Case Study: Gastrointestinal Illness on a College Campus

FACILITATOR VERSION

Objectives

  • List three active case finding strategies for a gastrointestinal outbreak on a college campus
  • Create a case definition, given an outbreak scenario
  • Describe the purpose of initial interviews with case-patients
  • Identify the responsibilities of the environmental health specialist during a foodborne outbreak investigation

Instructions

Convene your local Epi Team, and provide each Epi Team member with a copy of the case study (participant version). Choose one person to act as the facilitator. This person should use the facilitator’s version of the case study.

Guidance for facilitating the case study can be found on the next page. The case study begins on page 3.

Time Allotted: 1.5 hours

Background Materials

The following trainings, found at the North CarolinaCenter for Public Health Preparedness WebSite ( are recommended for Epi Team members without prior outbreak investigation experience. They can be viewed prior to completing the case study.

  1. Embarking on an Outbreak Investigation (FOCUS on Field Epidemiology Volume 1, Issue 3)
  2. Designing Questionnaires(I is for Investigation Session III)
  3. Selecting a Study Design(FOCUS on Field Epidemiology Volume 2, Issue 4)
  4. Case-Control Studies for Outbreak Investigations (FOCUS on Field Epidemiology Volume 3, Issue 2)

Resources

Foodborne Disease Surveillance and Outbreak Investigation Toolkit, CDC

Guidance for Facilitators

Goal

The goal of working through a case study is active learning through engaged participation by each Epi Team member.

Role of Facilitator

As a facilitator, your job is to:

  • Guide the Epi Team through the case study
  • Involve every team member
  • Moderate discussion drawing on the suggested answers to discussion questions
  • Ensure key points are covered for each question
  • Keep an eye on the clock

Active Participation

All team members should be involved in the discussion. One strategy for getting everyone involved is to have team members take turns reading aloud and attempting to answer questions. Facilitators should:

  1. Ensure that the room set-up encourages group participation, ie. everyone seated around a table or chairs in a circle.
  2. Select one team member to read aloud the first update and the first question.
  3. Encourage that team member to attempt to answer the first question (regardless of his/her background).
  4. Encourage other team members to add information and discuss the question.
  5. Use the suggested answers to cover key point(s) that were not addressed during the discussion.
  6. Have the next person read aloud the next update or question and begin an attempted answer as above – move in sequence to include everyone.

Follow this sequence until the scenario is completed, pacing discussion in order to finish the entire scenario.

Facilitation Tips

  • Read the case study ahead of time so you are prepared for the discussion.
  • Always remember the learning objectives for the case study.
  • If one person is dominating the conversation, call on other team members.
  • If someone in the group is not participating, ask his or her opinion.
  • When someone asks a question, encourage other team members to provide the answer.
  • Do not spend too much time providing the suggested answers to the group. Instead, encourage the team to share their opinions about possible responses to the given scenario.

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

Group Brainstorm

Choose one member of the Epi Team to record responses on a flip chart. As a group, discuss the following question.

Question 1: What questions do you ask Sheila during this phone call?

Suggested answer: Ask questions in the following categories:

1)Contact information for the person making the report (Sheila);

2)Contact information for the ill persons;

3)Total number of people that have been ill and characteristics of the ill people (e.g. are they all male? All freshmen? Both staff and students?);

4)Usual number of cases of GI illness that the Student Health Services would typically expect at this time of year;

5)Specific symptoms;

6)Any diagnoses and treatment;

7)Recent events on campus where large groups of students were gathered; and

8)Any common factors or connections that Sheila has noticed among the cases.

Also remember to ask Sheila for her theory about what is going on. Since she is familiar with the students and the campus, she may think of a potential cause of the outbreak that you would not have considered.

Question 2: What diseases would you suspect at this point?

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

Suggested answer: At this early point, you would suspect virtually any gastrointestinal illness, including Salmonella, Shigella, norovirus, Campylobacter, Yersinia, E. coli O157, Bacillus cereus, and Staphylococcus aureus. You would contact the hospital to find out whether any students had positive laboratory results for any of these diseases.

Question 3: What, if any, clinical specimens should be collected from the ill students? Who is responsible for collecting the specimens? Where should the specimens be sent?

Suggested answer: Stool samples should be collected from the ill students and sent to a laboratory capable of testing for a wide range of gastrointestinal pathogens, including Salmonella, Shigella, norovirus, Campylobacter, Yersinia, E. coli O157, Bacillus cereus, and Staphylococcus aureus. The Student Health Service at the college may be able to collect stool samples from some of the ill students. Otherwise, the responsibility would lie with your local health department. Since a few of the students went to the hospital, you should also call the hospital to determine whether any laboratory tests have already been conducted. Here in North Carolina, the stool samples would likely be sent to the State Laboratory of Public Health, one of the regional laboratories in Asheville, Greenville, or Charlotte, or a private laboratory.If a stool sample is not available for every student, it may be possible to collect a vomitus sample.

