PU520 Unit 5 Community Dinner Outbreak Investigation Example

Seventy-Four of Two Hundred Ten After the Community Dinner: An Outbreak Investigation Report Working a Composite Thanksgiving Meal From the First Call to a Probable Vehicle, With a Case Definition, an Epidemic Curve in Two-Hour Bins and Food-Specific Attack Rates

[Student Name]

Purdue University Global

PU520: Principles of Epidemiology

Unit 5 Outbreak Investigation

[Instructor Name]

[Date]

The event, the attendees and every figure are composite and bracketed for the writer's own data; the investigation steps and research cited are real.

Notification and Verification

On the Friday after Thanksgiving, the hall's coordinator called the county health department: several guests from Thursday's free community dinner reported diarrhea and cramps overnight. A nurse epidemiologist confirmed [11] ill people in the first hour of calls, all of whom had eaten at the dinner between [13:00] and [15:00]. That clustering by person, place and time verified an outbreak.

Case Definition

Confirmed case: a person who ate at the dinner, developed diarrhea within [24] hours, and had Clostridium perfringens enterotoxin detected in stool. Probable case: a person who ate at the dinner and developed diarrhea or abdominal cramps within [24] hours. A definition comes before any count, because without one the outbreak cannot be counted at all.

Finding Cases

The coordinator's sign-in sheet listed [210] attendees with phone numbers. Staff reached [196] by telephone within three days and asked each person the same questions about symptoms, when they began and which dishes they ate. [74] met the case definition: [5] confirmed and [69] probable.

Epidemic Curve

Onset times, plotted in two-hour bins with the meal marked on the horizontal axis, form a single sharp peak. The first case began about [6] hours after the meal, the median about [11] hours, and nearly all cases occurred within [20] hours. Illness lasted under [24] hours for most people, with diarrhea and cramps and little vomiting or fever. A single peak with an incubation of six to twenty hours and a short, mild illness points to a toxin-producing bacterium and a single shared meal. A single peak indicates a point-source exposure, everyone exposed at the same time, which is the classic shape for a shared meal (Celentano & Szklo, 2019). The incubation and symptoms fit Clostridium perfringens better than norovirus, which more often causes vomiting and has a longer incubation.

Food-Specific Attack Rates

Each food is compared between those who ate it and those who did not, out of [210] attendees. Gravy: ate [68] ill of [140], [48.6] percent; did not eat [6] of [70], [8.6] percent; relative risk [5.7]. Turkey: [70] of [180], [38.9] percent; [4] of [30], [13.3] percent; relative risk [2.9]. Stuffing: [52] of [150], [34.7] percent; [22] of [60], [36.7] percent; relative risk [0.9]. Mashed potatoes: [55] of [160], [34.4] percent; [19] of [50], [38.0] percent; relative risk [0.9]. Green beans: [40] of [120], [33.3] percent; [34] of [90], [37.8] percent; relative risk [0.9]. Pumpkin pie: [50] of [150], [33.3] percent; [24] of [60], [40.0] percent; relative risk [0.8]. Gravy has the highest attack rate among those who ate it, the lowest among those who did not, and the largest relative risk. Most people who ate turkey also took gravy over it, which likely explains turkey's elevated risk.

Environmental Investigation

The gravy was made the day before from turkey drippings, cooled overnight in a deep stockpot in a walk-in refrigerator and reheated to warm, not boiling, before serving. Clostridium perfringens spores survive cooking and germinate as food cools slowly through warm temperatures, multiplying rapidly; inadequate reheating leaves the bacteria alive. Leftover gravy grew Clostridium perfringens at high counts.

Why the Setting Fits

Clostridium perfringens outbreaks reported nationally occur year round but most often in November and December, frequently involving meat, poultry and gravies prepared in large quantities (Grass et al., 2013).

Ruling Out Other Explanations

Norovirus was considered first, because it is the most common cause of foodborne outbreaks and spreads easily at large gatherings. Three features argued against it: vomiting was uncommon, the incubation was shorter than norovirus usually produces, and there were no secondary cases among household members who did not attend. A volunteer food handler with recent illness was also considered; none of the [14] volunteers reported illness before the dinner, and the attack rates did not point to foods one person handled alone. A water problem at the hall was ruled out because illness tracked a specific food rather than everyone who drank from the tap.

Who Was at Risk

Attack rates were similar across age groups, though two older guests with chronic illness visited an emergency department for dehydration and were discharged the same day. No one was admitted.

Control Measures

Discard leftover gravy. Teach volunteers to cool large batches quickly in shallow pans, and to reheat to at least [165] degrees Fahrenheit. Provide the hall with a cooling log and a food thermometer. Notify other congregations that host holiday meals.

Steps Followed

The investigation moved through the usual stages, from confirming that an outbreak existed and counting cases, through describing them and testing an explanation, to tracing the source and stopping further illness (Reingold, 1998). Writing the steps into the report lets a reader see that none was skipped.

Limits

Recall of foods eaten two days earlier may be imperfect. Only five people provided stool specimens.

References

Celentano, D. D., & Szklo, M. (2019). Gordis epidemiology (6th ed.). Elsevier.

Grass, J. E., Gould, L. H., & Mahon, B. E. (2013). Epidemiology of foodborne disease outbreaks caused by Clostridium perfringens, United States, 1998-2010. Foodborne Pathogens and Disease, 10(2), 131-136. https://doi.org/10.1089/fpd.2012.1316

Reingold, A. L. (1998). Outbreak investigations: A perspective. Emerging Infectious Diseases, 4(1), 21-27. https://doi.org/10.3201/eid0401.980104