32 Volume 76 • Number 5
A D V A N C E M E N T O F T H E PRACTICEA D V A N C E M E N T O F T H E PRACTICE
Introduction State environmental health professionals are critical members of the public health work- force, accounting for 10% of public health professionals (Center for State and Local Government Excellence, 2008). Ensuring a safe food supply is a key responsibility of environmental health programs, particu- larly in state and local health departments. According to a 2006 survey conducted by the
Association of State and Territorial Health Offi cials, 82% of state environmental health units administer food protection programs (Association of State and Territorial Health Offi cials, 2007). In Louisiana, the Center for Environmental Health in the Offi ce of Public Health oversees the food safety program and ensures safety for almost 32,000 food estab- lishments, including full service restaurants; fast food businesses; cafeterias in hospitals,
prisons, nursing homes, and schools; conces- sion stands; delis; bars; and grocery stores. Among other functions, the Louisiana food safety program conducts food establishment inspections and responds to consumer com- plaints and reports of foodborne illness.
Food safety inspections serve as one of the most fundamental public health activities intended to prevent foodborne illness from occurring, but no defi nitive evidence exists indicating that poor restaurant inspection results accurately predict the potential for food- borne illness. The Seattle-King County (Wash- ington) Health Department performed a study in 1987, which found that “restaurants with poor inspection results were at increased risk of foodborne outbreaks (Irwin, Ballard, Grendon, & Kobayashi, 1989),” and that routine inspec- tions by the health department could help identify restaurants with an increased risk of an outbreak. Florida Department of Health epi- demiologists conducted a similar study, how- ever, and found no correlation between lower inspection results and increased outbreak risk (Cruz, Katz, & Suarez, 2001).
The literature is in confl ict about how often food establishments should be inspected and whether less frequent inspections result in poorer inspection outcomes. Allwood and co-authors found that restaurant inspec- tion scores were signifi cantly higher (better) when restaurants were inspected four times a year, compared to three times a year (All- wood, Lee, & Borden-Glass, 1991). In con-
Lindsey Realmuto, MPH Environmental Health Section
San Francisco Department of Public Health
Katherine L. Hunting, MPH, PhD Rebecca Parkin, MPH, PhD
School of Public Health and Health Services
The George Washington University
A b s t r a c t The study described in this article evaluated the effects of public health workforce cuts on routine food safety inspections
and the occurrence of critical violations. Routine inspection information was
collected from two Louisiana databases for permanent food establishments
categorized as risk category 3 or 4 in East Baton Rouge Parish, Louisiana,
for the years 2005, 2007, and 2009. The length of time between routine
inspections nearly quadrupled from 2005 to 2009. For risk category 4
establishments, a significant increase occurred in the proportion of
inspections that resulted in a critical violation between the three years. The
amount of time between routine inspections was signifi cantly higher for
inspections that resulted in a critical violation versus those that did not.
Lastly, the amount of time between routine inspections, an establishment’s
risk category, and history of complaint were found to have signifi cant
predictive effects on the incidence of a critical violation during a routine
inspection, although results varied by year. Study results indicate that
decreased workforce capacity in Louisiana may negatively affect the
outcomes of routine food safety inspections.
State Health Agency Workforce Shortages and Implications for Public Health: A Case Study of Restaurant Inspections in Louisiana
4 tables, 0 fi gures
JEH12.13_PRINT.indd 32 10/31/13 11:21 AM
December 2013 • Journal of Environmental Health 33
A D V A N C E M E N T O F T H E PRACTICE
trast, a 2008 Canadian study did not find a statistical difference between food safety outcome measures based on inspection fre- quency, but concluded that “food prem- ise inspections should continue to play an important role in protecting the public from foodborne illnesses by educating workers (Newbold, McKeary, & Hart, 2008).”
The El Paso County (Colorado) health department has lacked staff resources to per- form two inspections per year of restaurants and other food providers as required by law (Auge, 2009). A report on the health care infrastructure in Colorado Springs (in El Paso County) revealed complaints about restau- rant cleanliness and employee hygiene had increased from 60 in 2005 to 219 in 2007 (Emery, 2008). Although the county had not seen an increase in major foodborne out- breaks, individual complaints of sickness from food poisoning increased nearly threefold from 2005 to 2006 (Limbert & Beard, 2008). Lim- bert and Beard’s report suggests that decreased restaurant inspections due to workforce short- ages may present increased public health risks.
