Assessment of shellfish practices in licensed retail food establishments in response to increased Vibrio illnesses in a landlocked area.

Introduction

Vibrio bacteria are odorless and colorless, occur naturally in seawater worldwide (Oliver, Warner, & Cleland, 1983), and multiply rapidly when water temperatures increase (Cabrera-Garcia, Vazquez-Salinas, & Quinones-Ramirez, 2004; Elhadi, Radu, Chen, & Nishibuchi, 2004; Motes et al., 1998). Filter-feeding molluscan shellfish such as oysters, clams, and mussels can accumulate large numbers of these bacteria, but they have also been found occasionally in raw crustaceans and fin fish (Bean et al., 1998; Cabrera-Garcia et al., 2004). Among Vibrio cases reported to the Centers for Disease Control and Prevention's (CDC's) Cholera and Other Vibrio Illness Surveillance System, three-quarters of cases during 1997-2006 were a result of raw or undercooked foods and 25% were nonfoodborne (Dechet, Yu, Koram, & Painter, 2008). Non-food-associated Vibrio infections have occurred after exposure of wounds or broken skin to seawater (Howard 8z Bennett, 1993). Because many Vibrio patients in the United States become infected by consuming raw oysters, several post-harvest processing (PHP) methods, including low temperature pasteurization, high hydrostatic pressure, and individual quick freezing (IQF) have been developed to reduce the number of bacteria in raw oysters to undetectable levels (Interstate Shellfish Sanitation Conference, 2007). The most common PHP method, IQF, is a system of freezing half-shelled oysters in trays going through a nitrogen (or carbon dioxide) freezing tunnel, followed by ice glazing, before the oysters are packed and stored in freezers or sold to customers (Mississippi Department of Marine Resources, 2008).

Of the more than I1 species of Vibrio known to infect humans, the most common agents of non-cholera Vibrio gastroenteritis in the United States are Vibrio parahacmolyticus and Vibrio vulnificus (Centers for Disease Control and Prevention [CDC], 2009; Morris & Black, 1985). Symptoms of acute Vibrio parahaemolyticus infection typically begin 24 hours after exposure and can include watery diarrhea, abdominal cramps, nausea, vomiting, fever, and chills. Illness is usually self-limited and lasts three days (CDC, 2008a). Bloodstream infections and death are rare and occur more commonly in persons with weakened immune systems (CDC, 2008a). Vibrio vulnificus infections typically have an onset of 12 to 72 hours after exposure. Among healthy people, symptoms can include vomiting, diarrhea, and abdominal pain (CDC, 2008a; Tacket, Brenner, & Blake, 1984). In persons with underlying medical conditions, however, especially liver disease, Vibrio vulnificus can infect the bloodstream, causing a severe and life-threatening illness characterized by fever and chills, decreased blood pressure (septic shock), and blistering skin lesions (CDC, 2008d). Vibrio vulnificus bloodstream infections are fatal about 50% of the time (CDC, 2008c). Although the severity of symptoms depends on the species of Vibrio and the underlying medical conditions of the patient, overall hospitalization and death rates for Vibrio illnesses are higher than those for many other enteric pathogens (CDC, 2006).

CDC estimates that as many as 4,500 cases of Vibrio parahaemolyticus occur every year, of which an average of 215 culture-confirmed cases, 30 hospitalizations, and one to two deaths are reported (CDC, 2008b). The exact number of Vibrio vulnificus cases is unknown, but an average of 95 cases (half of which are culture confirmed), 85 hospitalizations, and 35 deaths are reported each year nationwide (CDC, 2008d). Approximately half of these Vibrio vulnificus cases are reported from the Gulf Coast region. Before January 2007, CDC relied on reports from five states in the Gulf Coast region and 10 states that participate in the Foodborne Diseases Active Surveillance Network (FoodNet) to monitor cases of Vibrio (CDC, 2009). In January 2007, all Vibrio infections became nationally notifiable. If a Vibrio case reports consuming seafood, it is the responsibility of the county and state public health departments to collect information for a Food and Drug Administration (FDA) traceback investigation of the source of the bacteria and its potential further risk to consumers using the CDC Cholera and Other Vibrio Illness Surveillance Report form (CDC, 2007).

