Environmental Surveillance Unit, Health Protection Agency Communicable Disease Surveillance Centre, 61 Colindale Avenue, London NW9 5EQ, UK

Transcription

1 256 Journal of Food Protection, Vol. 68, No. 2, 25, Pages Copyright, International Association for Food Protection Salmonella and Raw Shell Eggs: Results of a Cross-Sectional Study of Contamination Rates and Egg Safety Practices in the United Kingdom Catering Sector in 23 R. ELSON,* C. L. LITTLE, AND R. T. MITCHELL Environmental Surveillance Unit, Health Protection Agency Communicable Disease Surveillance Centre, 6 Colindale Avenue, London NW9 5EQ, UK MS 4-266: Received 6 June 24/Accepted 3 July 24 ABSTRACT This study was prompted by epidemiological investigations of the unusual number of Salmonella Enteritidis outbreaks associated with the use of eggs in catering premises in England and Wales during 22. The aims of the study, performed between April and May 23, were to establish the rate of Salmonella contamination in raw shell eggs from catering premises, investigate any correlation between the origin and type of eggs and the presence of particular serotypes or phage types (PTs) of Salmonella, and examine the use of raw shell eggs in catering premises in the United Kingdom. A total of 34,6 eggs (5,686 pooled samples of six eggs) were collected from 2,4 catering premises, most of which were eggs produced in the United Kingdom (88%). Salmonella was isolated from 7 pools (.3%) of eggs. Of these, 5 were Salmonella Enteritidis, which were further characterized to PTs as follows: PT6 (.%), PT4 (.7%), PT2 (.4%), PT (.4%), and PT4b (.2%). Salmonella Livingstone and Salmonella Typhimurium definitive type 7 resistant to ampicillin, streptomycin, sulfonamides, and tetracycline were also isolated. The Salmonella contamination rate of eggs produced in the United Kingdom appears to have decreased significantly since 995 and 996. This trend is reflected in the decrease of Salmonella Enteritidis and, in particular, Salmonella Enteritidis PT4. The impact of the United Kingdom Food Standards Agency s advice on the use of eggs, issued in January 23, is discussed. Egg production and handling, from lay to final use, remain the subject of food safety policy in many countries of the world. The decade following the United Kingdom Salmonella and eggs crisis in 988 saw the introduction of legislation, industry codes of practice, and egg assurance schemes intended to control Salmonella within laying flocks and eggs. Most notably, the Lion Code of Practice and the Laid in Britain scheme have been credited, at least in part, with the reduction of human Salmonella Enteritidis phage type (PT) 4 infections and outbreaks in England and Wales since 996 (). The Lion Code of Practice accounts for more than 8% of egg production in the United Kingdom and applies to production from lay to pack and includes vaccination of pullets against Salmonella Enteritidis, biosecurity measures, hazard analysis critical control point (HACCP), and full traceability and feed controls. The Laid in Britain Scheme accounts for approximately 5% of egg production in the United Kingdom and includes the use of competitive exclusion and biosecurity measures but does not include an element of vaccination. The association between Salmonella infection and the consumption of eggs has been highlighted by epidemiological studies (5, 26, 27); however, other studies have failed to make an association (7, 2), although this may have been due to study design, recall bias, confounding, or differing food contamination rates, particularly when comparing dif- * Author for correspondence. Tel: ; Fax: ; richard.elson@hpa.org.uk. ferent countries. Outbreaks of salmonellosis have been attributed to the use of raw shell eggs, including their consumption raw (4) or undercooked (5, 6) and/or acting as a source of cross-contamination (29). Eggs continue to be implicated as a source or vehicle of infection in outbreaks, including those that occur in premises that serve vulnerable groups (23). Between September 22 and November 23, 59 outbreaks of Salmonella Enteritidis occurred in England and Wales, affecting more than 2,262 people; this accounted for more than a fifth (22%) of all human Salmonella Enteritidis cases confirmed by the Health Protection Agency (HPA) Laboratory of Enteric Pathogens in the same period. More than half (58%) of these outbreaks were linked to consumption of foods prepared with raw shell eggs purchased from catering premises (24). Results of an HPA public health investigation of eggs associated with these outbreaks showed an overall Salmonella Enteritidis contamination rate of 6.7% in both Spanish and unlabeled eggs (country of origin not known) (2). In response, the United Kingdom Food Standards Agency (FSA) reiterated advice to the public and the catering industry concerning the proper use of raw shell eggs (, 2); they advised all egg importers and wholesalers in the United Kingdom to send all eggs from Spain for commercial heat treatment (). The FSA renewed this advice to caterers in January 24 following an outbreak of Salmonella Enteritidis associated with a cafe where lightly cooked dishes were made with Spanish eggs (2).