The Food and Drug Administration (FDA) recommends that to sustain an effective risk- based food safety program, state agencies should maintain well-resourced programs; i.e., at least one-full time staff member devoted to every 280–320 food establishments (FDA, 2009). In Louisiana, the ratio of sanitarians to the number of expected food establishments in need of inspection is already much higher than FDA’s recommendation. For example, in the East Baton Rouge area in 2010, approxi- mately 2,775 food establishments were open but only seven sanitarians were employed, cre- ating a ratio of 396 retail food establishments to be inspected per sanitarian. Thus, these san- itarians are expected to perform nearly 25% more inspections than FDA recommends for an effective program. Additionally, food safety inspections constitute only one aspect of the environmental health sanitarian’s overall job duties. Sanitarian inspection responsibilities cover a wide range of program areas, includ- ing beach monitoring, building and premises, commercial seafood, disease/vector control, food and drug, infectious waste, milk and dairy control, mollusks and shellfish, onsite wastewater, and retail food establishments.
Nationwide, the existing environmental health workforce shortage (Perlino, 2006) and even more severe budgetary constraints
on environmental health programs (Gurwitt, 2009) threaten food safety programs. In Loui- siana, budget cuts have directly affected the state’s ability to maintain core environmental health staff. For example, in fiscal year 2010– 2011, the environmental health department was required to reduce its budget by 25% to meet state budgetary constraints. To do this, the department removed six environmen- tal health sanitarian staff from payroll, thus reducing the capacity to perform public health inspections and reducing response time by the department for other public health issues. These cuts exacerbated the potential impact of prior year workforce reductions. As state environmental health programs are forced to defend and justify their programs in the face of increased budgetary constraints, they must demonstrate the value of environmen- tal health services. The purpose of our study was to evaluate Louisiana’s electronic food inspection database to assess potential trends in inspection results and inspection frequency over time, as Louisiana Department of Health and Hospitals has faced workforce shortages.
Methods Our study compared food safety inspec- tions among 2005, 2007, and 2009 in East Baton Rouge to identify potential predic- tors of critical violations. A critical violation as defined by the FDA 2001 Food Code is a violation “that, if in noncompliance, is more likely than other violations to contribute to food contamination, illness, or environmen- tal health hazard (FDA, 2004a).” Inspection records of establishments in risk categories 3 and 4 (RC3 and RC4) were utilized because they represent sites at most risk for food- borne outbreaks, with RC4 higher than that of RC3 establishments. Risk category desig- nations are generally defined by the types of food served, required preparation steps, vol- ume of food served, population served, and compliance history (FDA, 2004b). Examples of RC3 and RC4 establishments include full service restaurants that have extensive menus and handle raw ingredients, as well as restau- rants that involve the cooking, cooling, and reheating of potentially hazardous food. A food establishment may also be rated as an RC3 or RC4 if the primary service population may be at increased risk of foodborne illness, such as schools and nursing facilities. The incidence of critical violations was used as a
surrogate to indicate greater potential risk of foodborne disease.
Two Louisiana Department of Health and Hospitals data systems provided information. The Automated Inspection Records System is used by sanitarians in the field to record inspection results. The Sanitarian Event Track- ing System is used by sanitarians to record establishment location, risk category, permit status, etc. These two systems are linked elec- tronically and help inform sanitarians when they need to perform routine inspections.
Variables compiled in an Excel 2007 data- base were restaurant permit number, date of routine inspection, and previous routine inspection; the number of days between the date of routine inspection and previous routine inspection; number of critical viola- tions received; number of noncritical viola- tions received; history of complaint since last routine inspection; risk category; and year of inspection. A “complaint” occurs when a member of the public contacts the depart- ment regarding an issue with a food facility and the environmental health department is obligated to follow up on the complaint by performing an inspection at the facility to ver- ify the validity of the complaint and enforce remediation if necessary. Up to six critical violations per inspection were abstracted.
