Abstract
The aim of this study was to survey the microbial levels of food contact surfaces in hotels. Microbiological levels of 4611 surfaces (chopping machines, kitchenware, knives, worktops, and cutting boards) from 280 different facilities in Spain were determined in a 3-year period. The contact-plate technique was used throughout the survey. Overall, the mean of the log of total aerobic count cm−2 was 0.62, better than those reported for child-care and assisted living facilities. Significant differences were detected among different types of surfaces, time of sampling, season, and year. The majority (74%) of food contact surfaces sampled in Spanish hotels was within the recommended standard of <1.3 log CFU cm−2, and differences depend on several factors. Our results set a representative picture of the actual situation in our resorts and establish the basis for the development of educational programs to improve food handlers' knowledge of foodborne diseases and their transmission via food contact surfaces.
Introduction
Cross contamination via inanimate surfaces is an important factor in foodborne infections (Deza et al., 2005). Pathogens can easily spread from contaminated food or surfaces to cooking utensils, surfaces, and other food products (Bradford et al., 1997; Kusumaningrum et al., 2003). The transfer of bacteria to food surfaces via dirty cloths has been well documented (Scott and Bloomfield, 1990). Some bacteria, including food pathogens, are able to attach onto surfaces as biofilms (Joseph et al., 2001; Kusumaningrum et al., 2003). Biofilm formation allows bacteria to persist in adverse conditions and become a continuous contamination source of spoilage bacteria and pathogens in the food-processing facilities (Joseph et al., 2001; Midelet and Carpentier, 2002; Silagyi et al., 2009). Because of this adaptation, biofilm bacteria are less sensitive to cleaning and disinfection than planctonic forms (Frank and Koffi, 1990).
Several studies have evaluated the microbial levels of food contact surfaces in processing plants (Eisel et al., 1997; Gounadaki et al., 2008), schools (Henroid et al., 2004; Yoon et al., 2008), assisted living (Sneed et al., 2004) or child-care (Staskel et al., 2007; Cosby et al., 2008) centers, and even domestic kitchens (Haysom and Sharp, 2005). Caterers such as restaurants, cafés and bars, or retail premises (butchers, delicatessens, and bakers) have been also studied (Sagoo et al., 2003). However, data from touristic institutions remain scarce. Spain is a major tourist destination, accounting for ∼7% of arrivals worldwide (WTO, 2008). Tourism is a key economic activity in our population; therefore, important measures are implemented from the public and private sectors to ensure tourist safety. The main objectives of this study were (1) to determine the microbiological levels of food contact surfaces in Spanish hotels and (2) to analyze the effects of different parameters on these levels. This will allow us to evaluate the appropriateness of the current hygiene practices, particularly the cleaning and sanitizing procedures. Further, data will set a basis for the development of educational programs to improve, where needed, the prerequisite programs in our hotels.
Materials and Methods
Sampling procedures
Hotels evaluated in this study were located in the Balearic and Canary Islands and Andalusia, which are among the most important tourist regions in Spain (ITS, 2008). Facilities were audited without prior notice in a monthly basis from April 16, 2007 to December 22, 2009. Food contact surfaces considered clean by the establishment were randomly selected at the time of the visit, with one cutting surface (knives, choppers, blades, etc.) and one flat surface (cutting boards, worktops, etc.) being sampled on each occasion where possible. Different researchers were responsible for sampling and analyzing through the study. Initial training sessions were conducted by the microbiologist, and different internal controls were introduced from beginning to end of the study to ensure homogeneity. Written working instructions were available to all people involved in data compilation.
Microbiological analysis
Microbiological levels were determined using gridded contact plates (Niskanen, 1977; Roberts et al., 2005). Twenty-five-cm2 samples were taken by presenting a contact plate with plate count agar+TLHTh (Sharlab, Barcelona, Spain) against the surface for 10 s, as described in the manufacturing instructions. The plates were refrigerated at 4°C and transported to the laboratory within 24 h. On arrival, samples were incubated at 31°C for 48 h, and the total aerobic count (TAC) was calculated.
Statistical analysis
TAC was transformed to CFU cm−2. As a bacterial count is not a continuous variable, the logarithm of this value was calculated to obtain a continuous variable to be used for descriptive statistics. Means, standard deviation, and frequencies were calculated for each parameter. Statistical analyses were calculated with the SPSS statistical package for Windows (release 10.0.6). Data from different parameters were subjected to one-way analysis of variance, followed by post hoc comparisons by Tukey's multiple-range test to determine significant differences (p<0.05). Based on the recommendations for the food industry proposed by Henroid et al. (2004) and use in several studies (Sneed et al., 2004; Cosby et al., 2008), the standard for food contact surfaces was stated as <1.3 log CFU cm−2.