2 J. Food Prot., Vol. 68, No. 2 SALMONELLA CONTAMINATION OF EGGS AND THEIR USE BY CATERERS 257 TABLE. Salmonella serotypes and phage types (PTs) isolated from hens eggs (7 pools of six eggs) Salmonella serotype and PT No. of positive pools (% of total pools) (n 5,686) Lion Code No. of positive pools per egg assurance scheme Laid in Britain Neither scheme or not recorded Enteritidis PT Enteritidis PT4 Enteritidis PT6 Enteritidis PT2 Enteritidis PT4b Livingstone Typhimurium DT7 ASSuT a 2 (.4) 4 (.7) 6 (.) 2 (.4) (.2) (.2) (.2) Total 7 (.3) 7 9 a Salmonella Typhimurium definitive type 7 resistant to ampillicin, streptomycin, sulfonamides, and tetracycline. This study was prompted by the multiple common source outbreaks of Salmonella Enteritidis, coupled with a marked change in the epidemiology of Salmonella Enteritidis attributed to a number of non-pt4 Salmonella Enteritidis strains. The aims of the study were to establish the rate of Salmonella contamination of raw shell eggs used by catering premises, investigate any correlation between the origin and type of eggs and the presence of particular serotypes and PTs of Salmonella, and examine the use of raw shell eggs in catering premises in the United Kingdom. MATERIALS AND METHODS Sample collection. Samples of raw shell eggs were collected by staff from 33 local authority environmental health departments in the United Kingdom between April and 3 May 23 using a standardized protocol. Samples were transported to 38 official food control laboratories in accordance with the Food Safety Act 99 Code of Practice no. 7 (8) and guidance on microbiological food sampling (22). Information on the catering premises was obtained by observation and inquiry and recorded on a standard form. This included information on the premises and practices with regard to the preparation, storage, and use of raw shell eggs; potential for crosscontamination; presence and documentation of a hazard analysis system; and the level of food hygiene training received by the manager. In the United Kingdom, food hygiene inspections are performed in accordance with the Food Safety Act 99 Code of Practice no. 9 (9), which specifies that, among other factors, the number of consumers at risk and confidence in management control systems (including the application of HACCP-based systems) should be assessed to produce a risk rating of the premises. The risk rating determines the frequency of inspection and ranges from category A (highest risk, inspected every 6 months) to category F (lowest risk, inspected every 5 years). Confidence in management scores range from (high confidence) to 3 (no confidence). Sample examination and detection of Salmonella. On receipt of the sample, the examining laboratory recorded the appearance of the eggs (i.e., whether they were clean, dirty, contaminated with feces, or visibly cracked). In designing the study, eggs were regarded as a potential source of Salmonella contamination in the catering environment irrespective of the site of contamination; therefore, no distinction was made between contamination of egg or shell contents. Pooled samples of six eggs were cracked against the rim of a sterile screw-top container, and the shells and egg contents were placed into the container. Where sterile screwtop containers were not available, eggs were cracked against the rim of a sterile honey jar, and the egg contents were transferred to a stomacher bag with closure and homogenized for 5 to 3 s followed by the addition of the shells that had been crushed by a gloved hand in a stomacher bag. An equal volume of buffered peptone water was added, and the solution mixed and incubated at 37 C for 8 2 h. The incubated buffered peptone water broths were then subcultured, incubated, and examined in accordance with the HPA standard method F3 (4) with the following modifications: the Rappaport Vassiliadis soya peptone and selenite cysteine broths were reincubated for another 24 to 26 h and subcultured to selective agars. On examining selective agar plates for the presence of Salmonella, plates with no Salmonella colonies were reincubated for a further 22 2 h and reexamined for the presence of Salmonella. Isolates of Salmonella were sent either to the Laboratory of Enteric Pathogens at the HPA Specialist Reference Microbiology Division or to the Scottish Salmonella Reference Laboratory for confirmation, serotyping and phage typing for Salmonella Enteritidis and Salmonella Typhimurium, and antimicrobial resistance typing. Data analysis was performed using Microsoft Excel, and proportions were compared using the 2 test and 2 test for trend. Results with P.5 were deemed to be of statistical significance. RESULTS A total of 34,6 eggs (5,686 pooled samples of six eggs) were collected from 2,4 catering premises. Almost all eggs sampled were hens eggs (33,796; 99.%); 4 (.3%) were ducks eggs, and for 26 eggs (.6%), the species of origin was not recorded. Figures and proportions reported from this point forward are based on pooled samples of six eggs rather than individual eggs. Salmonella contamination. Seventeen pools (.3%) were found to contain Salmonella, all of which were hens eggs. Fifteen of the Salmonella isolates were identified as Salmonella enterica serovar Enteritidis (Table ). Of these, five different PTs were identified, with PT6 being the most common (six pools;.%) followed by PT4 (four pools;.7%), PT2 (two pools;.4%), PT (two pools;.4%), and PT4b (one pool;.2%). Salmonella Livingstone (one pool;.2%) and Salmonella Typhimurium definitive type 7 resistant to ampicillin, streptomycin, sulfonamides, and tetracycline (one pool;.2%) were also isolated (Table ).

