A strategic cleaning assessment program: menu cleanliness at restaurants.
Cleanliness of the environment in which food is being prepared, served, and consumed is critical in reducing the potential for foodborne illness. Foodborne illness outbreaks can certainly damage a restaurant's reputation and lead to a loss of revenue. Previous research found that 70% of consumers would no longer buy food from a food service establishment where they had concerns about hygiene (Food Safety Agency, 2008). Knight and co-authors (2007) found that people who perceived that a restaurant was "not at all" committed to food safety were less likely to choose that restaurant when eating out. In fact, at least one study found that cleanliness was the most important determinant for consumers' perceptions of restaurant food safety (Henson et al., 2006).
Consumers are likely to judge the cleanliness of a restaurant on visual perceptions. Similarly, although health inspectors use an inspection manual and the food code to inspect restaurants, their judgments also rely heavily on visual assessment. Where visual observations are used, subjective assessments may also be needed to quantify cleanliness. Moore and Griffith (2002) state, "'Cleanliness' is a relative concept--what is acceptable as being 'clean' in one situation may be unacceptable in another (p. 318)." This perceptual difference was found in a previous study in which health inspectors showed variations in their opinions of cleanliness (Lee, Almanza, Nelson, & Ghiselli, 2009). The typical assessment of what is clean, therefore, relies heavily on visual assessment, which may be subjective and is likely to differ from one individual to another.
As bacterial and viral contaminations are not detectable by visual assessment, the lack of microbiological analysis can be problematic. The results of studies using hygiene swabs and agar contact plates have shown that visual inspection is a poor indicator of cleaning (Griffith, Cooper, Gilmore, Davis, & Lewis, 2000; Moore & Griffith, 2002). Microbiological assessment of restaurants is generally not conducted, however, as part of the inspection process since traditional microbiological analyses require 48-72 hours after the sample is collected to obtain results. Equipment such as an adenosine triphosphate (ATP) meter provides a faster assessment of cleaning, but ATP meters assess organic soils (which include food soil and other organic residues in addition to microorganisms) and are expensive for routine inspections.
Furthermore, consistent cleaning of certain surfaces outside the kitchen may not be done in all restaurants. This may be particularly true for furniture, equipment, and other frequently used items such as menus. Contaminants on menus can be transferred to guests' hands and subsequently to food being consumed. Cleaning of menus is commonly done in many restaurants; however, the need for cleaning is based more on esthetic considerations and may simply be done by visual inspection or by touching the menu. Standards or protocols to clean menus or even to determine when the menu needs to be cleaned have not yet been established. This is potentially a concern as it has been demonstrated that microorganisms could be transferred from damp menus to fingertips for up to 24 hours for certain types of menus (Sirsat, Choi, Almanza, & Neal, 2013). The purpose of our study was therefore to assess the cleanliness of menus in a restaurant and the impact of different factors (such as type of cleaning method and how menus are stored or distributed) on menu contamination.
Determination of the Cleanliness of a Surface Using ATP Meters
The cleaning of food service equipment or furniture in a restaurant depends on the protocols of that facility. Capable restaurant managers institute their own cleaning and sanitizing schedules for the restaurant to facilitate cleaning and sanitizing procedures. Factors influencing the choice of hygiene practice methods include cost, time, staff, ease of use, management needs, and nature of the food contact surfaces (Griffith, Blucher, Fleri, & Fielding, 1994).
More recently, ATP meters have been suggested as a less time-intensive and acceptable method for determining the cleanliness of a surface (Griffith et al., 2000). They measure the bioluminescence from ATP (an energy-containing substance present in living cells) from microorganisms, food residues (or other organic materials), and humans (Worsfold & Griffith, 1996). Because of the potential for contamination on menus and the possible transfer of contamination onto customers' hands, it is important to consider menu cleaning practices and their effectiveness. More specifically, the purpose of our study was to assess the contamination on restaurant menus to determine if typical cleaning methods are effective and the impact of different factors (such as type of cleaning method and how menus are stored or distributed) on menu contamination.