Results
A total of 4611 samples from 280 different hotels were analyzed throughout the study. Overall, 1197 (26.0%) of the analyzed surfaces showed microbiological levels higher than the recommended standards. Taken as a whole, the mean of the log TAC cm−2 was 0.62±0.85. Results obtained from the different regions were not significantly different (p=0.361): 25.8% of surfaces sampled in the Balearic Islands, 25.7% samples from the Canary Islands, and 29.1% samples from Andalusia were found to have unsatisfactory microbial levels. TAC values were 0.61±0.84, 0.62±0.85, and 0.69±0.87, respectively.
Significant differences were detected among diverse types of food contact surfaces (Table 1). Cutting boards were by far the surfaces with higher microbiological counts (p<0.001), and more than 50% of the analyzed samples did not meet the standard. On the contrary, kitchenware was usually clean and presented lower levels than knives and chopping, mincing, and related machines (p<0.001). Worktops showed intermediate levels.
Different symbols indicate significant differences among types of surfaces (p<0.05).
SD, standard deviation; CFU, colony forming units.
Time of sampling was classified into three categories (Table 2): prelunch (from 8:00 a.m. to 10:59 a.m.), lunch (from 11:00 a.m. to 1:59 p.m.), and after lunch (from 2:00 p.m. to 6:00 p.m.). Differences were detected among the different periods. Surfaces sampled before lunch time clearly presented the best hygienic conditions (p=0.019 and p=0.001 when compared with lunch and postlunch time, respectively), followed by samples analyzed during lunch time (p=0.49 with respect to after lunch).
Significant differences in all cases (p<0.05).
As the present study was carried out during a 3-year period, differences throughout the time, that is, among seasons (Table 3) and years (Table 4), were investigated. Microbiological levels of the surfaces analyzed during summer were higher than in spring and winter (p<0.001). No significant differences were detected between spring, autumn, and winter levels. When surfaces sampled in different years were compared, best results were obtained in 2009 (p<0.001; Table 4).
Different symbols indicate significant differences among seasons (p<0.05).
Different symbols indicate significant differences among years (p<0.05).
Discussion
The establishment of good hygiene practices is a key factor for the prevention of foodborne outbreaks (DeBess et al., 2009). In this study, we have used microbiological monitoring to evaluate the cleaning and sanitization procedures in tourist facilities. The majority (76%) of the sampled food contact surfaces were satisfactory, with the mean TAC of log 0.62 cm−2 being clearly below the recommended standard of <1.3 log CFU cm−2 (Henroid et al., 2004; Sneed et al., 2004; Cosby et al., 2008). In fact, microbial levels are lower that those reported for child-care (3.76 log CFU cm−2) (Henroid et al., 2004) and assisted living facilities (1.51–3.25 log CFU cm−2) (Sneed et al., 2004) and processing plants (2.2–3.7 log CFU cm−2) (Eisel et al., 1997). Different factors may play a role in these differences. Personal hygiene, appropriate kitchen design, and cleaning and sanitizing procedures adapted to public facilities are generally considered as main points to reduce food contact surfaces contamination (Cosby et al., 2008). Probably, the control systems operating in Spanish installations, ranging from good manufacturing practices to full HACCP systems, are a key tool to ensure these satisfactory levels.
Different types of food contact surfaces have been compared, including chopping/mincing machines, kitchenware, knives, worktops, and cutting boards. The last group shows the higher microbiological levels, although the mean TAC value remains under the <1.3 standard. Nevertheless, more than 50% of the analyzed cutting boards presented inappropriate levels. Tebutt (1999) already highlighted the dangers of cross contamination from cutting boards. These results are similar to those obtained in assisted living and child-care centers (Sneed et al. 2004; Staskel et al., 2007). It was noted that these high values almost always corresponded to scored and damaged boards, which, presumably, made cleaning and disinfection of the surface more difficult. This was also observed in food premises in the United Kingdom (Sagoo et al., 2003). Specific training on this issue must be addressed, and food handlers should be informed about the hazards derived from the used of these boards as well as the corrective measures, such as scraping or replacement. In contrast, the kitchenware surfaces had the lowest overall counts. Moreover, when microbiological levels are compared with those obtained for similar surfaces in assisted living facilities (mixing bowls: 2.64 log CFU cm−2), the results obtained for hotels in our study are about four times lower (Sneed et al., 2004). Microbial levels of worktops in our hotels are also lower than those reported for food premises (1.51 log CFU cm−2) (Sagoo et al., 2003), domestic kitchens (3.9 log CFU cm−2) (Kennedy et al., 2005), and Iowa schools (3.76 log CFU cm−2) (Henroid et al., 2004). Contrary to our results, worktops' sanitary conditions in a university food service in New Jersey were more frequently unacceptable than those presented by cutting boards (Buckalew et al., 1996). Finally, in our study, machinery and knives are significantly dirtier than kitchenware. Probably, this is partially due to inappropriate cleaning procedures: commonly, machinery dismantling is necessary before cleaning, and this operation may be sometimes omitted. Again, this is a point that must be emphasized in food handlers training. With respect to knives, it is not unusual to see a food handler cleaning the knives just using a dishcloth. This practice does not clean the knives and may even become a source of contamination (Scott and Bloomfield, 1990). Specific training on these topics is needed to ensure correct cleaning and sanitizing procedures.
