by Cynthia B. Morrow, MD, MPH
First thing Monday morning, the communicable disease nurse was successful in reaching the patient. When told of his positive results, he declared that he wasn’t surprised and noted that he went out with friends to a dining establishment the prior week. He shared that several of them were also sick. The communicable disease nurse immediately became interested. She asked for more details about the friends, about the restaurant, and of course, about the symptoms. She went upstairs to talk to her colleagues in Food Protection to share the news. When she got there, she found that one of the sanitarians was on the phone, talking to a different individual who just called in to file a complaint about a restaurant. The caller claimed that he had gone out the previous week and that he and several of his friends were also ill. The sanitarian wrote the name of the restaurant and showed it to the communicable disease nurse who nodded. “We’ve got more…” the sanitarian whispered to her friend. She too obtained contact information from the caller including contact information for his ill friends. As soon as she got off the phone, the colleagues compared notes. The symptoms and time frames being reported were consistent with Campylobacter infection. They called in the supervisor for Food Protection: “We think we may have a problem.”
They explained to the supervisor that they had one confirmed case of Campylobacter jejuni and two clusters of similar gastrointestinal illness that were epidemiologically linked to an event that had occurred at a popular venue, Hinerwadel’s Grove, the previous week. The establishment, a family owned business known for its festive clambakes, provided customers with the experience of enjoying its famous clams and salt potatoes while playing horseshoe, softball, corn hole, volleyball, and other games on its 34-acre property. In addition to catering large outdoor parties, Hinerwadel’s had indoor seating and a retail seafood store on the premise. The Bureau of Food Protection immediately went into action. Two of the department’s sanitarians were sent to inspect Hinerwadel’s and an emergency meeting between communicable disease and food protection staff was called to coordinate the interviews with the all of the newly identified suspected cases.
The sanitarians arrived at the establishment by noon. The proprietors were appropriately concerned and offered to help the investigation in any way that they could. Campylobacter, which causes about 1.3 million illnesses a year in the United States, is typically associated with raw or under cooked poultry. While Hinerwadel’s was best known for its raw, fried, and steamed clams and its salt potatoes, it also served cooked mussels, shrimp, chowder, hamburgers, hot dogs, barbecue beef and pork, chicken sandwiches, and many dessert items. The sanitarians did a thorough inspection, noting that there were separate stations for preparing the chicken, the seafood, the hamburgers and hot dogs, as well as the ready-to-eat food such as salads and desserts. The inspectors found that stations were physically separated, in different areas of the establishment. They noted that each station had its own sinks, preparation surfaces, knives, and so forth. The environmental health staff felt confident that the risk of cross-contamination was extremely low. During the inspection, several food items, including all available bags of clams, were confiscated. After a thorough inspection, the staff concluded that there was no smoking gun to explain a cluster of suspected cases of Campylobacter.
Meanwhile, back at the health department, Dr. Cynthia Morrow, then the Commissioner of Health, had been updated about the situation, as had the regional office of the New York State Department of Health. By Monday afternoon, there was an additional confirmed case of Campylobacter jejuni and several other suspected cases. It became very clear that the confirmed and suspected cases had all attended an event sponsored by the local hospital’s union. The health department issued a press release to inform the public about the incident and ask those who attended the specific clambake to call in. The OCHD’s Communicable Disease and Food Protection staff utilized the hypothesis-generating questionnaire to interview the clambake attendees to build a detailed line list. To the health department’s staff’s surprise, chicken was not a commonly implicated food in the interviews; however, clams were. The fact that all of the cases had consumed clams wasn’t particularly surprising given that clams were what the establishment was famous for, but it didn’t make a great deal of sense because clams were not known to be associated with this particular bacterium. Were the clams the actual source of the infections, or was another food?
