* A Foodborne Outbreak of Gastroenteritis in a Teaching Hospital 07/16/96 A Foodborne Outbreak of Gastroenteritis in a Teaching Hospital Developing Family Medicine: Challenges for the Next Decade AKUH, February, 22, 2003 Rashid A. Chotani, MD, MPH Assistant Professor & Director Global Infectious Disease Surveillance & Alert System (GIDSAS) Center for International Emergence, Disaster & Refugee Studies Johns Hopkins Schools od Medicine & Public Health Phone: 410-614-8330 E-mail: rchotani@jhsph.edu Rashid A. Chotani, MD, MPH 11/22/2018 *

Hospital 940 bed hospital with 4 cafeterias. * 07/16/96 Hospital 940 bed hospital with 4 cafeterias. Cafeteria A, located in the OPD, serves approximately 600 visitors and employees daily. This outbreak occurred at one of the cafeterias located in the out patient department which serves approximately 600 visitors and employees daily. We will call this cafeteria ‘cafeteria A’. To give you an idea we are dealing with a 940 bed hospital with 4 cafeterias and a few restaurants. On December 9, 1997 individuals who had eaten at cafeteria A reported nausea and projectile vomiting after eating a noon meal. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

* 07/16/96 Events On 12/9/97 individuals who ate at Cafeteria A reported nausea and projectile vomiting after eating a noon meal. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Method: case definition * 07/16/96 Method: case definition Any person who ate lunch prepared at cafeteria A on December 9, 1997 and developed sudden onset of vomiting or diarrhea or abdominal cramps and Any of the following symptoms including nausea, fever, body aches, headache, chills or fatigue. The first thing we did was to come up with a case definition. A case was defined as any person who ate lunch prepared at cafeteria A on December 9, 1997 and developed sudden onset of vomiting or diarrhea, or abdominal cramps and any of the following symptoms including nausea, fever, body aches, headache, chills or fatigue. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Method: case finding We identified all cases who identified * 07/16/96 Method: case finding We identified all cases who identified individuals who ate with them. All non-Ill persons were used as controls. Additional cases were found when we contacted: Nurse managers Hospital managers Directors of nursing, functional unit directors, JHH vice-presidents Several employee groups were notified via e-mail and asked to identify cases. We identified all cases who identified individuals who ate with them. All non-ill persons were used as controls. Additional cases were found when we contacted nurse managers, hospital managers, directors of nursing, functional unit directors, and JHH vice-presidents We also notified several employee groups via e-mail and asked them to identify cases. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Methods: questionnaires * 07/16/96 Methods: questionnaires Standard questionnaires were used to obtain medical and food history from Food service workers Ill and non-ill employees Menu reviewed at Cafeteria A. Standard questionnaires were used to obtain medical and food history from food service workers and ill and non-ill employees. The menu of Cafeteria A was reviewed for the day in question. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Control Measures and Interventions: Cafeteria A * 07/16/96 Control Measures and Interventions: Cafeteria A Cafeteria: Kitchen was inspected Leftover foods recovered and cultured Food preparation, kitchen cleaning procedures reviewed Certain food items quarantined Employees were interviewed, inspected for sores, boils, cuts, IV tract marks and sent to occupational health services We obtained nares swabs 3 cafeteria staff members submitted stool samples or rectal swabs All staff (n=17) were questioned daily for symptoms Simultaneously, as we were conducting the case investigation we had to put forth control measures and interventions at cafeteria A. The kitchen was inspected, leftover food was recovered and cultured. Food preparation and kitchen cleaning procedures were reviewed. Certain food items were quarantined. Employees were interviewed and were inspected for sores, boils, cuts, IV tract marks and were sent to occupational health services. Nares swabs were obtained on all the cafeteria workers. 3 cafeteria staff members who reported either cold symptoms or diarrhea or changes in bowel habits submitted stool samples or rectal swabs. All 17 staff members were questioned daily for symptoms. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Salmonella, Shigella, Aeromonas, Campylobacter, Yersinia * 07/16/96 Laboratory Food was processed in standard fashion Blood agar plate R/O Bacillus CAN plate R/O Staphylococcus Plates for enteric pathogens R/O Salmonella, Shigella, Aeromonas, Campylobacter, Yersinia Samples sent to city, state, FDA, and commercial laboratory Sequencing preformed Heavy metal testing In the lab, food was processed in standard fashion. Blood agar plate was used to rule out Bacillus; CAN plate was used to rule out Staphylococcus; and plates for enteric pathogens were used to rule out Salmonella, Shigella, Aeromonas, Campylobacter, and Yersinia. Samples were immediately sent to the city and state health departments, FDA, and commercial laboratory. 16 S ribosomal RNA sequencing preformed at our lab and heavy metal testing was done at the private commercial lab. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Results N = 75 (ill = 40; non-ill = 35) * 07/16/96 Results N = 75 (ill = 40; non-ill = 35) Mean age: 39 years (range 25-56) Sex: 85% female Results N was 75 with 40 ill and 35 non-ill individuals. The mean age was 39 years. The range being 25-56 years. 85% of the study population was female. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Incubation period: JHH employees (n=40) * 07/16/96 Incubation period: JHH employees (n=40) 12 Point source outbreak 10 Number of Cases Reported 8 6 4 The incubation period was between 15 minutes to 19 hours, the mean being 4 hours. If we excluded the one outlier, the incubation period was between 15 minutes to 8 hours. Thus, the disease was of a short incubation period. 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Incubation Period (Hours) *

