Food Protection Trends - July 2010 - 414

of this study, health inspection reports from the Kansas Department of Health and Environment (KDHE) were posted on the Internet in the public domain for anyone to review. KDHE reports include detailed information on violations of Kansas food codes, which help researchers identify and target food code violations. The Web site provides information about the different types of health inspections conducted for a one-year period. The types include routine, licensure, customer complaint-driven, and follow-up re-inspection completed after the restaurant has received poor performance ratings on a routine or complaint inspection. The KDHE is mandated to inspect all establishments once annually, but more inspections are required to ensure that an establishment has become compliant with the food code. In September 2008, the Kansas Department of Agriculture assumed the role of restaurant inspectors in the state, and they too make inspection results available on the Internet. As a proxy measure for food handling behaviors in independent ethnic restaurants, restaurant inspection reports were analyzed from a readily available, government internet site. The purpose of this research project was to identify the food safety training needs of independent ethnic restaurants. Specific objectives were to identify the frequencies and types of food code violations and to compare and contrast food safety inspection reports of ethnic and non-ethnic independent restaurants.

were eliminated. Of the 2,097 remaining, 541 were identified as ethnic and 1,556 as non-ethnic restaurants. Although there were about three times as many non-ethnic restaurants as ethnic restaurants, equal numbers as restaurants (250 ethnic and 250 non-ethnic) were randomly selected as the study sample to assure a balanced sample for statistical analysis. Because publicly available data were used without contacting human subjects, no approval from the institutional review board was necessary.

Variables and data collection
For the 500 restaurants selected, health inspection results were reviewed online (http://kansas.kdhe.state.ks.uspls/ certop/fssearch). Using the KDHE inspection guide, a data collection form was created so that research assistants could easily record necessary information from the online report. Code numbers of 275 Kansas food codes were listed on the data collection form so that the number of violations of each code could be entered. Spaces were provided for recording the numbers of critical and non-critical violations, the date of the health inspection with the greatest number of critical violations between September 1, 2007 and August 31, 2008, the total number of inspections during the 12-month period, and the inspection type (e.g., routine inspection, complaint-driven inspection, re-inspection after poor performance on previous inspections, etc.). For each randomly-selected facility, the one inspection report that indicated the most violations was selected from the multiple inspection reports during the specified period. This decision was made because reviewing the most recent inspection may result in selecting re-inspection reports, which are usually better than the preceding report. During October and November 2008, two research assistants reviewed the KDHE Web site and completed the data collection form. After the data had been recorded, the information was cross-checked by a different research assistant to ensure accuracy of data entry. Data were then entered into a Microsoft Access database, cross-checked to verify correct data entry, and converted to SPSS Version 15.0 for statistical analyses (28).

and standard deviations of continuous variables (i.e., number of violations and inspections), were calculated. Pearson χ2 analyses were conducted to evaluate differences between numbers of independent ethnic and non-ethnic restaurant establishments where individual food code violations were cited. Independent sample t-tests were conducted to compare differences in mean numbers of critical, non-critical, and within-category violations. Individual food codes were grouped based on KDHE categories. KDHE categories were further combined to reduce the number of variables. For example, KDHE had separate categories for toilet facilities, hot and cold water availability, sewage and waste water disposal, and garbage and refuse disposal. Because these were all related to maintenance of physical facilities, we created the category “Physical Facility Maintenance” by combining them. The number of violations within each category was calculated using the “compute” function of SPSS for further analyses. Independent t-tests and ANOVA with Scheffe’s post hoc analyses were conducted to evaluate differences in number of violations in different categories between ethnic and non-ethnic independent restaurants and among different categories of ethnic restaurants. Statistical significance was established at P < 0.05.

RESULTS
Of all inspection reports included (n = 500), 360 (72.0%) were reports of routine inspections, and 79 (15.8%) were triggered by customer complaints. There was no statistical difference in types of inspections between ethnic and non-ethnic independent restaurants.

MATERIALS AND METHODS
Study sample
Five hundred randomly selected independent restaurant inspection reports (for 250 ethnic and 250 non-ethnic restaurants) in 14 Kansas counties with the highest numbers of Hispanic and Asian populations were selected as the study sample. The sample was drawn from the list of foodservice establishments, including both commercial and non-commercial establishments, that were subject to KDHE health inspections. From the list of foodservice establishments in the 14 counties, all chain restaurants, non-commercial foodservices (i.e., schools, churches, and community centers), and convenience stores

Numbers of critical and non-critical violations and inspections
Independent restaurants were cited for an average of 3.71 ± 2.90 critical and 2.28 ± 2.40 non-critical food code violations on average per inspection. Independent t-tests revealed significant differences in the number of critical and non-critical inspections between ethnic and non-ethnic restaurants. Ethnic restaurants had more critical (4.52 ± 2.97) and non-critical violations (2.84

Statistical analyses
Descriptive statistics, consisting of frequencies, cross-tabulations, means,

414 FOOD PROTECTION TRENDS | JULY 2010

http://kansas.kdhe.state.ks.uspls/certop/fssearch http://kansas.kdhe.state.ks.uspls/certop/fssearch

Food Protection Trends - July 2010

Table of Contents for the Digital Edition of Food Protection Trends - July 2010