The Tri-County Health Department (TCHD) serves Adams, Arapahoe, and Douglas counties, which are three of the seven counties that participate in FoodNet in the Denver metropolitan area of Colorado. In 2004, six non-cholera Vibrio cases were reported in TCHD's jurisdiction, increasing the TCHD incidence rate 2.8-fold to 0.52 per 100,000 population. Two TCHD Vibrio cases and three nearby metro area cases reported local consumption of raw shellfish, including one severe illness and one death, associated with consuming food in restaurants located in the TCHD jurisdiction. The shellfish traceback for these cases was complicated by poor restaurant practices such as not keeping the harvest information tags for the required 90 days, which prompted concerns about further risks to consumers from raw shellfish served or sold in local retail food establishments. In response to these concerns, TCHD conducted a survey of shellfish practices in retail food establishments in the Tri-County area.

Methods

TCHD developed a questionnaire to assess compliance with the Colorado Retail Food Establishment Rules and Regulations governing safe shellfish handling practices (Colorado Department of Public Health and Environment, 2007a) and to assess methods to improve safety, including serving post-harvest processed (PHP) oysters and providing consumer advisory notices. The questionnaire focused on raw oysters, the raw shellfish most commonly served in Colorado. The questionnaire was administered from January to March 2005 to licensed retail food establishments known to serve or sell raw shellfish within the Tri-County area. These establishments were identified by first excluding all retail food establishments known to sell only prepackaged foods or other specialty foods such as ice cream. A list of the remaining establishments was then generated and given to environmental health (EH) staff, who were assigned geographical areas to identify which establishments served or sold raw seafood. Establishments were divided into three categories based on characteristics of serving shellfish: restaurants, caterers, and markets. Information was collected from restaurants, caterers, and independent markets through on-site inspections and interviews conducted by TCHD EH staff. For three major grocery chains, questionnaires were self-administered by the seafood managers of individual stores after food safety managers attended a presentation given by TCHD. Self-administered questionnaires that were returned with less than half of answers completed were followed up with a phone call or site visit to try to collect missing information.

Results

Of 3,926 licensed retail food establishments in the Tri-County jurisdiction, 158 establishments were identified as highly likely to serve or sell raw seafood. Of these 158 establishments, 96 were confirmed to serve or sell raw shellfish. Together, the self-reported averages from these 96 establishments added up to 7,496 raw oysters served or sold per week. Restaurants made up 34 (36%) of these establishments and accounted for most of the oysters sold. The median sales volume for an individual restaurant was three dozen raw oysters per week with peak serving days on Friday and Saturday. Fifteen (44%) of the restaurants purchased whole oysters and shucked them in their facility; the other 19 (56%) restaurants purchased pre-shucked oysters. The five caterers meeting the survey criteria sold raw oysters upon request only, usually for holidays and special occasions. The remaining 57 (59%) establishments were chain grocery markets and independent markets, most of which sold raw, unshucked oysters in small quantities to individual consumers.

Twenty (59%) restaurants, one (20%) caterer, and 41 (72%) markets surveyed did not offer any post-harvest processed (PHP) oysters (Table 1). Overall, only 17 (18%) of the 96 establishments claimed to serve PHP IQF oysters, which can reduce the risk of Vibrio infection. Only eight establishments, however, could be verified to have received them from suppliers approved to use PHP safety claims on their product (Interstate Shellfish Sanitation Conference, 2007). Use of PHP oysters was higher among restaurants (41%) and caterers (40%) than markets (2%). Two caterers (40%) and fifteen markets (26%) did not have any product on hand at the time of the survey and did not have PHP information in their records.

Shellfish handling violations of one or more sections of the Colorado Retail Food Establishment Rules and Regulations (Colorado Department of Public Health and Environment, 2007a) were identified in 38 (43%) of the 89 establishments found to serve raw shellfish who provided information for the survey (Table 2). Self-administered questionnaires from chain markets underreported commingling violations that were recorded during follow-up visits. Overall, the most frequent violations included commingling of shellfish (22%), failure to keep shellfish tags for 90 days (10%), failure to keep original containers (10%), and cross-contamination (8%).