Statistical Analysis SAS 9.1 was used for statistical analysis. Descriptive analyses determined the num- ber of inspections performed each year and within each risk category; the mean number of violations cited during an inspection; the frequencies of inspections resulting in criti- cal violations, noncritical violations, and no violations; the frequency of inspections with a history of complaint; and the average days between routine inspections for estab- lishments of different risk categories and between years. Chi-square tests determined if trends over time and between risk catego- ries were statistically significant. Analysis of variance compared the average days between routine inspections from year to year. One- sided t-tests determined whether the aver- age days between inspections resulting in critical violations versus ones that did not were statistically different. One-sided t-tests also determined if the average days between inspections with a history of complaint were statistically different from those without such
JEH12.13_PRINT.indd 33 10/31/13 11:21 AM
34 Volume 76 • Number 5
A D V A N C E M E N T O F T H E PRACTICE
history. Logistic regression analysis evaluated whether days since last routine inspection, history of complaint, and risk category pre- dicted the incidence of critical violations.
Results A total of 3,488 (57.3%) inspection records were eligible for study among 6,090 inspec- tion records evaluated. A previous inspection date could not be found for 1,067 records, 670 records collected were risk category 1 or 2, and 865 records were missing risk category information. Routine inspection records were not found for January–April, 2005; thus, this analysis included only May–December 2005. Of all eligible inspections, 1,615 (46.3%) resulted in critical violations, 2,988 (85.7%) resulted in noncritical violations, and 445 (12.8%) resulted in no cited violations (Table 1). Forty-two different types of critical viola- tions were cited across the three years. The most commonly cited were related to equip- ment cleanliness (n = 610), toxic storage (n = 452), and employees eating or drinking in a food preparation area (n = 341).
A greater proportion of RC4 establish- ments had a routine inspection resulting in a critical violation. The number of critical viola- tions received by an establishment during an inspection ranged from 0 to 17 with a mean of 0.95, while the number of noncritical viola- tions ranged from 0 to 40 with a mean of 4.37. The total number of violations per inspection ranged from 0 to 48 and averaged 5.32. Chi- square tests for trend showed that the propor- tion of critical violations varied significantly (p < .0001) by year for all inspections as well as for RC4 establishments only (Table 1).
The average number of days between rou- tine inspections (regardless of risk category) was 252 days, ranging from 2 to 1,421 days. The annual means increased over time from about 100 to over 400 days for RC3, RC4, and all establishments combined. Variance F sta- tistics revealed that the time between inspec- tions increased significantly (p < .0001) from 2005 to 2009 for RC3, RC4, and all inspec- tion categories. Table 2 indicates that for RC4 establishments, the average number of days between routine inspections that resulted in critical violations was significantly greater than those that did not (p < .0001); RC3 results were not statistically significant.
Chi-square analyses of an establishment’s history of complaint, stratified by risk category
and year, demonstrated statistically signifi- cant (p < .0001) positive trends in complaints for RC3, RC4, and all inspection categories.
Chi-square tests (Cochran-Mantel-Haenszel option) supported (p < .0001) the hypotheses that a history of complaint would increase the
Violation Details by Year and Risk Category
Risk Category Year Chi-Square (p-Value)
2005 2007 2009
# of Inspections (% of Total Within Year) Risk category 3
No violations cited 45 (13) 49 (8) 77 (13) Only noncritical violations cited
167 (48) 340 (53) 250 (44)
Critical violation cited 136 (39) 254 (40) 245 (43) 1.51 (.22) Total 348 643 572
Risk category 4
No violations 85 (17) 107 (14) 82 (13) Only noncritical violations cited
188 (38) 285 (37) 198 (30)
Critical violation cited 225 (45) 380 (51) 375 (57) 17.2 (<.0001)
Total 498 772 655 All inspections
No violations cited 130 (15) 156 (11) 159 (13) Only noncritical violations cited
355 (42) 625 (44) 448 (37)
Critical violation cited 361 (43) 634 (45) 620 (51) 13.5 (<.0001)
Total 846 1415 1227
TABLE 1
Number of Days Between Inspections That Resulted in Critical Violations Versus Those That Did Not, for Each Risk Category and All Inspections
Risk Category Resulted in a Critical Violation
# Mean # of Days Between Inspections
(95% CI a)
t-Test Statistic (p-Value)*
Risk category 3 No 928 251.9 (238.5–265.3)
-0.92 (.36)
Yes 635 261.8 (245.4–278.1)
Risk category 4 No 945 225.9 (213.8–237.9)
-5.08 (<.0001)
Yes 980 272.0 (258.9–285.2)
All inspections No 1873 229.7 (238.8–247.8)
-4.02 (<.0001)
Yes 1615 257.8 (268.0–278.2)
aCI = confidence interval. *For all inspections and risk category 4 establishments, the Satterthwaite t statistic was used. For risk category 3 establishments, the pooled t-test was used.
