19. Outbreak Investigation Step by Step: Reports to Final Report
After this lesson you can
- List the steps of an outbreak investigation in order and explain why they overlap and loop
- Establish an outbreak against its baseline and write a tiered case definition
- Build a line list, read the epidemic curve, and estimate the exposure window
- Compute food-specific attack rates and risk ratios, or an odds ratio when no guest list exists
- Map control measures onto the chain of infection and declare an outbreak over after two maximum incubation periods
- Structure the final report in five headings
Video coming soon
Scheduled for release on November 16, 2026.
The course capstone, a step-by-step outbreak investigation of a fictional Oakhaven picnic with 100 attendees and 40 cases. It runs from observed-versus-expected counts and a tiered case definition through the line list, epidemic curve, exposure window and food-specific attack-rate table to control measures, the end-of-outbreak rule and the written report.
Questions and answers
Questions a learner would ask about this lesson. Think through your own answer before you open each one.
- How do you know an outbreak is really happening?
Show answer
Compare observed counts with expected counts. In Oakhaven, the same week in five past years brought 10, 12, 16, 17 and 20 enteric reports, a mean of 15, with an alert threshold of 23; this week brought 26. Then rule out artifacts such as a new case definition, a new laboratory test, changed reporting rules or a population change, and note whether many cases share one event, here a Friday picnic.
- What goes into a case definition, and why use tiers?
Show answer
A case definition has clinical criteria plus person, place and time: for the picnic, diarrhea or vomiting in an attendee with onset after the Friday noon meal. Tiers of confirmed, probable and possible describe certainty, not different illnesses. Use a loose, sensitive definition early to find cases, and a tight, specific one in the analysis so that non-cases do not dilute the comparison.
- What does the epidemic curve tell the investigators?
Show answer
The 40 picnic cases form one peak with a steep rise and a slower fall, the point-source shape of a single shared exposure. Working back from the earliest onset (hour 25 minus a minimum incubation of 12), the median onset (about hour 40 minus an average incubation of about 28) and the latest onset (hour 59 minus a maximum incubation of 48) places exposure at hours 11 to 13, around the Friday noon meal. The right tail reflects the spread of incubation periods, not continued exposure.
- Why read both columns of the attack-rate table?
Show answer
In an outbreak that made 40 of 100 people ill, every food has a high attack rate among eaters. Only the comparison with non-eaters implicates one food. Potato salad had 36 of 60 eaters ill against 4 of 40 non-eaters, a risk ratio of 6.0, while chicken, ice cream and bread stayed near 1.0.
- When can an outbreak be declared over?
Show answer
The rule of thumb is two maximum incubation periods after the last onset with no new case. For the picnic agent that is 2 x 48 = 96 hours after the last onset at hour 59, which reaches hour 155, about Thursday noon. One quiet day is not the end, because a secondary case in a household could still appear.
Practice quiz
- Question 1After establishing that an outbreak exists and verifying the diagnosis, what is the next step?
- Question 2Sixty attendees ate the salad and 36 became ill; 40 did not eat it and 4 became ill. What is the risk ratio?
- Question 3A banquet with a complete guest list is best investigated with which design?
- Question 4Without a guest list, investigators interview 40 cases and 40 well controls. Of the cases, 36 ate the salad and 4 did not; of the controls, 16 ate it and 24 did not. What is the odds ratio?
- Question 5Attack rates are 10 percent among the unvaccinated and 2 percent among the vaccinated. What is the vaccine efficacy?
- Question 6The last onset was at hour 59 and the maximum incubation period is 48 hours. How long after the last onset, with no new cases, can the outbreak be declared over?
- Question 7The picnic curve has one peak, a steep rise and a slower fall, with every onset inside one incubation range. What does this shape suggest?
- Question 8Chicken had 29 of 70 eaters ill and 11 of 30 non-eaters ill. Why does this not implicate the chicken?
- Question 9When should control measures begin?
- Question 10Why should the analytic step use a tight, specific case definition rather than a loose one?
Saved in your browser only: no account, no server.
Principles of Epidemiology is education in epidemiologic reasoning, not clinical, screening, or policy guidance. Every example is set in the fictional Oakhaven Health District, and every number is invented for teaching.