ATP values of 500 relative light units (RLU) for a clean surface are considered a realistic upper critical limit (Griffith et al., 2000). The use of ATP meters in research studies is thought to be advantageous in that it is a rapid test that provides results within minutes and is a more cost-effective means to monitor surface cleanliness than traditional microbiology (Griffith et al., 1994). The use of the ATP technique has also been suggested for the restaurant and food service industry to indicate the level of potential cross contamination of food (Leon & Albrecht, 2007). A comparison of ATP bioluminescence and traditional swabbing methods for the determination of surface cleanliness at a hospital kitchen showed both techniques were highly correlated (Aycicek, Oguz, & Karci, 2006). Hence, because of the advantages of ATP testing and its correlation to traditional swabbing methods, our study used ATP meters to assess the cleanliness of food contact surfaces.
Materials and Methods
In order to validate the most appropriate areas of the menu to test for the impact of cleanliness, a pretest was conducted to identify the high-touch areas of the menu by consumers. Because restaurants use a variety of different menu styles, this pretest included four styles of menus. Two sizes were tested (letter size and legal size) for two different menu formats (single page and multipage). The menus were made with a high-quality color copy glossy 32 lb. weight paper similar to that used by many restaurants. To standardize consumers' visual and tactile use of the menus, both single-page menus had a restaurant's name on one side and the menu printed on the other. The two single-page menus listed the same menu items using the same font and type size (Times New Roman, 12 point); the only difference was the spacing between the menu items for the letter vs. legal sizes. To standardize consumers' visual and tactile use of the menus for the multipage formats, the same food items using the same font and type size were again used for the letter vs. legal sizes. The multipage menus were created by folding the paper on the long side resulting in a four-page menu that measured either 8 1/2" x 5 1/2" for the letter size paper or 8 1/2" x 7" for the legal size paper. For both multipage menus, the name of the restaurant was printed on the front and the menu was printed on the two inside pages.
The research investigator and three field workers visited with a group of 36 students enrolled in a hospitality and tourism management program to explain the purpose of the study and ask if students were willing to participate in the pretest. Seventeen students agreed. Participants were asked to rub their hands with a fluorescing liquid that would leave traces on menus when they were touched and then be visible under ultraviolet light. Menus were presented one at a time to each study participant. Each participant evaluated all four menus following the protocol described above. To ensure participants used the menu as they would in a restaurant, they were handed the menu after being seated and asked to select their choice of entree, beverage, and dessert as if they were eating in that restaurant. After collecting the first menu from the participants, the second menu was presented to the participants. Similarly, after completing the second menu, the third menu and finally the fourth menu were presented. A coupon for a gourmet cookie was then given to the participants in appreciation for their participation. Menus were stacked on clean sheets of paper so that the liquid could dry thoroughly without smearing or contacting other menus.
Menus were then analyzed visually using a 5-watt ultraviolet disclosing lamp black light. To quantify the areas of highest touch and identify the areas for testing in the later study, the menu was divided into smaller units. A transparent grid marked with 2.8" squares was laid over the menu to ascertain if any touch contact occurred in these smaller units. This grid was then used to create a map of consumer contact on the menus. Contact patterns were determined by recording a positive result each time an area had been touched as indicated by the ultraviolet visible fingerprints left on the menu.
Results of the pretest were then used to determine which areas of the menus to evaluate for the main study. Results showed that the lower and outer sides of the menus were most likely to be touched for both sizes of the two page menus (Figure 1). One difference for the 8.5" x 11" menu was that the lower half of the menu was most likely to be touched, whereas for the 8.5" x 14" menu, it was not the bottom half but an area representing a similar distance from the top as the shorter menu that was most likely to be touched. The lowest section of the 8.5" x 14" menu was in fact less likely to be touched, particularly for the one page menus. Multiple page menus were unique in that the cover of the menu was less likely to be touched on the left side of the menu. The one-page letter-size menu did not show distinct patterns and was therefore excluded from the analyses.