TABLE 2
JEH12.13_PRINT.indd 34 10/31/13 11:21 AM
December 2013 • Journal of Environmental Health 35
A D V A N C E M E N T O F T H E PRACTICE
odds of having an inspection that resulted in a critical violation in all (odds ratio [OR] = 1.88) and RC4 (OR = 4.09) establishment categories; RC3 results were nonsignificant. A t-test anal- ysis (Table 3) evaluated history of complaint, and shows that, regardless of risk category, the mean number of days between routine inspec- tions for establishments that had a complaint since their last routine inspection was sub- stantially and significantly (p < .0001) longer compared to those that did not.
A logistic regression analysis evaluated predictive factors for the occurrence of a critical violation during a routine inspection. The logistic model included risk category (4 vs. 3), complaint history (yes vs. no), and number of days between routine inspec- tions, and was run for all years combined and for each year separately. Table 4 displays the Wald confidence interval adjusted ORs. For all years, having a history of complaint (OR = 1.74) and being a RC4 establishment
(OR = 1.55) were strong predictors of the incidence of a critical violation, while the number of days between routine inspec- tions was a weaker but still-significant pre- dictor. Every additional day between routine inspections increased the odds of a critical violation by about 0.1%. In other words, if a routine inspection was supposed to occur every 100 days, waiting another 30 days to do the inspection increased the odds of a critical violation by about 3.5%.
When stratified by inspection year, the logistic model results varied substantially. The days between inspections demonstrated a slightly stronger effect in 2005 compared to all years (OR = 1.006). Risk category demon- strated the strongest predictive effect in 2009 (OR = 1.84). A history of complaint was the strongest predictive variable for inspections in 2007 (OR = 2.69). The 2005 analysis was markedly different in that a history of com- plaint showed no predictive effect (OR = 0.55). The Hosmer and Lemeshow goodness- of-fit test stratified by year revealed good fits for 2007 and 2009, but the odds ratios for 2005 and all years combined should be inter- preted with caution.
Discussion These results document several very impor- tant trends in food safety inspections in East Baton Rouge, Louisiana. From 2005 to 2009, the amount of time between routine inspec- tions almost quadrupled. In 2005, the average time between routine inspections for all estab- lishments was approximately three and a half months. By 2009, however, the average time between routine inspections had increased to over a year. This was most likely a result of decreased workforce capacity within the depart- ment from layoffs and hiring freezes caused by state budget cuts. The data also show only minor differences in the days between routine inspections for RC3 versus RC4 establishments.
RC4 establishments tend to have more haz- ardous food production processes than RC3 establishments and present greater risk of potential foodborne illness. Our study found that a larger proportion of RC4 establishment inspections resulted in critical violations and that risk category demonstrated the most consistent predictive effect on critical viola- tions. These findings support FDA’s rationale that RC4 establishments should be inspected more often than other establishments.
Mean Number of Days Between Inspections With a Complaint History Versus Those That Did Not, Stratified by Risk Category and for All Inspections
Risk Category History of Complaint
# Mean # of Days Between Inspections (95% CIa)
t-Test (p-Value)
Risk category 3 No 1442 239.3 (229.2–249.3) -9.1 (<.0001)Yes 121 462.9 (430.2–495.7)
Risk category 4 No 1816 236.0 (227.3–244.7) -9.67 (<.0001)Yes 109 472.3 (424.6–519.9)
All inspections No 3258 237.4 (230.8–244.0) -13.3 (<.0001)Yes 230 462.9 (430.2–495.7)
aCI = confidence interval.
Adjusted Odds Ratios (OR) and 95% Confidence Interval (CI) for Critical Violations—Logistic Regression Analysis for All Inspections and Stratified by Year
Year Variable (Effect) Adjusted OR Estimate*
95% CI Goodness-of-Fit Chi- Square (p-Value)
All inspections Date_difference 1.001 1.000 1.001 20.28 (.009) Risk category 1.545 1.349 1.77 Complaint 1.742 1.309 2.319
2005 (n = 846)
Date_difference 1.006 1.002 1.009 15.926 (.04) Risk category 1.30 0.983 1.723 Complaint 0.55 0.16 1.89
2007 (n = 1415)
Date_difference 0.99 0.998 1.000 7.407 (.493) Risk category 1.514 1.244 1.910 Complaint 2.688 1.644 4.394
2009 (n = 1227)
Date_difference 1.001 1.000 1.001 12.61 (.126) Risk category 1.835 1.46 2.307 Complaint 1.484 1.02 2.159
Note. Values in bold are statistically significant. *Wald confidence interval adjusted OR estimate.