Question 4: Of the diseases listed above, norovirus is the only one which is not routinely tested by the North Carolina State Laboratory of Public Health (SLPH) when stool samples are submitted. However, SLPH does have the capacity to test for norovirus. What steps should you take to ensure that the specimens are tested for norovirus?

Suggested Answer: First, you should call the Communicable Disease Control Branch at the NC Division of Public Health. There are two reasons to call: 1) to report the possible outbreak, and 2) to authorize testing for norovirus, since the local health department cannot authorize such testing. Next, you need to be sure that at least 5 specimens are collected, since SLPH will not test for norovirus with less than 5 specimens. Finally, be sure to write “outbreak” on the laboratory submittal slip to ensure that testing is rapid.

Question5: What specific actions will you take to determine whether there are other possible cases? Discuss which Epi Team members will be responsible for each case finding task.

Suggested answer: First you will create a working case definition based on the information provided by Sheila. Next, you should conduct active case finding, using one or more of the following strategies:

1)Call the hospital where the students were sent to determine whether there are additional cases from MadisonCollege;

2)Talk to the ill people and find out if they know other people who have gotten ill;

3)Call local laboratories to find out if they have had an increase in testing requests for gastrointestinal (GI) pathogens;

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

4)Contact other healthcare providers in the area to determine if they have seen increased number of people with GI illness;

5)Send an email to the entire campus of MadisonCollege (including students, faculty, and staff) asking them to visit Student Health Services if they have gastrointestinal symptoms. (This email would also be a good place to provide some infection control guidance);

6)Set up a hotline for people to call if they have symptoms of illness;

7)Send resident assistants to visit all dormitory residents to ensure that very ill students are not isolated in their rooms, and consider ways to do the same for students living in sorority and fraternity houses on campus; and

8)Call other local colleges to notify them of the outbreak and find out whether they have observed similar cases.

The Epi Team member who is responsible for each case finding activity will differ by local health department. You may want a single Epi Team member to be a liaison between MadisonCollege and your health department.

Activity

As a team, decide upon an initial (working) case definition. Keep in mind that your case definition will change as you obtain more information. A case definition should always include clinical information, and elements related to person, place, and time.

Clinical Information (e.g. disease signs and symptoms, lab results)
Person
Place
Time

Suggested answer: At the beginning of an outbreak, it is usually best to create a broad case definition that will capture all possible cases.

One suggested case definition is included below:

Clinical Information (e.g. disease signs and symptoms, lab results) / Vomiting or diarrhea (3 or more loose stools in a 24 hour period)
Person / Anyone affiliated with MadisonCollege (including students, faculty, staff, and visitors)
Place / MadisonCollege
Time / Symptom onset between March 9 and present

Question6: What is the purpose of these initial interviews?

Suggested answer: The purpose of the early interviews is to generate one or more hypotheses about potential sources of disease exposure. The hypothesis-generating interviews do not need to be conducted with all of the case-patients, but rather with a subset. Ideally, these initial interviews will allow you to narrow down the potential sources of exposure, create a plausible hypothesis, and then conduct hypothesis-testing interviews with both ill and healthy persons to determine whether or not the hypothesis is correct.

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

Question7: Will you create a new questionnaire specifically for this outbreak, or will you use an existing questionnaire? If you choose to use an existing questionnaire, please discuss where this questionnaire is located, and how it will need to be modified for this outbreak.

Suggested answer: Although you do not yet know the cause of the outbreak, most GI illnesses result from exposure to contaminated food or water. You should obtain a food history and gather information about water exposure. You may want to start by using a standardized foodborne illness questionnaire for hypothesis-generating interviews.Keep in mind that you may need to add or subtract specific questions as appropriate. Once you have a better idea of possible sources, you will refine this questionnaire, or create a new questionnaire based on the suspected sources of infection.

Two standard questionnaires for foodborne disease outbreaks can be found in the Foodborne Disease Surveillance and Outbreak Investigation Toolkit from the Centers for Disease Control and Prevention ().

Question 8: Should your health department activate its Incident Command System (ICS) at this time? If so, who is responsible for activating the ICS?

Suggested answer: Yes, it is probably a good idea to activate the Incident Command System for several reasons. First, there are at least 29 cases, which represent a rather large outbreak. Second, this is a good opportunity to practice using the ICS and to work out problems or kinks in the system. The purpose of the ICS structure is to facilitate effective emergency management across multiple agencies and jurisdictions. For a local health department, the ICS structure can be used to deal with disease outbreaks. The person responsible for activating the ICS structure may vary, but is often the local Health Director.