In addition to these violations, our survey found that nine (10%) of the establishments had no method for organizing shellfish tags by date of use, making it difficult or impossible to determine the date that a specific batch of oysters was served or sold to consumers. Also, 18 (32%) of the 57 markets were receiving oysters in small quantities repackaged by the secondary shipper in a manner that, although technically in compliance with the regulations, omitted harvest information that would be necessary for traceback investigations. We inquired whether establishments would be interested in providing advisories to consumers via several different methods. Interestingly, 83 (86%) of the 96 establishments who responded were willing to provide voluntary consumer advisory notices about eating raw shellfish (Table 3).

Conclusion

TCHD's assessment of shellfish handling practices identified widespread deficiencies. The four most common deficiencies were commingling of shellfish, cross-contamination, failure to keep original containers, and failure to keep shellfish tags for 90 days. Three of the four most common violations found at retail food establishments greatly increase the difficulty of tracing back the source of infection to a particular harvest site, thus increasing the chances that additional contaminated product may be sold to consumers. The deficiencies identified during the assessment are likely not unique to Tri-Country's retail food establishments; however, we did not assess establishments outside our jurisdiction and did not find data from other jurisdictions in our literature review.

Health departments can take several actions to help reduce the risk of Vibrio illnesses. First, it may be possible to increase the focus on shellfish handling practices and education of retail food operators during routine EH inspections of retail food establishments that serve raw shellfish. Similar to many inland health departments, TCHD had little previous experience with Vibrio investigations, so we trained our EH staff in shellfish investigation and traceback procedures, and then they educated operators. Second, operators can be encouraged to purchase PHP oysters, in which Vibrio bacteria have been reduced to nondetectable levels. Although PHP oysters are readily available, this assessment revealed that most establishments were not using this risk-reduced product. The main reasons cited for not serving PHP oysters were lack of awareness of the product and increase in cost for the establishment. The small price increase per PHP oyster and small change in texture are accompanied by a significant increase in safety.

Third, retail food establishment operators and consumers can be educated about the disease risks of eating raw shellfish and provided with educational materials in several languages. Tri-County provided education for operators during our inspections and for consumers through press releases and a fact sheet on the risk of eating raw oysters or clams, which is available on our Web site. Our educational materials for consumers provide information on medical conditions that can increase risk and severity of Vibrio illness, and instructions for destroying Vibrio by various cooking methods (Tri-County Health Department, 2005). Although it was not done during this response, education campaigns that target those at risk for Vibrio vulnificus infection, such as immune-compromised persons and health care providers that treat such patients, may also be advantageous. Another form of education is the placement of consumer advisory notices, which a majority of our establishments are willing to make available for their customers even though they are not required in Colorado.

Finally, state and local health departments can improve their ability to monitor the incidence of Vibrio illnesses. Prior to this assessment, Vibrio illnesses in Colorado were only reported and investigated by the seven county health departments that participate in FoodNet. Vibrio has since been added to Colorado's list of reportable conditions for laboratories and Vibrio illnesses are now investigated statewide as of January 2006 (Colorado Department of Public Health and Environment, 2007b). With the increased importation and consumption of specialty seafood such as raw oysters, inland environmental health departments need to improve their vigilance and knowledge of these potentially hazardous foods. We hope that the combined efforts we suggest can help minimize the risk of Vibrio illness to consumers' health and retail food establishment operators' liability.

Acknowledgments: The authors thank the environmental health specialists of Tri-County Health Department who completed inspections and questionnaires for the shellfish practices assessment. The authors are also grateful to Alicia Cronquist, Susan Parachini, and Clark Wilson of the Colorado Department of Public Health and Environment, and Mario Seminara and Charlotte Wilkins of the United States Food and Drug Administration for their assistance and guidance of the Vibrio illness and shellfish traceback investigations.

Authors' Note: The results of this study have been presented at the following professional meetings:

* International Conference on Emerging Infectious Diseases, Atlanta, GA, March 22, 2006: Assessment of Shellfish Practices in Retail Food Establishments in Response to Increased Vibrio Illnesses in a Landlocked Area, Abstract #474.

* Colorado Environmental Health Association Annual Educational Conference, Estes Park, CO, September 23, 2005: Shellfish Practices in Tri-County Health Department Retail Food Establishments.