3 258 ELSON ET AL. J. Food Prot., Vol. 68, No. 2 Egg assurance schemes. Most eggs (88%) sampled were produced in the United Kingdom (Table 2). A third (34%) of the eggs were produced under the auspices of the Lion Code, 7% were produced as part of the Laid in Britain scheme, and 47% were produced under neither of these schemes (Table 2). Of the pools that contained Salmonella, seven (.36%) were recorded as being produced under the auspices of the Lion Code scheme, one (.2%) was produced under the Laid in Britain scheme, and eight (.3%) were not produced under either of these schemes (Table 2). There was no evidence of any difference between the Salmonella contamination rate in eggs produced in the United Kingdom of Lion Code quality (7 of,947;.4%) and eggs produced in the United Kingdom not of Lion Code quality (eggs produced under either the Laid in Britain or no assurance scheme) (9 of 3,4;.3%) (P.89). However, it is possible that eggs not marked with the Lion Code symbol could also have come from vaccinated flocks. Egg size, production method, and appearance. Egg sizes varied from small (2%), medium (69%), large (22%), to very large (%) (Table 2). Salmonella was detected only from medium (.3%) or large (.6%) eggs (Table 2). More than a third (36%) of the eggs were labeled as cage, 5% as free range, 9% barn, and 2% as other (direct from farm, home reared), and the production method for 38% of eggs was either not known or not recorded (Table 2). Of the Salmonella-positive pools of eggs, seven (.3%) were cage produced, four (.5%) were free range, and the production method for the remaining five positive pools (.3%) was not recorded (Table 2). Salmonella was not isolated from the small proportion (%) of egg samples labeled as organic (Table 2). Two-thirds (67%) of the egg shells in the pooled samples were recorded as being visually clean, 4% as being contaminated with feces, 2% as dirty,.7% as having visible cracks present, and 25% of pooled samples contained various combinations of clean, cracked, soiled, and/or presence of feathers (Table 2). More than half (53%) of the positive pools contained eggs that appeared clean; 35% contained various combinations of clean, cracked, soiled, and/or presence of feathers; and 6% had visible cracks in their shells (Table 2). Labeling details. Two-thirds (67%) of the eggs sampled had a best before date available either on the egg or its packaging compared with more than a quarter (27%) of eggs that were not accompanied by this information (Table 2). Packing station details were only available for half the samples (5%) (Table 2). Premises visited and managers awareness of the FSA s egg safety advice. Cafes (22%), restaurants (7%), and hotels (6%) formed more than half the premises visited (Table 3). Other premises visited are listed in Table 3. Fifty-nine percent (,248 of 2,4) of the proprietors or managers of the catering premises visited were aware of the FSA s egg safety advice, 7% (363 of 2,4) were partly aware, and 2% (42 of 2,4) were not aware of this advice. TABLE 2. Details of egg samples and Salmonella-positive pools Egg details Country of origin United Kingdom Spain Germany Portugal Republic of Ireland Holland Italy Not known Egg assurance scheme Lion Code Laid in Britain Neither scheme Egg production method Cage Free range Barn Other (farm gate, home reared) Not known or not recorded Organic production Labeled as organic Not labeled as organic Appearance of eggs Clean Combination a Feces Dirty Visible cracks Egg size Small Medium Large Very large Unknown Best before date Present Not present Packing station details Available Not available Total no. (%) of pools of eggs examined (n 5,686) 4,987 (88) 22 (.4) (.2) 7 (.) 3 (.) 3 (.) 3 (.) 65 (),947 (34) 49 (7) 2,669 (47) 66 (2) 2,34 (36) 845 (5) 52 (9) 9 (2) 2,76 (38) 3 () 5,78 (89) 577 () 3,827 (67),427 (25) 249 (4) 4 (2) 4 () 29 () 43 (2) 3,9 (69),228 (22) 6 () 6 () 348 (6) 3,8 (67),56 (27) 323 (6) 2,822 (5) 2,76 (38) 688 (2) No. (%) of pools Salmonella positive (n 7) 6 (.3) (.2) 7 (.4) (.2) 8 (.3) (.2) 7 (.3) 4 (.5) 6 (.3) 5 (.3) 2 (.4) 9 (.2) 6 (.4) (2.4) (3.5) (.3) 7 (.6) (.3) 4 (.3) 2 (.6) 9 (.3) 5 (.2) 3 (.4) a Pooled sample contained combinations of eggs that were clean, cracked, soiled, and/or had feathers present on their surface.