The owner of a casual-family dining restaurant chain agreed to allow our study to be conducted in one of his restaurants using his menus. The restaurant represented a small chain that offered sandwiches, salads, pizza, and desserts. Information was collected regarding menu storage, menu distribution and collection procedures, and menu cleaning procedures. After discussing possible research questions in this real-world setting, permission was received from the owner to evaluate the impact of three factors on menu cleanliness. The first question was to find out the impact of typical cleaning methods on menu cleanliness (before being cleaned vs. after being cleaned by the staff). The second question was to find out how the type of cleaning method (spray vs. wet cloth) impacted cleanliness. Finally, the third question was to determine the impact of menu storage and distribution method (servers handing them out vs. leaving menus in a holder on the table).
Prior to the start of data collection all menus were collected and thoroughly cleaned using 91% isopropyl alcohol. This standardized the beginning level cleanliness among the restaurant menus. The menus were circulated in the restaurant for two weeks and the staff cleaned menus according to their routine cleaning policies. Routine cleaning practices included the use of a commercial grade chlorine-based solution as a sanitizer. The standard cleaning practice in the restaurant was to spray the menus with this sanitizer and then wipe it from the menus using a washable cotton cloth. All menus were normally cleaned at the end of each shift. A fresh cleaning cloth was used for each shift.
Two rooms were used in the restaurant to determine the impact of handing out menus vs. leaving them on the table in a storage rack. This was already a standard practice in the restaurant (one room normally had servers hand out menus that were stored at the host station, the other room had wire racks on each table that stored the menus and were readily available for customers to use without the need to have one handed to them by a server). Identical menus were used in each of these two rooms. The staff were asked not to exchange menus between these two rooms during this time period. All menus were cleaned with isopropyl alcohol prior to start of our study to standardize cleanliness levels between the two distribution treatments.
The first part of our study determined the effectiveness of current cleaning procedures in the restaurant. Menus were collected for sampling at two time periods (1:30 p.m. and 4:30 p.m.). Those times were specifically chosen to determine how dirty the menus became after the lunch rush (the 1:30 collection time) and how well the menus were cleaned by the staff during their routine cleaning (prior to the 4:30 collection time) and before the evening meal service period. Menus were collected without the staff being alerted as to the reason why they were being collected. A sample of menus was collected for three days in each of two weeks (six data collections) for this portion of our study. For the first week, 50 menus were sampled (15 menus on the first day, 15 menus the second day, and 20 menus the third day) and 55 menus were sampled the second week (20 menus for the first day, 15 menus the second day, and 20 menus the third day) for a total of 105 menus.
To determine the impact of the spray vs. wet cloth cleaning method, a total of 120 menus were selected for sampling (30 menus in each of the two weeks for each of the two cleaning methods). The second variable tested the effect of cleaning method (wiping vs. spraying) and sought 60 menus for sampling (15 menus for each of the two weeks for each cleaning method). For the third variable, the effect of handing out vs. storing menus on the table, a total of 40 menus were selected for sampling. These included 10 menus from each of the two rooms (one room had servers hand out the menus and the other room kept menus on the table) in each of the two weeks.
ATP sampling on menus was conducted in the following manner. Menus were partitioned into squares measuring 10 cm x 10 cm (100 [cm.sup.2]) as recommend by Moore and Griffith (2002). The areas used for sampling were those determined in the pretest for multipage menus. Based on the results of the pretest, a 100-[cm.sup.2] area in the lower right hand side of the front of the menu was swabbed for testing. After swabbing, menus were cleaned with 91% isopropyl alcohol before they were handed back to the restaurant for customer use.
The second part of our study examined the effect of two cleaning methods. The first cleaning method involved spraying the menus with a Lysol brand product then wiping with a clean towel. Each menu was sprayed once per cleaning. To evaluate the amount of spray used, the volume of Lysol was measured. Volumes used for the two weeks were fairly similar with a total of 25 mL of Lysol used to clean the 15 menus in the first week and a total of 22 mL used for the menus in the second week. After spraying, the menus were wiped with a towel with anti-microbial protection. Each towel was used to clean a maximum of four menus before it was then discarded to minimize the effect of any carryover from one menu to another.
