NURS 8310 Week 4 Outbreak Design Memo Example

Memorandum

To: County communicable disease team and shelter partners

From: DNP student, Epidemiology and Population Health

Date: Month Day, Year

Re: Choose a case-control design for the shigellosis cluster, and switch to a cohort only if one shared meal explains it

The Situation

In the past three weeks, the county has confirmed 14 cases of shigellosis, with laboratory results showing the same Shigella sonnei strain, among adults who are either staying at the county's largest shelter or camping in the riverside area within a mile of it. Several more people with diarrheal illness have not yet been tested. Shigellosis spreads easily from person to person through the fecal-oral route, and outbreaks among people experiencing homelessness have been reported in which limited access to toilets and handwashing played a part (Hines et al., 2016; Murti et al., 2015). This cluster and its figures are composite, assembled for the purpose of this memo.

Before the team can test any hypothesis about the source, it must choose a study design. This memo recommends a case-control study, names the one circumstance in which a cohort study would serve better and places the decision within the usual sequence of an outbreak investigation.

Where the Decision Sits in the Investigation

Field investigations follow a familiar sequence: confirm the diagnosis and that an outbreak exists, write a case definition, search for more cases and describe who they are, where they were and when they fell ill, develop hypotheses, test them with an analytic study, implement control measures and communicate findings (Rasmussen & Goodman, 2019). The team has confirmed the diagnosis and the cluster and has begun describing cases. Descriptive findings so far suggest three hypotheses: a meal served at the shelter, shared use of the shelter's restrooms, and use of a single portable toilet and handwashing station set up near the camping area, which staff report ran out of water on several days. The analytic study must be able to test all three at once. Control measures, including hygiene supplies, restroom cleaning and exclusion of ill food handlers, should begin now and not wait for the study's results.

Why Case-Control

A case-control study compares people who have the disease with people who do not, looking back at their exposures (Schulz & Grimes, 2002). Four features of this cluster favor that design.

The population at risk is not well defined. A cohort study needs a clearly bounded group whose members can all be followed and asked about exposure. Here, the at-risk population includes shelter guests on different nights and people camping nearby who may never enter the shelter. There is no complete list of who was exposed to what.

Cases are few. With 14 confirmed cases, a case-control study can enroll all of them and a larger number of controls, which is efficient when the outcome is uncommon relative to the population.

Several exposures must be tested. A case-control questionnaire can ask about meals, restroom use, the portable toilet, handwashing, sharing of food or drink and contact with ill people at the same time.

Speed matters. Cases and controls can be interviewed within days, with the help of outreach workers who know the camping area.

Selecting Controls

Controls should be drawn from the community the cases came from: shelter guests and people camping nearby who have not had diarrhea in the past month. The team should enroll two or three controls per case, recruited on the same days and in the same places where cases were found, by outreach workers and shelter staff. Recruiting controls only from the shelter would miss the camping population and could bias the comparison toward shelter exposures. Each interview should cover the same time window before illness onset for cases and before the interview date for controls.

When a Cohort Would Serve Better

There is one circumstance in which the team should switch to a retrospective cohort study: if descriptive data show that most cases ate one particular shelter meal, and the shelter has a roster of everyone who ate it. In that situation, the exposed group is defined and listed, attack rates can be calculated for those who ate each food item and those who did not and the relative risk can be estimated directly. That is the classic design for a point-source food outbreak. At present, onset dates are spread across three weeks, which suggests ongoing person-to-person spread rather than a single meal, so the cohort design does not fit yet.

Limits to Anticipate

Recall of exposures over several weeks will be imperfect, and people who are ill may remember exposures differently from those who are not. The team can reduce this by asking about specific, concrete exposures and by using shelter records to confirm who stayed on which nights. Some people with mild illness who were never tested will be counted among controls, which would weaken any association found. Finally, if nearly everyone in both groups used the portable toilet, the study may not be able to separate its effect, and environmental findings will carry more weight.

Recommendation

Proceed with a case-control study of all confirmed cases and two to three community-matched controls per case, testing the meal, restroom and portable toilet hypotheses together. Review descriptive data daily and switch to a cohort design if a single shared meal emerges. Begin control measures now.

References

Hines, J. Z., Pinsent, T., Rees, K., Vines, J., Bowen, A., Hurd, J., Leman, R. F., & Hedberg, K. (2016). Notes from the field: Shigellosis outbreak among men who have sex with men and homeless persons: Oregon, 2015-2016. MMWR. Morbidity and Mortality Weekly Report, 65(31), 812-813. https://doi.org/10.15585/mmwr.mm6531a5

Murti, M., Louie, K., Bigham, M., & Hoang, L. M. N. (2015). Outbreak of shigellosis in a homeless shelter with healthcare worker transmission: British Columbia, April 2015. Infection Control & Hospital Epidemiology, 36(11), 1372-1373. https://doi.org/10.1017/ice.2015.211

Rasmussen, S. A., & Goodman, R. A. (Eds.). (2019). The CDC field epidemiology manual. Oxford University Press.

Schulz, K. F., & Grimes, D. A. (2002). Case-control studies: Research in reverse. The Lancet, 359(9304), 431-434. https://doi.org/10.1016/S0140-6736(02)07605-5