4 J. Food Prot., Vol. 68, No. 2 SALMONELLA CONTAMINATION OF EGGS AND THEIR USE BY CATERERS 259 TABLE 3. Details of catering premises and managers awareness of Food Standards Agency s egg safety advice Premises details No. (%) of premises (n 2,4) Managers awareness of egg safety advice, no. (%) a Aware Unaware 47 (22) 366 (7) 339 (6) 25 () 26 (6) 96 (5) 94 (4) 92 (4) 85 (4) 72 (3) 58 (3) 56 (3) 23 () (2) 244 (52) 25 (59) 26 (64) 65 (77) 8 (64) 4 (43) 49 (52) 4 (45) 32 (38) 57 (79) 47 (8) 42 (75) 2 (25) 6 (55) 3 (22) 78 (2) 58 (7) 23 () 28 (22) 3 (3) 22 (23) 26 (28) 28 (33) 7 () 3 (5) 7 (3) 3 (3) 4 (36) 82 (4) 635 (3) 9 (43) 85 (9) 23 () 269 (3) 64 (78) 426 (67) 59 (56) 84 (45) 4 (6) 5 (56) (2) 9 (9) 86 (2) 5 (28) 5 (22) 49 (8),5 (53) 425 (2) 269 (3) 295 (4) 763 (68) 23 (54) 96 (36) 58 (54) 8 (6) 85 (2) (37) 55 (9) Premises type Café Restaurant Hotel Residential care home Public house Take-away establishment Sandwich bar Bakery Mobile caterer Schools or educational establishments Hospital Canteen Other (leisure and conference centres, catering companies) Confidence in management High (good record of compliance) Moderate Some Little No (poor record of compliance) Hazard analysis In place and documented In place and undocumented Not in place Management food hygiene training Received training and attended a(n) Advanced course Intermediate course Basic 6-h course Other recognized course Training not specified Not trained Hand washing facilities Accessible and available for use And used And not used Unable to assess during visit Not accessible and available for use,895 (9) 59 (8) 349 (8),274 (67) 42 (2) 7 (4) 73 (3) 36 (7),59 (6) 23 (77) 26 (75) 78 (56) 25 (6) 42 (59) 27 (37) 62 (46) 363 (9) 8 () 45 (3) 276 (22) 7 (7) 7 (24) 28 (38) 29 (2),993 (95),334 (67) 29 () 44 (22) 39 (2) 72 (3),2 (6) 926 (69) 76 (35) 84 (42) 2 (3) 5 (69) 392 (2) 2 (6) 9 (42) 89 (2) 8 (46) 2 (7) a Managers only partly aware of the Food Standards Agency s egg safety advice were not included in this calculation. Almost half (43%) of the premises visited had a confidence in management score of (some confidence) (Table 3). Awareness of the FSA s food safety advice on the use of eggs decreased significantly with decreasing confidence in the management (high confidence, 78%; moderate confidence, 67%; some confidence, 56%; little confidence, 45%; no confidence, 6%) ( 2 for trend P.) (Table 3). More than half (53%) of the premises visited had a documented hazard analysis system in place, 2% had an undocumented system, and 3% had no hazard analysis system in place (Table 3). Awareness of the FSA s egg safety advice was significantly higher among managers of premises where there was some form (documented or undocumented) of hazard analysis in place (65%) compared with managers of premises that had no hazard analysis system in place (36%) (P.) (Table 3). Most premises visited (9%) had managers who were

5 26 ELSON ET AL. J. Food Prot., Vol. 68, No. 2 TABLE 4. Food safety practices intended to prevent cross-contamination and managers awareness of Food Standards Agencys egg safety advice Egg safety practice b No. (%) of total premises (n 2,4) Managers awareness of egg safety advice, no. (%) a Aware Unaware 325 (5),634 (78) 45 (7) 76 (54),6 (62) 66 (46) 77 (24) 36 (9) 27 (9) Eggs stored in direct contact with RTE foods Yes No Use of raw shell eggs Cracked open in preparation areas used for: RTE foods Raw food only Eggs beaten near to RTE food or surfaces or equipment used for RTE food Yes No,265 (6) 727 (35) 2 (5) 696 (33),34 (63) 94 (4) 74 (59) 457 (63) 5 (45) 378 (54) 829 (63) 4 (44) 267 (2) 38 (9) 5 (3) 46 (2) 259 (2) 5 (6) a Managers only partly aware of the Food Standards Agency s egg safety advice were not included in this calculation. b RTE, ready-to-eat. trained in food hygiene, and two-thirds of these (67%) had received basic food hygiene training, with 8% and 8% having undertaken intermediate and advanced training, respectively (Table 3). An additional 2% of managers had received other forms of training (in-house training, formal catering qualifications, and HACCP training) (Table 3). Awareness of the FSA s egg safety advice increased among managers as their level of training increased ( 2 for trend P.). In most premises visited (95%), hand washing facilities were available and accessible for use (Table 3). These facilities were used, as judged by the sampling officer, in 67% of the premises visited but were not used in % of premises (Table 3). Egg safety practices and managers awareness of the FSA s egg safety advice. Eggs were stored separately from ready-to-eat (RTE) foods in more than three-quarters (78%) of the premises visited but were not in 5% of premises (Table 4). Of the premises where eggs were stored separately from RTE foods, a higher proportion of managers were aware of the FSA s egg safety advice (62%) compared with managers of premises who stored eggs in contact with RTE foods (54%) (P.4) (Table 4). By cracking eggs in RTE food preparation areas, RTE foods or preparation surfaces were exposed to the risk of contamination in 6% of premises. Eggs were cracked near RTE food areas or RTE food preparation areas in a lower proportion of premises where the manager was aware of the FSA s egg safety advice (59%) compared with those premises where the manager was not aware of this advice (63%) (Table 4). Eggs were beaten near RTE food, surfaces, or equipment used for RTE food in a third (33%) of premises compared with 62% of premises where they were not. Eggs were beaten near RTE foods and surfaces in a lower proportion (54%) of premises with managers who were aware of the FSA s egg safety advice compared with those premises where the manager was not aware of this advice (63%) (Table 4). There was an almost equal split between premises that stored eggs below or equal to 8 C (5%) and those that stored them above 8 C (48%) (Table 5). However, awareness of the FSA s egg safety advice was more common among managers of premises where eggs were stored below or equal to 8 C (69%) than managers of premises where eggs were stored above this temperature (5%) (P.) (Table 5). Eggs without a best before date were significantly more likely to be stored above 8 C (894 of 2,679; 33%) compared with those provided with a best before date (657 of 2,922; 22%) (P.). Eggs were used as an ingredient in lightly cooked foods in just more than a quarter (26%) of the premises visited and in uncooked foods in only 3% of the premises visited. Dishes that contained lightly cooked or uncooked eggs were more commonly served immediately or refrigerated ( 8 C) than stored above refrigeration temperatures (Table 5). Where eggs were prepared and served fried, boiled, or as omelets or similar dishes, they were sometimes served lightly cooked in 48% of premises and always cooked lightly in 6% of premises (Table 5). Awareness of the FSA s eggs safety advice was more common among managers of premises where egg dishes were never lightly cooked (66%) compared with managers where these dishes were sometimes or always lightly cooked (56%) (P.2) (Table 5). Premises that serve vulnerable groups. Thirty-nine premises (2%) served food to vulnerable groups (groups of 2 persons or more who are either younger than 5 or older than 65 years, sick, or immunocompromised). All of these premises were residential care homes, schools, and hospitals. Most premises (95%) had a hazard analysis system in place and trained management (95%), most of whom (85%)

6 J. Food Prot., Vol. 68, No. 2 SALMONELLA CONTAMINATION OF EGGS AND THEIR USE BY CATERERS 26 TABLE 5. Food safety practices intended to prevent Salmonella growth and survival and managers awareness of Food Standards Agency s egg safety advice Egg safety practice No. (%) of total premises (n 2,4) Managers awareness of egg safety advice, no. (%) a Aware Unaware Storage temperature of raw shell eggs Stored equal or below 8 C Stored above 8 C,68 (5),3 (48) 33 () 734 (69) 56 (5) 8 (24) 7 (6) 247 (25) 3 (9) Eggs used in lightly cooked foods and: Served immediately Equal or below 8 C Above 8 C Storage temperature not recorded Not used in lightly cooked foods 539 (26) 4 (74) 75 (4) 2 (4) 43 (7),454 (69) (5) 326 (6) 242 (6) 46 (6) 2 (57) 26 (6) 863 (59) 59 (53) 5 (9) 79 (2) 6 (2) 6 (29) 4 (9) 35 (2) (9) Eggs used in uncooked foods and: Served immediately Equal or below 8 C Above 8 C Storage temperature not recorded Not used in uncooked foods 66 (3) 3 (47) 23 (35) 3 (5) 9 (3),67 (76) 43 (2) 4 (6) 2 (65) 3 (57) (33) 6 (67) 968 (6) 24 (56) 4 (2) 5 (6) 7 (3) (33) () 32 (2) 85 (2) Eggs fried, boiled, or used in omelets that are: Never lightly cooked Sometimes cooked lightly Always cooked lightly Degree of cooking not recorded Dishes not prepared,886 (9) 768 (4) 898 (48) 9 (6) (5) 46 (7) 72 (3),3 (6) 58 (66) 496 (55) 72 (6) 55 (54) 85 (58) 32 (44) 373 (2) 3 (7) 92 (2) 25 (2) 26 (26) 24 (24) 7 (7) a Managers only partly aware of the Food Standards Agency s egg safety advice were not included in this calculation. were aware or had some awareness of the FSA s egg safety advice (Table 6). A higher proportion (27 of 39; 69%) of managers from premises that served vulnerable groups were aware of this advice compared with other premise types (,76 of,8; 59%). Surprisingly, approximately half of the premises that served vulnerable groups used cracked (56%) or beaten (46%) eggs near RTE foods and/or surfaces, equipment, and preparation areas used for RTE foods (Table 6). No premises that served vulnerable groups used eggs in uncooked foods; however, a fifth (2%) used eggs as an ingredient in lightly cooked foods (Table 6). Fried and boiled TABLE 6. Food safety practices and managers awareness of Food Standard Agency s egg safety advice in premises that serve food to vulnerable groups Premises and food safety criteria b No. (%) of premises (n 39) Managers awareness of egg safety advice, no. (%) a Aware Not aware Manager trained Hazard analysis in place Hazard analysis documented Eggs cracked near RTE or RTE and raw foods 37 (95) 37 (95) 28 (72) 22 (56) 27 (73) 27 (73) 2 (75) 4 (64) 6 (6) 6 (6) 4 (4) 4 (8) Eggs beaten near RTE foods or surfaces or equipment used for RTE foods 8 (46) 2 (67) 3 (7) Fried or boiled eggs or omelets Sometimes lightly cooked Always lightly cooked Eggs used as an ingredient In lightly cooked foods In uncooked foods 6 (4) 2 (5) 8 (2) (63) (5) 5 (63) a Managers only partly aware of the Food Standards Agency s egg safety advice were not included in this calculation. b RTE, ready-to-eat. 3 (9) (5) 3 (38)