The other cleaning method involved wiping the menus with a damp towel. The towels were impregnated with 200 parts per million U.S. Environmental Protection Agency-approved sanitizing solution of quaternary ammonium. To release the sanitizer from the towel (as given in the manufacturer's directions) they were immersed in one gallon of water at approximately 75[degrees]F (24[degrees]C) then squeezed 10 times. One towel was used to clean all 10 menus in that week. To minimize the effect of carryover of contamination from one menu to another, the towel was placed back into the bucket after four menus were wiped down and then squeezed to remove excess liquid. After the cleaning process, menus were air dried for 20-30 minutes and then swabbed with the ATP meter. All menus were cleaned with 91% isopropyl alcohol before they were given back to the restaurant for use by customers.
Lastly, the effect of menu storage and handling was assessed. This part of our study hypothesized that menus handed out by the staff would be cleaner than menus displayed on tables since customers could easily touch the menus on the tables while they were eating and food debris could contaminate the menus. One room in the restaurant had servers hand out menus that were stored at the host station. The second room left the menus in a holder at the table. This portion of our study differed slightly in its collection method in that the menus were sampled prior to staff cleaning to assess the differences in contamination associated with these two storage and handling methods. Of the 45 menus typically used in each room, 10 menus were randomly collected for sampling from each room at approximately 4:00 p.m. (prior to cleaning) for each of the two weeks. The menus were swabbed with ATP meters using the method described above and cleaned with 91% isopropyl alcohol before they were given back to the restaurant.
Results and Discussion
Because of changes in menu availability and handling in the restaurant and one dropped sample during testing, the actual number of menus sampled changed slightly for the first test, but were still considered adequate for analysis. For the first test (before and after restaurant cleaning), 53 menus were sampled prior to cleaning and 58 menus after cleaning (Table 1). The number of samples used for the second test (60 menus) and third test (40 menus) were the intended numbers described in the methodology.
A t-test was used to compare the ATP values of the menus after lunch (and before they were cleaned) at 1:30 p.m. to those after they were cleaned (and before the supper period) at 4:30 p.m. by the staff. As expected, menus had higher ATP values prior to being cleaned (M = 306.20, SD = 687.42) as compared to menus that had been cleaned (M = 125.36, SD = 125.90). This difference was statistically significant: t(103) = 1.92, p < .05. The average of RLU values (306.20) for the uncleaned menus did not reach the critical upper limit of 500 RLU; however, approximately 11% or 6 out of the 53 usable individual menu results were at values greater than 500 RLU (657; 700; 4,838; 752; 1,152; and 508). Interestingly, one cleaned menu even showed a critical ATP value > 500 RLU (546).
The assessment of cleaning methods using a t-test showed a statistically significant difference in cleanliness between the two methods: t(58) = 2.303, p = .025. Menus cleaned with the spray cleaning method (M = 81.23, SD = 46.97) gave significantly higher ATP values than menus cleaned by the wiping method (M = 52.97, SD = 48.12) and were therefore considered less clean. None of the menus had values higher than 500 RLU and all of the menus were in an acceptable range of 0 to 249 RLU (Table 2). These results suggest that the use of a wet wipe to clean menus may be more effective, but that the spray method using a dry cloth was also able to adequately clean the menus according the ATP test.
Results of the third study showed that menus stored on the table (M = 370.35, SD = 243.70) gave statistically significant higher RLU values than menus that had been handed out (M = 96.80, SD = 47.07; t = 4.93, p < .001), indicating that menus stored on the table were dirtier than menus handed out. None of the individual menus that were handed out exceeded 500 RLU; however, 30% or 6 out of the 20 menus left on the table had levels greater than 500 RLU (576; 1,035; 533; 746; 529; and 563; Table 3). As predicted, menus stored on the tables may have had more chances for contamination by customers as food was being eaten.