Touristic facilities were visited at different times during the day. Sampling time was organized within three groups: before, during, and after the lunch time. Obviously, activities such as food preparation during sampling had a direct effect on the results obtained. As a consequence, results demonstrate that microbiological levels are higher during lunch time with respect to the samples acquired before lunch time. Similar findings have been previously described for child-care centers (Cosby et al., 2008) and domestic kitchens (Haysom and Sharp, 2005). However, microbiological levels were higher after lunch in our study, opposite to what was described by Cosby et al. (2008). This fact indicates that cleaning procedures in this period may be not appropriate. Therefore, special care should be taken and procedures should be modified to obtain similar levels to those obtained in the rest of day.
Our survey monitored the hygienic conditions of surfaces for a 3-year period. This allows us to analyze the presumed differences among seasons and/or years. Values for food contact surfaces collected in different seasons demonstrate that levels of microorganisms are higher in summer (Table 3). It has been well established that seasonality is a major factor in gastroenteritis outbreaks. Food poisoning peaks in summer because, among others factors, hot and wet climate favors bacterial growth. This aspect is often related to the growth of bacteria in the food itself. An increase in Escherichia coli levels in summer was described in school food services (Yoon et al., 2008). In addition, our results indicate that food contact surfaces are likely to become contaminated more easily as well. Therefore, food handlers must be especially careful in food preparation during warm periods.
Results from the evaluation on a yearly basis indicate that hygienic levels of food contact surfaces improve through time. Assessment on different topics was performed during the visits, including information on (i) cleaning of raw and ready-to-eat food must be separated in time or area; (ii) proper storage of equipment including covering equipment, separating the used for raw and ready-to-eat food, and storing gastronorms, bowls, and similar recipients upside down; and (iii) specially accurate cleaning procedures for knives and machines, as stated earlier. Some of these statements have been highlighted elsewhere (Sagoo et al., 2003; Sneed et al., 2004). The results obtained throughout our study indicate that the assessment achieved by the consultants is appropriate and helps the hoteliers to improve the sanitary conditions of their kitchens.
The aim of this study was to survey the cleaning and sanitizing procedures applied to food contact surfaces in touristic services. Overall, food contact surfaces in hotels located in tourist regions of our country meet the internationally recognized standards. Moreover, when compared with other sectors, surfaces present better hygienic conditions that those reported for child-care centers, assisted living facilities, and domestic kitchen. The large number of samples tested and the long period of time and diverse origin of the analyzed facilities make this report a representative picture of the actual situation. Besides, it will be of great help to develop educational strategies based on the differences detected among type of surface, time of sampling, season, and year. They will improve the sanitary conditions in our hotels and avoid undesirable food poisoning episodes that may occur among tourists. Some important aspects should be taken into account when preparing the teaching activities. First, immigrant population with poor Spanish language skills is frequent among food handlers. And, second, we have detected some degree of illiteracy among the population surveyed. Then, educational material must be adapted to ensure effective food safety trainings (DeBess et al., 2009).
Conclusions
Food contact surfaces in Spanish tourist resorts have been surveyed during 2007–2009. Overall, sanitary conditions meet the international recognized standards, indicating that prerequisite systems operating in our facilities are appropriate. In fact, our figures are better than those reported for more sensitive facilities, such as child-care centers and assisted living facilities. Nevertheless, significant differences based on type of surfaces and time parameters have been identified. The compiled data will be useful to elaborate educational programs tailored to improve food handler's knowledge of particular aspects on foodborne diseases, including the role of specific food contact surfaces and environmental temperature in the sanitary conditions that may cause cross contamination of food.
Footnotes
Acknowledgments
The authors thank all the staff who collected samples for this survey. Dr. Sebastián Albertí is acknowledged for the helpful critical review of the article. The authors are especially grateful to David Farrés for his invaluable assistance in data processing.
Disclosure Statement
No competing financial interests exist.
