). ICMSF. (1980) Microbial Ecology of Foods. Vol. I, Academic Press, New York. Jepsen, A. and Skovgaard, N. (19"/5) Levnedsmiddelm/krobiologi og Hyglejne. 3rd. edn., Carl Ft. Monensen, Copenhagen. 3uul Szre~sen, I. (1985) Storage results II of chilled meat products. Danish Meat Products Laboratory. Roskilde, Denmark (unpublished).