Our study showed that menus should be cleaned after each shift or approximately every four hours. Skipping even one cleaning might result in a high level of contamination increasing the potential risk of cross contamination. If the menus are not clean, contamination could easily be transferred from the menus to the customers' hands or to the staff's hands when they serve food to the customers. Therefore, restaurant staff need to pay close attention to cleaning the menus with the appropriate cleaning methods after each shift.
Better cleaning was associated with the wet wiping cloths, although both methods achieved adequate cleaning scores according to the ATP results. In addition, our study found that menus stored on the table had greater levels of contamination than menus that had been handed out. Restaurant managers should consider handing out menus rather than leaving menus on the table. Menus stored on the table appear to have more chances for contamination from food or the hands of customers. If a restaurant's policy is to put menus on the table, the staff may need to clean the menus more frequently.
Previous studies have assessed consumers' perceptions of risk when purchasing food (Mitchell, 1998; Mitchell & Harris, 2005) and when selecting a restaurant (Henson et al., 2006; Knight et al., 2007). Future studies may wish to consider the perception of consumers about clean or unclean menus and whether an unclean menu affects their revisit intention. Although our study did not measure consumers' perception of the importance of clean menus, it is possible that clean menus may be an indicator of restaurant cleanliness to consumers in the same way that odors (Fatimah, Boo, Sambasivan, & Salleh, 2011) and workers' behavior and appearance (Fatimah et al., 2011; Henson et al., 2006) have been found to be important cues to the cleanliness of the restaurant.
Our study measured contamination on the surfaces of a multipage restaurant menu. One limitation of our study is that only one type of menu was assessed. Future studies may wish to evaluate the differences with other types of menus and other variables. For example, laminated menus, disposable menus (including children's menus), and menus that fit in leather-type holders may offer different results. In addition, the length of time that menus are used prior to disposal may make a difference. Finally, the types of foods offered on the menu may make a difference. Foods that are eaten with fingers or hands (particularly when menus are kept at the tables) may offer different results when menus are handled by customers.
It would appear that cleaning can reduce contamination if it is done correctly. Better cleaning and lower levels of contamination may be possible with certain cleaning methods such as the sanitizer towel tested in our study Differences in the level of contamination were also found with different service methods (leaving the menu on the table vs. the server handing it to the customers). One of the most significant findings of our study is that critical levels of contamination (as assessed by ATP values) may occur if regular cleaning is not done. High levels were found after one service period. Cleaning may not have been a priority in the past for restaurants, but should be considered in the future based on the results of our study. Regular menu cleaning can make a difference in the cleanliness of menus.
Corresponding Author: Jinkyung Choi, Assistant Professor, Department of Foodservice Management, Woosong University, Daejeon, South Korea 300-718.
Aycicek, H., Oguz, U., & Karci, K. (2006). Comparison of results of ATP bioluminescence and traditional hygiene swabbing methods for the determination of surface cleanliness at a hospital kitchen. International Journal of Hygiene and Environmental Health, 209(2), 203-206.
Fatimah, U.Z.A.U., Boo, H.C., Sambasivan, M., & Salleh, R. (2011). Foodservice hygiene factors--the consumer perspective. International Journal of Hospitality Management, 30(1), 38-45.
Food Safety Agency. (2008). Scores on the doors. Retrieved from http://www.food.gov.uk/multimedia/pdfs/board/fsa080306.pdf
Griffith, C.J., Blucher, A., Fieri, J., & Fielding, L.M. (1994). An evaluation of luminometry as a technique in food microbiology and a comparison of six commercially available luminometers. Food Science and Technology Today, 8(4), 209-216.
Griffith, C.J., Cooper, R.A., Gilmore, J., Davis, C., & Lewis, M. (2000). An evaluation of hospital cleaning regimes and standards. Journal of Hospital Infection, 45(1), 19-28.
Henson, S., Majowicz, S., Masakure, O., Sockett, P., Jones, A., Hart, R., Carr, D., & Knowles, L. (2006). Consumer assessment of the safety of restaurants: The role of inspection notices and other information cues. Journal of Food Safety, 26(4), 275-301.
Knight, A.J., Worosz, M.R., & Todd, E.C.D. (2007). Serving food safely: Consumer perception of food safety at restaurants. International Journal of Contemporary Hospitality Management, 19(6), 476-484.