Symptoms Percent fever body aches chills headache diarrhea fatigue * 07/16/96 Symptoms 18 fever 25 body aches 43 chills 48 headache 63 diarrhea fatigue 63 68 abdominal cramps vomiting The predominant symptom were nausea (93%) and vomiting 78%; followed by abdominal cramps (68%); fatigue and diarrhea (63%) each; headache (48%); chills (43%); body aches (25%); and fever (18%) 78 93 nausea 20 40 60 80 100 Percent *

* 07/16/96 Results: outcomes Duration of illness: mean--24 hr. (range <24 - 72 hr.) Bedridden 62.5% Sought medical care 27.5% Hospitalized 2.5% The duration of illness ranged between less than 24 hours to 72 hours. Mean being 24 hours. 62.5% of the ill were bedridden, 27.5% sought medical care and 2.5% were hospitalized. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Results: univariate analysis * 07/16/96 Results: univariate analysis Food item OR СI 95% p-value Green beans 36.4 6.93,341.60 <0.0001 Tortellini 5.50 1.03, 54.50 0.02 Corn soup 0.23 0.04, 1.10 0.03 Veg soup 0.23 0.04, 1.10 0.03 Not associated: bread, breadsticks, chicken salad, broccoli salad, cheese salad, caesar salad, havarti cheese, swiss cheese, beef stew, Thai beef, couscous, honey turkey, chicken fingers, cheese pizza, sausage pizza, chow mein noodles, marinated tomatoes, onions, mixed greens, cucumbers, dressing tomato-bacon/peppercorn, creamy, sunflower seeds, crackers, chips, cookies, yogurt, and fresh fruits The results of the univariate analysis indicated that the odds ratio for green beans was 36.4 which was statistically significant, with a rather wide 95% confidence interval. Tortellini had an odds ratio of 5.5 which was not statistically significant and corn soup and vegetable soup had a protective affect. None of the other foods served at Cafeteria A were statistically associated. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Results: multivariate analysis * 07/16/96 Results: multivariate analysis Variable OR CI 95% p-value Green beans 1.84 1.49,2.27 <0.005 Tortellini 1.25 0.98,1.59 NS Corn soup 0.89 0.69,1.16 NS Veg soup 0.93 0.71,1.21 NS The multivariate analysis demonstrated clearly the green beans were the potential problem with an odds ratio of 1.84 which was statistically significant. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

* 07/16/96 Results: laboratory Bacillus sp. recovered from garlic mix (opened/unopened jars), Moroccan beef stew and vegetable soup. Bacillus subtilis was identified based on the library profiles. Heavy metals negative. Patient/employees culture negative. In the lab we recovered Bacillus species from opened and unopened jars of garlic mix, beef stew, and vegetable soup. GLC preformed at our lab identified bacillus subtilis based on the library profiles. The 16 S ribosomal RNA sequencing confirmed these results. No viable staph aureus were recovered. The FDA lab reported no staph aureus and a minute quantity of bacillus cereus that would not cause the toxin mediated illness we saw. The commercial lab assayed for staph enterotoxin and bacillus cereus diarrheal toxin, which were both negative. They also reported no heavy metal toxins. There currently exists no practical test for bacillus cereus emetic toxin, so this could not be tested even though nausea and vomiting were the predominant symptoms. All patient (ill) and employee nares and stool cultures or rectal swabs were negative. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Food preparation Frozen green beans steamed for 10 minutes. * 07/16/96 Food preparation Frozen green beans steamed for 10 minutes. Seasoned with salt, pepper, olive oil and garlic mix. Baked in oven for 15 minutes at 375 OF. Stored in warmer at 180 degrees. Placed in pan and sent to serving line (140 degrees); maximum time 4 hours. Here we feel that it is important to review the food preparation of green beans. The frozen green beans were first steamed for 10 minutes, then seasoned with salt, pepper, olive oil and garlic mix. This was followed by 15 minutes of baking at 375 degrees Fahrenheit. The green beans were then stored in a warmer at 180 degrees, placed in a pan and sent to the serving line (140 degrees), where they were served rather quickly, as most of the lunch items had a side order of green beans. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Summary 40 persons became ill after eating green beans. * 07/16/96 Summary 40 persons became ill after eating green beans. We under-estimated magnitude of problem because case ascertainment difficult. The symptoms pointed to a toxin mediated illness. The process of preparing green beans with garlic (in soy oil base) most likely led to the illness. Bacillus was isolated from opened/unopened jars. In summary 40 people became ill after eating green beans. We feel that the magnitude of the problem was under-estimated because of the difficulty in case ascertainment. The symptoms pointed to a toxin-mediated illness. The process of prepating green beans with garlic (in a soy bean base) most likely lead to the illness and bacillus was isolated from opened and unopened jars. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Conclusions FDA Inspected the manufacturing facility * 07/16/96 Conclusions FDA Inspected the manufacturing facility Inspected food supplier Ordered recall of all garlic jars produced by company A Mandated new control protocols Aggressive control measures should be taken to prevent the spread of any outbreak particularly in a hospitals in order to protect not just the patients but the staff. Rarely bacillus subtilis has been associated in food poisoning. In conclusion I would like to say that aggressive control measures should be taken to prevent the spread of any outbreak in a hospital in order to protect not just the patients but the staff. And rarely bacillus subtilis has been associated in food poisoning. 11/22/2018 Rashid A. Chotani, MD, MPH GIDSAS-JHU *

Thank you 11/22/2018 * 07/16/96 * Rashid A. Chotani, MD, MPH GIDSAS-JHU *