7 262 ELSON ET AL. J. Food Prot., Vol. 68, No. 2 eggs along with omelets and similar egg dishes were sometimes served lightly cooked in 4% of premises that served vulnerable groups, and, at the time of sampling, 5% of premises that served vulnerable groups always served these dishes lightly cooked (Table 6). DISCUSSION No egg can be guaranteed to be free of Salmonella irrespective of its origin or production method. However, compared with a previous study performed in 995 and 996, the Salmonella contamination rate of eggs produced in the United Kingdom appear to have decreased significantly (995 to 996,.%; 23,.3%; rate ratio [RR],.32; 95% confidence interval [CI],.9 to.54; P.). This trend has been reflected in the decrease in the overall rate of Salmonella Enteritidis contamination (995 to 996,.9%; 23,.3%; RR,.33; 95% CI,.9 to.57; P.54) and in particular Salmonella Enteritidis PT4 (995 to 996,.6%; 23,.%; RR,.4; 95% CI,.5 to.37; P.9), although there appears to have been no significant change in the rate of non- PT4 Salmonella Enteritidis contamination (995 to 996,.3%; 23,.2%; RR,.76; 95% CI,.38 to.52; P.536) (). These findings are mirrored in the.3% Salmonella Enteritidis prevalence reported by the FSA in its survey of eggs on retail sale performed in 23 (3); however, the apparent status quo in the contamination rate of non-pt4 Salmonella Enteritidis emphasizes the need for continued veterinary, food, and human surveillance to detect and manage future Salmonella problems should they arise. Vaccination of layer flocks, combined with improved flock hygiene, appears to have made a significant impact on the prevalence of Salmonella contamination of eggs and on human Salmonella infection (). The results of this study did not provide any evidence to show a difference in the rate of Salmonella contamination between eggs produced in the United Kingdom from vaccinated flocks (Lion Code quality) and those from unvaccinated flocks (non Lion Code quality). This finding may reflect that a proportion, or all, of the non Lion Code eggs were in fact from flocks partially or fully vaccinated against Salmonella but were not identifiable as such at the time of sampling. It is noteworthy that the egg supply chain to the United Kingdom catering market differs from that of the retail market; caterers receive most of the small volume of imported eggs and eggs not produced under United Kingdom assurance schemes. Previous studies have shown imported eggs to have a higher rate of Salmonella contamination compared with eggs produced in the United Kingdom. In 993, de Louvois (3) reported a Salmonella contamination rate of.6%, higher than that of eggs produced in the United Kingdom but with a lower rate of Salmonella Enteritidis contamination (United Kingdom,.7%; non United Kingdom,.2%) and Salmonella Enteritidis PT4 (United Kingdom,.5%; non United Kingdom,.2%) (3). A 996 to 997 survey of eggs imported by the United Kingdom and intended for retail sale found that 2% of samples contained Salmonella,.3% contained Salmonella Enteritidis, and.% contained Salmonella Enteritidis PT4 (). The rate of contamination found in this study was lower than that found by the HPA public health investigation of eggs that commenced in October 22 and continued to April 24; 6.7% of Spanish eggs and 6.7% of unlabeled eggs (country of origin not known) were contaminated with Salmonella serotypes, most of which were non-pt4 Salmonella Enteritidis (2). The eggs sampled for the public health investigation were suspected food vehicles associated with outbreaks at catering premises, which undoubtedly had an influence on the high contamination rates reported. The volume of eggs imported to the United Kingdom fluctuates throughout the year and closely reflects supply and demand. Although salmonellae were not found in the small number of non United Kingdom eggs in this study, the nature of the United Kingdom catering egg market may allow the sporadic introduction of highly contaminated eggs. Caterers can reduce the risk of their customers becoming infected with Salmonella by ensuring that eggs are stored correctly, avoiding contamination of RTE foods from egg shells and/or contents, and ensuring that salmonellae are killed through proper cooking (7). Combining