The objectives of this study were to examine domestic food safety knowledge levels of consumers, establish the levels and incidence of bacterial contamination and operational temperatures in domestic refrigerators, and identify areas in which consumer food safety education is necessary in Ireland. A food safety knowledge questionnaire applied to a representative sample of households (n = 1,020) throughout the island of Ireland found the gaps in consumer food safety knowledge. Analysis of swab samples (n = 900) recovered from the domestic refrigerators in these households showed average total viable counts of 7.1 log CFU/cm2 and average total coliform counts of 4.0 log CFU/cm2. Analysis of swab samples also detected the incidence of Staphylococcus aureus (41%), Escherichia coli (6%), Salmonella enterica (7%), Listeria monocytogenes (6%), and Yersinia enterocolitica (2%). Campylobacter jejuni and E. coli O157:H7 were not detected in domestic refrigerators. The temperature profiles of a subset of the sampled refrigerators (100) were monitored for 72 h, and 59% were found to operate, on average, at temperatures above the recommended 5 degrees C. Knowledge and temperature survey results varied considerably, but consumers who scored better in terms of basic food safety knowledge had reduced levels of bacterial contamination in their refrigerators and reported a reduced incidence of food-associated illnesses. This study confirms the effect of basic food hygiene knowledge on hygienic practice and identifies specific areasfor emphasis in the development and delivery of effective food safety risk communication messages to consumers.

No abstract

Domestic food contact surfaces can play an important role in the transmission of foodborne disease, yet debate continues as to which surface materials pose the greatest risk to consumer health in terms of cross-contamination during food preparation. Salmonella Typhimurium was inoculated onto stainless steel, Formica, polypropylene, or wooden surfaces (25 cm2) in the presence or absence of protein (tryptic soy broth supplemented with 5% horse serum) and held at room temperature. The pathogen was recovered from the test surfaces immediately after inoculation (T=0) and every hour for up to 6 h, by a conventional microbiological sampling technique and by direct transfer onto a model ready-to-eat food (cucumber slices). On all surfaces, pathogen numbers declined during the 6-h holding period, with the most rapid reductions occurring within the first hour. The presence of protein significantly increased (P < 0.05) the number of bacteria recovered from all surface types. However, regardless of application medium or holding time, the number of bacteria recovered from Formica (in all cases) and stainless steel (in most cases) was significantly higher than were the numbers on polypropylene or wood. Similarly, regardless of application medium or holding time, significantly higher bacterial numbers were transferred to the model food from Formica or stainless steel than from polypropylene or wooden surfaces. These differences were greater when the bacteria were applied in a protein-rich medium and the test surfaces held for 1 h or more. The results of this study emphasize that differences, both in recoverability and in the number of bacteria transferred to the model food rather than simply reflecting differences in pathogen survival, may also reflect differences in the ability of the test bacteria to remobilize from the different surface types. However, the results also demonstrate a fundamental problem when choosing food contact surfaces, i.e., that those characteristics that make a surface "easy to clean" may also render it more likely to release contaminating pathogens during common food preparation practices.

The survival characteristics of Escherichia coli O157:H7 in silage derived from contaminated grass were investigated. The survival of other enteric bacteria was also investigated to determine if E. coli O157:H7 demonstrates enhanced acid tolerance in comparison. Samples of chopped grass were treated as follows: (i) no additive (control); (ii) inoculation with E. coli O157:H7 to a final concentration of log10 4.0 CFU g(-1); (iii) addition of an 85% solution of formic acid at 3.0 ml kg(-1) grass; and (iv) addition of both E. coli O157:H7 and formic acid, at the above concentrations. Treated 6-kg grass samples were packed into laboratory silos, sealed, and stored at 15 degrees C for up to 180 days. Individual replicate silos were removed from storage periodically and subjected to microbiological and chemical analyses. Chemical analyses of the silage samples indicated that lactic acid-dominant fermentations, with a rapid drop in pH, occurred. Numbers of enteric bacteria decreased from log10 7.0 to 8.0 CFU g(-1) to undetectable levels within 19 days' storage. E. coli O157:H7 did not survive the silage fermentation process, with numbers declining from approximately log10 4.0 CFU g(-1) to undetectable levels within 19 days of ensiling. The pattern of decline in numbers of E. coli O157:H7 was the same as that for the enteric bacteria, indicating that under the conditions tested, the acid tolerance of E. coli O157:H7 was not significantly different from the acid tolerance of other enteric bacteria. This study found that E. coli O157:H7 did not survive a good silage fermentation process, indicating that properly ensiled grass that is correctly stored is unlikely to be a vector for the transmission of the pathogen among cattle.

1997. The use of a novel surface adhesion technique to isolate Listeria monocytogenes and Listeria innocua from an enrichment meat system was developed. Minced beef samples inoculated with L. rnonocytogenes (10 cfu g-') were incubated at 30°C for 1 4 1 8 h in a suitable enrichment broth. Listeria monocytogenes cells were isolated from the enriched meat sample by surface adhesion onto a polycarbonate membrane which was attached to a glass microscope slide. T h e Listeria cells on the membrane were subsequently visualized using an immunofluorescent microscopy procedure. T h e antibody used in this technique reacts with L. monocytogenes and L. innocua. T h e technique was demonstrated to have a detection level of log,, 3.11 cfu m1-l. There was excellent correlation ( 2~0 . 9 8 ) between the counts obtained by this surface adhesion immunofluorescent (SAIF) technique and counts obtained using traditional methods, i.e. plate counts on PALCAM. When the regression equation relating the rapid and standard methods was validated using the data from 50 retail beef mince samples, an rsd value of f 0.25 was obtained. No false-negative or false-positive results were recorded for L. rnonocytogenes or L. innocua species using the SAIF technique.