TABLE 3
TABLE 4
JEH12.13_PRINT.indd 35 10/31/13 11:21 AM
36 Volume 76 • Number 5
A D V A N C E M E N T O F T H E PRACTICE
While the average days between inspec- tions increased between 2005 and 2009, a significant positive trend also occurred in the proportion of inspections resulting in critical violations. Statistically significant differences were observed in days between inspections that resulted in critical violations versus those that did not, with more time between inspec- tions resulting in critical violations. The logis- tic regression analysis demonstrated that time between routine inspections was a weak but significant factor in predicting the incidence of critical violations. Thus, the significant increase in the amount of time between inspections from 2005 to 2009 could explain the positive trend in the proportion of inspections resulting in critical violations. Routine inspections pro- vide excellent opportunities for sanitarians to educate restaurant staff and managers on food safety. The three most frequently cited critical violations—equipment cleanliness, toxic stor- age, and employees drinking and eating in a food preparation area—are all behaviors and circumstances that can be minimized with reg- ular food safety education.
The proportion of inspections with a his- tory of complaint in East Baton Rouge also increased significantly between 2005 and 2009. A significant increased difference occurred in the days between routine inspec- tions that had a history of complaint com- pared to those that did not. One possible explanation for this association is that with reduced resources, inspections conducted to investigate a complaint are being substituted for routine inspections. Based on the inspec- tion records studied, it appears that when investigating a complaint, sanitarians remark on the purpose of the complaint but also per- form the duties of a routine inspection and note any other violations. Thus, the inspec- tion may act as a routine inspection, pushing back the time frame for the next inspection. A history of complaint also proved to be a fairly
strong predictor, however, although not as consistent as risk category, for the incidence of a critical violation during routine inspec- tions. Therefore, while inspections for inves- tigating complaints may serve informally as routine inspections, Louisiana sanitarians should try to prioritize these establishments for more frequent routine inspections.
Our study is one of a few efforts to evaluate a state food safety inspection program and can provide a model for similar evaluations else- where. One strength of our study was its large sample size, including nearly 3,500 inspection records. Close to 2,000 inspection records were missing a risk category designation or a previous inspection date, however, and could not be included in the analysis. This could affect the ability of the sample to represent the entire picture in East Baton Rouge.
The unexpected gap in inspection data for 2005 could have affected the results of our study; this gap may have resulted from issues being resolved in the electronic system, which was brought online in 2003. No reason exists, however, to believe that results of inspections from January to April would have been very different from May through December. None- theless, the missing data could have intro- duced unexplained biases that could account for the differences in results between 2005 and the other two years.
Data entry errors in the electronic system could have led to information bias within our study. Further, personnel changes over time could have contributed to differences in inspection result trends. Each region in Loui- siana has at least one “standardization offi- cer,” who trains and tests sanitarians in the field to ensure consistency across all inspec- tions (C. Bombet, personal communication, August 18, 2010). For the timeframe of our study, all inspections would have been per- formed by an inspector who had gone through training with a standardization officer. There-
fore, an inspection by any sanitarian in Loui- siana should have similar results. Budget cuts in the past two years have forced the Center for Environmental Health to reduce the num- ber of standardization officers in its central office in Baton Rouge. In the future, it may become more difficult to ensure consistency in inspections among sanitarians without the capacity to do adequate training for sanitar- ians entering the field.
Conclusion Based on our study, decreased frequency of inspections could be responsible for an increas- ing proportion of routine inspections that result in critical violations, especially among RC4 establishments. Additionally, increased time between inspections from 2005 to 2009 was paralleled by a significant increase in com- plaints. We recommend that the Louisiana Department of Health and Hospitals prioritize the inspection of RC4 establishments and those establishments with a history of complaint to reduce the risk of foodborne illnesses.