refrigeration and good stock rotation is an easily achieved control measure to prevent low levels of salmonellae from achieving harmful population sizes. However, this study showed that a half the caterers did not store their eggs under refrigerated conditions, and eggs without a best before date were more likely to be stored without refrigeration. European legislation now requires individual eggs to be stamped with details of their origin and production type; however, eggs are not required to be stamped with a best before date (25). Caterers should therefore pay close attention to egg stock rotation and refrigeration, particularly when using eggs that are not stamped with a best before date. Contamination can occur during egg preparation (particularly when egg mixes are aerosolized during whisking) or through contact with food preparation equipment, preparation surfaces, and food handlers hands (9). Once on surfaces, salmonellae can survive well (more than 24 h) in dried batter or egg mixes (9). Sagoo et al. (28) highlighted deficiencies in cleaning standards in food premises in the United Kingdom, including contamination of cleaning cloths with Salmonella and deficiencies in the use of cleaning products. It is therefore important that effective cleaning schedules are established, along with separation of cleaning equipment used to clean RTE and non-rte areas. Salmonellae in contaminated eggs can also survive certain cooking methods (8), and eggs should always be cooked in line with the FSA s advice. The FSA s guidance appears to have been absorbed to a greater extent by the managers of hospitals, schools, and residential nursing homes, premises that may be expected to serve food to vulnerable groups. However, other premises, such as mobile caterers and take-away establishments, appear not to have absorbed this information, which may reflect the peripatetic nature and high staff turnover of some of these types of businesses. This finding emphasizes the important role of local food authorities in ensuring that cen-

8 J. Food Prot., Vol. 68, No. 2 SALMONELLA CONTAMINATION OF EGGS AND THEIR USE BY CATERERS 263 trally issued advice is adopted locally and factored into programmed food hygiene inspections. In the United Kingdom, some caterers adopt practices that may allow infection of consumers should the eggs used be contaminated. Of particular concern is that these practices persist in a small number of premises that serve vulnerable groups, even where the manager is aware of the FSA s egg safety advice. In line with this advice, local food enforcement officers should always recommend that pasteurized egg be used in these premises. Hope et al. (6) have modeled the effect of five risk mitigation scenarios to estimate the number of and reduction in Salmonella Enteritidis cases from eggs and egg products in the United States. Although the model had limitations, the authors concluded that food safety policies directed at one segment of the farm-to-fork continuum alone are less effective than a broad-based approach (6). The United Kingdom Advisory Committee on the Microbiological Safety of Food has attributed the decrease in Salmonella Enteritidis cases in England and Wales to the raft of measures taken by the United Kingdom egg industry since 998. This is a welcome public health intervention; however, poor food safety practices in the final part of the food chain combined with the use of eggs that may be highly contaminated have the potential to undermine these efforts. Eggs are a commonly consumed food that may occasionally be contaminated with salmonellae. Caterers who continue to use raw shell eggs need to be aware of the Salmonella hazard presented by eggs and implement recommended control measures, particularly if serving food to elderly, very young, or immunocompromised persons. ACKNOWLEDGMENTS This study was performed on behalf of the Local Authority Coordinators of Regulatory Services and the HPA. The authors thank all the staff in the environmental health departments throughout the United Kingdom who collected samples for this study and all the staff in HPA, HPA collaborating, and non-hpa laboratories who performed the microbiological examinations. Thanks are also extended to the Laboratory of Enteric Pathogens (HPA Specialist and Reference Microbiology Division) and Stobhill Hospital for typing Salmonella isolates, to David Lock at Local Authority Coordinators of Regulatory Services for coordinating the participation of environmental health officers and advice from the Local Authority Coordinators of Regulatory Services Food Examination Focus Group, to the Regional Food, Water and Environmental Coordinators Forum for their advice in preparing the sampling protocol, and to Lilian Hucklesby and Louise Harte for entering the data. REFERENCES. Advisory Committee on the Microbiological Safety of Food. 2. Second report on Salmonella in eggs. The Stationery Office, London. 