Lee, J., Almanza, B.A., Nelson, D.C., & Ghiselli, R.F. (2009). Using health inspection scores to assess risk in food service. Journal of Environmental Health, 71(7), 28-33.
Leon, M.B., & Albrecht, J.A. (2007). Comparison of adenosine triphosphate (ATP) bioluminescence and aerobic plate counts (APC) on plastic cutting boards. Journal of Foodservice, 18(4), 145-152.
Mitchell, VW. (1998). A role for consumer risk perceptions in grocery retailing. British Food Journal, 100(4), 171-183.
Mitchell, VW., & Harris, G. (2005). The importance of consumers' perceived risk in retail strategy. European Journal of Marketing, 39(7-8), 821-837.
Moore, G., & Griffith, C. (2002). A comparison of traditional and recently developed methods for monitoring surface hygiene within the food industry: An industry trial. International Journal of Environmental Health Research, 12(4), 317-329.
Sirsat, S., Choi, J.K., Almanza, B.A., & Neal, J.A. (2013). Persistence of Salmonella and E. coli on the surface of restaurant menus. Journal of Environmental Health, 75(7), 8-14.
Worsfold, D., & Griffith, C.J. (1996). An assessment of cleanliness in domestic kitchens. Hygiene and Nutrition in Foodservice and Catering, 1, 163-173.
Jinkyung Choi, PhD
Department of Foodservice Management Woosong University, South Korea
Barbara Almanza, PhD
Douglas Nelson, PhD
School of Hospitality and Tourism Management Purdue University
Jay Neal, PhD
Sujata Sirsat, PhD
Conrad N. Hilton College of Hotel and Restaurant Management University of Houston
TABLE 1 The Effect on Adenosine Triphosphate Values (Relative Light Units) for Test 1 (Menu Cleanliness) * Before Cleaning After Cleaning 163 546 174 151 Dropped 283 217 431 367 403 218 440 295 329 208 220 121 306 208 85 157 203 752 109 1152 147 508 342 188 246 267 142 310 120 84 148 108 115 249 163 346 43 39 48 249 35 16 95 657 29 67 39 9 66 103 14 82 48 83 41 58 39 255 36 59 27 87 16 118 39 76 21 282 32 700 24 75 26 95 30 59 60 107 69 4838 50 152 82 169 24 121 72 222 91 66 67 57 26 150 74 81 113 * Relative light unit values greater than 500 indicate an unacceptable cleanliness level. TABLE 2 The Effect on Adenosine Triphosphate Values (Relative Light Units) for Test 2 (Type of Cleaning Method) * Spray Cleaning Damp Towel 26 60 117 69 69 50 125 82 83 24 77 72 70 91 120 67 115 26 151 74 63 113 32 249 117 66 77 106 119 69 7 16 26 12 41 61 36 64 76 16 82 21 207 23 28 32 93 0 77 10 69 4 12 32 116 32 47 32 159 16 * Relative light unit values greater than 500 indicate an unacceptable cleanliness level. TABLE 3 The Effect on Adenosine Triphosphate Values (Relative Light Units) for Test 3 (Menu Distribution) * Menu Distribution Hand Out Left on Table 47 130 27 47 121 176 16 576 82 102 67 1035 58 298 118 434 89 533 134 252 74 185 155 746 133 326 185 471 102 529 173 359 62 224 56 209 103 212 134 563 * Relative light unit values greater than 500 indicate an unacceptable cleanliness level.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||ADVANCEMENT OF THE SCIENCE|
|Author:||Choi, Jinkyung; Almanza, Barbara; Nelson, Douglas; Neal, Jay; Sirsat, Sujata|
|Publication:||Journal of Environmental Health|
|Date:||Jun 1, 2014|
|Previous Article:||Incorporating community-based participatory research principles into environmental health research: challenges and lessons learned from a housing...|
|Next Article:||Voluntary approaches to solid waste management in small towns: a case study of community involvement in household hazardous waste recycling.|