Acknowledgements: We would like to thank the staff at the Center for Environmental Health, Office of Public Health, Louisiana Department of Health and Hospitals for the use of their inspection data and cooperation in this study. Lindsey Realmuto would also like to thank The George Washington School of Public Health and Health Services for the Capital Connection Fund grant that allowed her to travel to Louisiana for this study.
Corresponding Author: Lindsey Realmuto, Health Program Planner, San Francisco Department of Public Health, Environ- mental Health Section, 1390 Market Street, Suite 822, San Francisco, CA 94102. E-mail: Lindsey.realmuto@sfdph.org.
References on page 37
? The International Federation of Environmental Health (IFEH) will bring its 13th World Congress to the NEHA 2014 AEC. IFEH is an organization whose members are national associations representing the interests of environmental health professionals throughout the world. The 2014 AEC is a unique opportunity to meet environmental health professionals from all over the globe!
Did You Know?
JEH12.13_PRINT.indd 36 10/31/13 11:21 AM
December 2013 • Journal of Environmental Health 37
A D V A N C E M E N T O F T H E PRACTICE
Allwood, P., Lee, P., & Borden-Glass, P. (1991). The vital role of res- taurant health inspections. Journal of Environmental Health, 61(9), 25–28.
Association of State and Territorial Health Offi cials. (2007). Ensur- ing healthy communities: Results of the 2006 SEHD survey. Retrieved from http://www.astho.org/Display/AssetDisplay.aspx?id=2552
Auge, K. (2009, March 1). Home of TABOR in a pinch. The Denver Post. Retrieved from http://www.denverpost.com/newsheadlines/ ci_11810109
Center for State and Local Government Excellence. (2008). The impending shortage in the state and local public health workforce. Retrieved from http://slge.org/publications/the-impending- shortage-in-the-state-and-local-public-health-workforce
Cruz, M., Katz, D., & Suarez, J. (2001). An assessment of the ability of routine inspections to predict foodborne outbreaks in Miami-Dade County, Florida. American Journal of Public Health, 91(5), 821–823.
Emery, E. (2008, January 23). Staffi ng shortages slice restaurant inspections thin. The Denver Post. Retrieved from http://www. denverpost.com/news/ci_8049606
Food and Drug Administration. (2004a). FDA 2001 food code— chapter 1: Purpose and defi nitions. Retrieved from http://www.fda. gov/ForConsumers/ConsumerUpdates/ucm089124.htm
Food and Drug Administration. (2004b). FDA 2001 food code— chapter 8: Compliance and enforcement. Retrieved from http://www. fda.gov/ForConsumers/ConsumerUpdates/ucm089235.htm
Food and Drug Administration. (2009). FDA food code 2009: Annex 5—conducting risk-based inspections. Retrieved from http://www. fda.gov/Food/FoodSafety/RetailFoodProtection/FoodCode/ FoodCode2009/ucm187947.htm
Gurwitt, R. (2009). Bleak outlook for state fi nances. Retrieved from http://www.governing.com/article/bleak-outlook-state-fi nances
Irwin, K., Ballard, J., Grendon, K., & Kobayashi, J. (1989). Results of routine inspections can predict outbreaks of foodborne illness: The Seattle-King County experience. American Journal of Public Health, 79(5), 586–590.
Limbert, W.M., & Beard, S. (2008). Health care infrastructure in Colo- rado Springs. Colorado Springs, CO: Western Strategies Center.
Newbold, K.B., McKeary, M., & Hart, R. (2008). Restaurant inspec- tion frequency and food safety compliance. Journal of Environmen- tal Health, 71(4), 56–61.
Perlino, C. (2006). The public health workforce: Left unchecked, will we be protected? Retrieved from http://www.apha.org/NR/ rdonlyres/597828BF-9924-4B94-8821-135F665E9D45/0/Public HealthWorkforceIssueBrief.pdf
References continued from page 36
Rapid. Repeatable. Robust.
Bactiquant®-surface MycoMeter®-airMycoMeter®-surface Bactiquant®-water
The Mycometer system can give you results onsite in 30 minutes. • Mould on sur faces • Bacteria on sur faces • Mould inside porous materials • Bacteria in water • Mould par ticles in air
For questions or to order go to: mycometer.com
Call Lisa Rogers at (813) 831-6511
JEH12.13_PRINT.indd 37 10/31/13 11:21 AM
Copyright of Journal of Environmental Health is the property of National Environmental Health Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.