The health department team took two approaches to address this question. The first was to try to establish a biologic connection to explain the outbreak. The second was to establish an epidemiologic association using a case-control study design. With respect to the former approach, the environmental health staff took two dozen clams from the confiscated bags of clams. The clams were sent to the New York State Department of Environmental Conservation and to the New York State Health Department for testing. With respect to the latter approach, first a more detailed questionnaire that included questions about every food/ beverage item on the menu was developed and implemented; next, a case definition that included symptoms, date of onset, and exposure to food from the venue was established; and finally, interviews were conducted for individuals who met the case definition as well as for those who ate at the venue but who did not become ill.
The investigation took an interesting twist when, during the interviewing process, one of the sanitarians learned that a medical resident from the local hospital had experienced diarrhea after the event and, out of curiosity, took his own stool sample to the lab. Vibrio parahaemolyticus was identified from his stool sample. Vibrio parahaemolyticus is frequently isolated from a variety of raw seafoods, particularly shellfish, and can cause human illness including watery diarrhea if ingested. There are about 80,000 cases of vibriosis reported in the United States every year, most commonly in warmer months when these naturally occurring bacteria are in higher concentration in coastal salt water.
The following Tuesday morning, Dr. Morrow had her annual budget presentation to the local legislature. Before starting her presentation, she informed legislators that the health department had an active investigation of foodborne illness and that there would be a press conference to try to encourage both “cases” and “controls” to contact the health department to get to the bottom of the mystery. The legislators, riveted by an ongoing investigation, voiced their support and expedited the budget hearing to allow the health department team to return to the investigation. By the end of the day, almost 40 people, including some people from neighboring counties, had called to report that they were ill after eating at the venue. The New York State Department of Health (NYSDOH) coordinated the interviews of out-of-county residents who were impacted by the outbreak. As an example, in one situation, a resident from another county had purchased a bag of clams from the venue’s retail store and held a clambake at his home. Several people who attended that private clambake became ill.
Over the next few days, as the health department appealed to the public to help with the investigation, hundreds of phone calls came in. Despite the flood of calls, the OCHD had the capacity to manage all of the incoming data, in part because of an experience with a large foodborne outbreak two years earlier. Reflecting on that outbreak during a press conference about the Hinerwadel’s investigation, Dr. Morrow stated, “The good news is that we’re really prepared to handle this. We’re able to process the information much more quickly because we learned a lot…” The department had developed a system to ensure that all data were entered into an electronic database that was tailored to the outbreak specific hypothesis-generating questionnaire. This allowed for immediate data entry as the interviews were being conducted. With this approach, which was coordinated with the NYSDOH, information from all of the interviews was entered into a database in real-time and odds ratios were calculated as the numbers grew. By the end of the investigation, data were available for 258 cases (7 of whom were hospitalized, 28 of whom had confirmed Campylobacter jejuni infection, and 1 of whom had Vibrio parahaemolyticus) and 328 controls.
The numbers told a clear story. With a final attack rate of 57.7% and an odds ratio of 3.73 (confidence interval 2.67-5.20) raw Mahogany clams were epidemiologically identified as the source of this foodborne outbreak. People who became ill either ate clams at the venue or ate clams purchased from the retail store between August 20 and September 18, 2008. There appeared to be bimodal curves for both the incubation period (one peak at 4-12 hours and a second at 40-48 hours) and the duration of symptoms (one peak at less than 48 hours and another peak at 73-180 hours), consistent with the hypothesis of a dual pathogen outbreak. Furthermore, Vibrio parahaemolyticus, V. alginolyticus and Campylobacter jejuni were isolated from the samples of clams that were collected during the investigation.
Now the mystery deepened, were the Mahogany clams the original source of these bacteria or had they been contaminated somewhere along the way?