* Colorado Department of Public Health and Environment Advanced Food Safety Workshops, Golden, CO, March 10, 2005, and Eagle, CO, April 5, 2005: Vibrio Infections Associated with Eating Raw Oysters: Tri-County Health Department's Response.

REFERENCES

Bean, N.H., Maloney, E.K., Potter, M.E., Korazemo, P, Ray, B., Taylor, J.P, Seigler, S., & Snowden, J. (1998). Crayfish: A newly recognized vehicle for Vibrio infections. Epidemiology and Infection, 121(2), 269-273.

Cabrera-Garcia, M.E., Vazquez-Salinas, C., & Quinones-Ramirez, E.I. (2004). Serologic and molecular characterization of Vibrio parahaemolyticus strains isolated from seawater and fish products of the Gulf of Mexico. Applied and Environmental Microbiology, 70(11), 6401-6406.

Centers for Disease Control and Prevention. (2006). FoodNet surveillance report for 2004 (Final Report). Retrieved June 30, 2006, from http://www.cdc.gov/foodnet/reports.htm

Centers for Disease Control and Prevention. (2007). Cholera and other Vibrio illness surveillance report. Retrieved May 15, 2009, from http://www.cdc.gov/nationalsurveillance/PDFs/CDC5279COVISvibriosis.pdf

Centers for Disease Control and Prevention. (2008a). Vibrio parahaemolyticus: General information. Retrieved May 15, 2009, from http://www.cdc.gov/nczved/dfbmd/discase listing/vibriop_gi.html

Centers for Disease Control and Prevention. (2008b). Vibrio parahaemolyticus: Technical information. Retrieved May 15, 2009, from http://www.cdc.gov/nczved/dfbmd/disease_listing/vibriop_ti.html

Centers for Disease Control and Prevention. (2008c). Vibrio vulnificus: General information. Retrieved May 15, 2009, from http:// www.cdc.gov/nczved/dfbmd/disease_listing/vibriov_gi.html

Centers for Disease Control and Prevention. (2008d). Vibrio vulnificus: Technical information. Retrieved May 15, 2009, from http:// www.cdc.gov/nczved/dfbmd/disease_listing/vibriov_ti.html

Centers for Disease Control and Prevention. (2009). Summary of human Vibrio cases reported to CDC, 2007. Retrieved May 15, 2009, from http:// www.cdc.gov/nationalsurveillance/cholera_vibrio_surveillance.html

Colorado Department of Public Health and Environment. (2007a). Colorado Retail Food Establishment Rules and Regulations: 3-102 Fish and Shellfish C.ER. [section] 3 Food.

Colorado Department of Public Health and Environment. (2007b). Conditions reportable by all laboratories collecting specimens or performing tests in Colorado. Retrieved June 21, 2007, from http:// www.cdphe.state.co.us/dc/Lablist.pdf

Dechet, A.M., Yu, PA., Koram, N., & Painter, J. (2008). Nonfoodborne vibrio infections: An important cause of morbidity and mortality in the United States, 1997-2006. Clinical Infectious Diseases, 46, 970-976.

Elhadi, N., Radu, S., Chen, C.H., & Nishibuchi, M. (2004). Prevalence of potentially pathogenic Vibrio species in the seafood marketed in Malaysia. Journal of Food Protection, 67(7), 1469-1475.

Howard, R.J., & Bennett, N.T. (1993). Infections caused by halophilic marine Vibrio bacteria. Annals of Surgery, 217(5), 525-531.

Interstate Shellfish Sanitation Conference. (2007). Post harvest processing. RetrievedJune 21, 2007, fromhttp://www.issc.org/php.aspx

Mississippi Department of Marine Resources. (2008). Post-harvest oyster processing technologies. Retrieved August 3, 2008, from http://safeoysters.org/industry/factsheet.pdf

Morris, J.G., Jr., & Black, R.E. (1985). Cholera and other vibrioses in the United States. New England journal of Medicine, 312(6), 343-350.

Motes, M.L., Depaola, A., Cook, D.W, Veazey J.E., Hunsucker, J. C., Garthright, WE., Blodgett, R.J., & Chirtel, S.J. (1998). Influence of water temperature and salinity on Vibrio vulnificus in Northern Gulf and Atlantic Coast oysters (Crassostrea virginica). Applied and Environmental Microbiology, 64(4), 1459-1465.