2. Anonymous. 24. Salmonella Enteritidis outbreak in central London linked to Spanish eggs. CDR Wkly. [serial online]. 4:3. Available at: ent. Accessed 8 June de Louvois, J Salmonella contamination of eggs. Lancet 342: Ejidokun, O. O., D. Killalea, M. Cooper, S. Holmyard, A. Cross, and C. Kemp. 2. Four linked outbreaks of Salmonella enteritidis phage type 4 infection the continuing egg threat. Commun. Dis. Public Health 3: Evans, M. R., P. G. Hutchings, C. D. Ribeiro, and D. Westmoreland A hospital outbreak of Salmonella food poisoning due to inadequate deep-fat frying. Epidemiol. Infect. 6: Evans, M. R., S. M. Parry, and C. D. Ribeiro Salmonella outbreak from microwave cooked food. Epidemiol. Infect. 5: Food Standards Agency. 2. A report of the study of infectious intestinal disease in England. The Stationery Office, London. 8. Food Standards Agency. 2. Food Safety Act 99 (as amended) Code of Practice no. 7: sampling for analysis and examination (revised October 2). Available at: multimedia/pdfs/cop7. Accessed 8 June Food Standards Agency. 2. Food Safety Act 99 (as amended) Code of Practice no. 9: food hygiene inspections (second revision October 2). Available at pdfs/cop9. Accessed 8 June 24.. Food Standards Agency. 22. Salmonella outbreak leads agency to issue hygiene alert: food businesses advised to use properly cooked or pasteurized eggs [press release R49 28]. Available at: Accessed 3 June 24.. Food Standards Agency. 22. Agency re-emphasizes advice on use and handling of all eggs and issues guidance on use of Spanish eggs [press release R57 28]. Available at: pressreleases/reemphasiseeggadvice. Accessed 2 June Food Standards Agency. 23. Eggs what caterers need to know. Available at: Accessed 3 June Food Standards Agency. 24. Report of the survey of Salmonella contamination of UK produced shell eggs on retail sale. Available at: Accessed 3 June Health Protection Agency. 23. Methods for food and dairy products detection of Salmonella species: standard method F3. Available at: esl su/sops docs/fsops/ F3i.3.pdf. Accessed 8 June Hennessy, T. W., L. H. Cheng, H. Kassenborg, S. D. Ahuja, J. Mohle-Boetani, R. Marcus, B. Shiferaw, and F. J. Angulo. 24. Egg consumption is the principal risk factor for sporadic Salmonella serotype Heidelberg infections: a case-control study in FoodNet sites. Clin. Infect. Dis. 5(38 Suppl. 3):S237 S Hope, B. K., R. Baker, E. D. Edel, A. T. Hogue, W. D. Schlosser, R. Whiting, R. M. McDowell, and R. A. Morales. 22. An overview of the Salmonella enteritidis risk assessment for shell eggs and egg products. Risk Anal. 22: Humphrey, T. J. 99. Growth of Salmonellas in intact shell eggs: influence of storage temperature. Vet Rec. 26: Humphrey, T. J., M. Greenwood, R. J. Gilbert, B. Rowe, and P. A. Chapman The survival of Salmonellas in shell eggs cooked under simulated domestic conditions. Epidemiol. Infect. 3: Humphrey, T. J., K. W. Martin, and A. Whitehead Contamination of hands and work surfaces with Salmonella enteritidis PT4 during the preparation of egg dishes. Epidemiol. Infect. 3: Kapperud, G., J. Lassen, and V. Hasseltvedt Salmonella infections in Norway: descriptive epidemiology and a case-control study. Epidemiol. Infect. 2: Little, C. L. 24. Personal communication ( christine. little@hpa.org.uk). 22. Local Authority Coordinators of Regulatory Services (LACORS). 22. LACOTS guidance on food sampling for microbiological examination. Available to subscribers at: tempbe/ pdf. Accessed 8 June Meakins, S. M., G. K. Adak, B. A. Lopman, and S. J. O Brien. 23. General outbreaks of infectious intestinal disease (IID) in hospitals, England and Wales, J. Hosp. Infect. 53: O Brien, S. J. 24. Personal communication ( sarah. o brien@hpa.org.uk). 25. Official Journal of the European Communities L2/ Council regulation (EC) no. 5/2 of 9 December 2 amending regulation (EEC) no. 97/9 on certain marketing standards for

9 264 ELSON ET AL. J. Food Prot., Vol. 68, No. 2 eggs. Available at: pdfs%5creg5.pdf. Accessed 8 June Palmer, S., S. Parry, D. Perry, R. Smith, M. Evans, L. Nehaul, R. Roberts, M. Walapu, and D. Wright. 2. The role of outbreaks in developing food safety policy: population based surveillance of Salmonella outbreaks in Wales Epidemiol. Infect. 25: Parry, S. M., S. R. Palmer, J. Slader, and T. Humphrey. 22. Risk factors for Salmonella food poisoning in the domestic kitchen a case control study. Epidemiol. Infect. 29: Sagoo, S. K., C. L. Little, C. J. Griffith, and R. T. Mitchell on behalf of the Local Authorities Coordinators of Regulatory Services and the Public Health Laboratory Service. 23. Study of cleaning standards and practices in food premises in the United Kingdom. Commun. Dis. Public Health 6: Wight, J. P., J. Cornell, P. Rhodes, S. Colley, S. Webster, and A. M. Ridley Four outbreaks of Salmonella enteritidis phage type 4 food poisoning linked to a single baker. Commun. Dis. Rep. CDR Rev. 9:R2 R5.