The Division of Environmental Health did extraordinary work tracing the clams’ path. They traced (and cleared) the trucks that transported the clams to the establishment. They tracked down the clam supplier, and in cooperation with the New York State Department of Environmental Health, the New York State Department of Environmental Conservation, and the Maine Department of Health, they learned how the Mahogany clams were harvested, stored in a bay, and washed before shipping. Interestingly, the clam supplier had an exclusive relationship with the venue: the supplier’s only customer was this venue and this venue only sold Mahogany clams from this particular supplier. The team also learned that the clams were harvested a few miles off the coast of Maine and then went into “wet bay” for storage. Finally, they learned that the clams were rinsed with water from a private well prior to be transported by truck to the clambake venue. Given jurisdictional rules, the OCHD was not directly involved in the investigation of the clam supplier, the private well, or any other part of the investigation in Maine. While Vibrio parahaemolyticus and Vibrio alginolyticus are commonly associated with shellfish, Campylobacter is not. Conclusive evidence of where contamination, if any, occurred was not provided to the OCHD. Perhaps the shellfish was contaminated with Campylobacter by the droppings of seabirds; perhaps they were contaminated when in the “wet bay”; or perhaps when they washed with water from a private well. Unfortunately, we will never know. Interestingly, two years later, in October 2010 a second, smaller, outbreak of Campylobacter jejuni was associated with consumption of raw Mahogany clams at the same establishment. In October 2018, for unrelated reasons, the venue announced that it was closing after being in business for over 100 years.
Acknowledgement: The author wishes to thank the Onondaga County Health Department for its assistance in providing information and for its review of this case.
- This case repeatedly uses the term “outbreak.” What are the criteria for appropriately using this term?
- The OCHD considered this most likely to be a dual pathogen outbreak based on the appearance of bimodal epidemiology curves for both incubation period and duration of symptoms consistent with vibriosis and camplyobacteriosis. On the other hand, the NYSDOH considered this to most likely to be a single pathogen outbreak based on the human specimen laboratory testing showing only one positive culture of Vibrio parahaemolyticus compared to 28 positive cultures for Campylobacter jejuni. What are your thoughts about the different interpretation of the cause of this outbreak?
- While the Centers for Disease Control and Prevention was notified of this outbreak, it was not actively involved in the investigation. Why do you think that is? When do you think the CDC should get involved?
- Bivalve mollusks (eg, clams and oysters) are the most frequently implicated seafood in seafood-associated foodborne outbreaks. During the outbreak, Dr. Morrow strongly advised the public not to eat raw shellfish, resulting in push back from the industry. Do you think that having warnings about the risks of consuming raw seafood on restaurant menus goes far enough?
Cynthia B. Morrow, MD, MPH, is an assistant professor at Virginia Tech Carilion School of Medicine. Previously, she was the Lerner Chair for Health Promotion at Syracuse University. Her prior academic positions include Professor of Practice in the Department of Public Administration and International Affairs at Syracuse University and an assistant professor with the Center for Bioethics and Humanities as well as with the Department of Public Health and Preventive Medicine at Upstate Medical University. Dr. Morrow served as Commissioner of Health for Onondaga County during which time she earned numerous awards from community-based organizations. She is a consulting editor for the Journal of Public Health Management and Practice and is also an editor of two books, including Public Health Administration: Principles for Population-Based Management and JPHMP’s 21 Public Health Case Studies on Policy & Administration. She is a graduate of Swarthmore College (BA) and Tufts University School of Medicine (MD, MPH).
- Bad Air, Bad Blood: A Homeric Admonition by John S. Marr, MD, MPH, and Robert M. Maulitz, MD
- It’s All in the Bottle by by Asim A. Jani, MD, MPH
- The World’s Deadliest Poison by John S. Marr, MD, MPH, and Marcus A. Horwitz, MD
- Mystery in the Pines by Gus Birkhead, MD, MPH, published in the Journal of Public Health Management and Practice
- Introducing Backstories in Epidemiology: True Medical Mysteries
- Introducing a New Series of Epidemiologic Case Studies: Podcast with Dr. John Marr
- Appendectomy Masquerade, an epidemic initially attributed to appendicitis in upstate New York
- The Babies Are Dying, newborn deaths in a West Virginia rural hospital nursery
- Of Bites and Men: The Most Dangerous Urban Animal, a story where the number of human bites exceeds shark bites