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

Question9: What actions will your environmental health specialist take in response to this information?

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

Suggested answer: Sanitarians and/or environmental health specialists should visit the student union cafeteria as soon as possible and take the following actions:

1)Inspect kitchen facilities and observe food handling practices;

2)Collect samples from the salad bar, deli sandwiches, and sweet tea;

3)Determine whether any of the food from March 9th and 10th is available – instruct staff not to discard those items;

4)Recommend thorough cleaning and maintenance of all cafeteria and kitchen facilities while students are away on spring break;

5)Interview staff members to determine whether any have been ill;

6)Collect stool samples from food handlers; and

7)Educate kitchen staff about handwashing, maintaining foods at proper temperatures, and regularly disinfecting kitchen facilities.

Question10: Is it time to conduct a case control or a cohort study? If so, which type of study design will you use? Explain your answer.

Suggested answer: Yes, it is appropriate to conduct a case control study at this time. Case control studies are best when it is difficult to identify every person who may have become ill, or when your time and resources are limited. A cohort study is the best choice when it is easy to identify everyone who may have become ill (e.g. all attendees at a wedding reception or all students in a single classroom), and there are enough resources to interview everyone. In this situation, a case control study is best because it would be very difficult to track down every student who ate in the student union cafeteria on March 9th and 10th, and there would probably be too many students to interview. Instead, by conducting a case control study, you will only need to interview the ill students and a sample of healthy students (controls). If possible, it would be ideal to identify and contact some controls before the weekend, when it will become more difficult to reach students on spring break.

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

Question 11: Based on the laboratory results, what infection control measures would you recommend?

Suggested answer: Norovirus is highly infectious, so there is certainly the possibility of secondary transmission. You and your Epi Team should consider appropriate infection control measures, which may include:

1)Sending reminders to the entire college community about the importance of handwashing with soap and water;

2) Providing a norovirus fact sheet to the college community;

2) Excluding foodhandlers from work if they have been ill in the past 72 hours; and

3)Recommending thorough disinfection of all cafeteria facilities.

Note that these infection control measures are not specific to norovirus, but rather are applicable to many foodborne diseases. It is not necessary to wait for laboratory confirmation of the exact pathogen before instituting disease control measures. You probably would have begun many of these control measures early in the outbreak.

Question 12: What activities, if any, is your health educator conducting at this time?

Suggested answer: This is an ideal time for your health educator to provide information to people at MadisonCollege about prevention of norovirus and other foodborne illnesses. The best way to share this information depends upon the specific setting, but you may consider the following:

1)Sending an email announcement to the college community with facts about norovirus, including prevention strategies such as handwashing;

2)Posting flyers in public restrooms and eating areas reminding people about the importance of handwashing;

3)Asking the student newspaper to write an article about foodborne illness;

4)Posting norovirus fact sheets or brochures at Student Health Services; and

5)Posting information about norovirus and other foodborne illnesses to college and health department websites.

In addition to educating the college community, this outbreak provides a good opportunity to educate the community at large, so consider strategies that will reach a larger audience as well.

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

Question13: Should you attempt to contact Marcia again? Why or why not? Are there confidentiality issues that you should consider?

Suggested answer: Yes, it is probably wise to try to speak with Marcia one more time, perhaps in an environment where she is more comfortable. It would be good to get a stool sample if possible, ask again about symptoms of illness, and determine whether she has close contacts who have been ill. However, keep in mind the importance of maintaining confidentiality in this situation.

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

1

10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 19 / 20 / 21 / 22 / 23 / 24 / 25

Question 14: How do you interpret these findings overall?After describing the findings generally, interpret the odds ratio for deli roast beef. Remember, an odds ratio is the odds of exposure among cases divided by the odds of exposure among controls.

Suggested answer: The data show that overall, the odds of eating deli ham, turkey, or roast beef were significantly higher among cases compared to controls. The odds of eating the salad bar items (lettuce, tomatoes, cucumbers, onions, green peppers) are not significantly different between cases and controls. The significant odds ratios for the deli meats may lead us to suspect the deli meats as a source of disease. The interpretation of the odds ratio for deli roast beef is: Cases were 2.66 times more likely to have eaten deli roast beef than controls.(For more information on calculating and interpreting odds ratios, see the FOCUS issue “Case-Control Studies for Outbreak Investigations” listed in the background materials.)

Odds ratio = 1: odds of exposure are the same among cases and controls (no association between exposure and disease)

Odds ratio > 1: odds of exposure are greater among cases than controls (a positive association between exposure and disease)

Odds ratio < 1: odds of exposure are less among cases than controls (a negative, or protective, association between exposure and disease)