Oliver, J.D., Warner, R.A., & Cleland, D.R. (1983). Distribution of Vibrio vulnificus and other lactose-fermenting vibrios in the marine environment. Applied and Environmental Microbiology, 45(3), 985-998.

Tacket, C.O., Brenner, E, & Blake, PA. (1984). Clinical features and an epidemiological study of Vibrio vulnificus infections. Journal of Infectious Diseases, 149(4), 558-561.

Tri-County Health Department. (2005). The risk of eating raw oysters or clams. Retrieved May 15, 2009, from http://www.tchd.org/disease.html

U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition. (2009). Foodborne pathogenic microorganisms and natural toxins handbook. Retrieved May 15, 2009, from http:// www.foodsafety.gov/-mow/intro.html

Laura DeGolier, MPH, REHS 

Katya Ledin, MPH, PhD

George Vargulich, MA, REHS

Jennifer L. Patnaik, MHS 

Bruce Wilson, MPA 

Richard L Vogt, MD

Corresponding Author: Laura DeGolier, Environmental Health Manager, Tri-County Health Department, 7100 E. Belleview Avenue, Suite 102, Greenwood Village, CO 80111. E-mail: [e-mail].

TABLE 1
Number of Retail Food Establishments (RFEs) that Report Serving
Post-Harvest (Pre-Consumer) Processed Oysters by Retail Food
Establishment Type

Type of Oysters Served            Restaurants   Caterers
                                   (n = 34)     (n = 5)

Individually quick frozen or       14 (41%)     2 (40%)
  other post-harvest processing
No post-harvest processing         20 (59%)     1  (20%)
Unknown to RFE operator             0 (0%)      2 (40%)

Type of Oysters Served             Markets    All RFEs
                                  (n = 57)    (N = 96)

Individually quick frozen or       1 (2%)     17 (18%)
  other post-harvest processing
No post-harvest processing        41  (72%)   62 (64%)
Unknown to RFE operator           15 (26%)    17 (18%)

TABLE 2
Deficiencies Found by Retail Food Establishment (RFE) Type

Deficiency Found                Restaurants      Caterers
                                  (n = 34)     (n = 4) (a)

No tags                           5 (15%)        1 (25%)
No original container             5 (15%)         0 (0%)
No source information              2 (6%)         0 (0%)
Commingling                        2 (6%)         0 (0%)
Temperature not <41[degrees]F      0 (0%)         0 (0%)
Cross-contamination               6 (18%)        1  (25%)
ANY deficiency                    14 (41%)       2 (50%)

Deficiency Found                  Markets        All RFEs
                                (n = 51) (b)   (N = 89) (c)

No tags                            3 (6%)        9 (10%)
No original container              4 (8%)        9 (10%)
No source information              3 (6%)         5 (6%)
Commingling                       18 (35%)       20 (22%)
Temperature not <41[degrees]F      2 (4%)         2 (2%)
Cross-contamination                0 (0%)         7 (8%)
ANY deficiency                    22 (43%)       38 (43%)

(a) One catering establishment not included because no oysters
were present during survey.

(b) Six markets not included because information was not provided
on self-administered survey.

(c) Total RFE number reduced by seven establishments as noted
above.

TABLE 3
Preferred Consumer Advisory Method by Retail Food Establishment
(RFE) Type

Advisory Method       Restaurants    Caterers
                       (n = 34)       (n = 5)

Willing to advise      29 (85%)       4 (80%)
Menu label             16 (47%)       0 (0%)
Table tent              9 (26%)       0 (0%)
Posted notice          14 (41%)       0 (0%)
Handout                 6 (18%)       3 (60%)
No method indicated     1 (3%)        1 (20%)

Advisory Method         Markets      All RFEs
                       (n = 57)      (N = 96)

Willing to advise      50 (88%)      83 (86%)
Menu label              0 (0%)       16 (17%)
Table tent             19 (33%)      28 (29%)
Posted notice          24 (42%)      38 (40%)
Handout                22 (39%)      31 (32%)
No method indicated    25 (44%)      27 (28%)

Note. Establishments were allowed to choose more than one method.