Anatomy of an Outbreak Investigation

Thursday, July 15th, 2021
man working on computer - photo from

By: Drew Barkley

Before I began working on my MPH, I knew very little about how foodborne disease outbreaks (or any outbreaks for that matter) were investigated. I just assumed that health departments would compare lab test results from a large group of people that got sick and look for commonalities in the foods they consumed or activities in which they participated. However, during my MPH and through my work at CFI, I have gained a greater knowledge and appreciation for the work that goes into an outbreak investigation. As a result, I would like to share a brief overview of the “Anatomy of an Outbreak Investigation” so that you might better appreciate the work that goes into such an investigation.

Outbreak Investigations can be divided into 3 main phases: descriptive, analytical, and intervention. The descriptive phase is first and is focused on information and data gathering. The analytical phase follows and uses the data gathered during the descriptive phase to look for potential risk factors and areas for intervention. The final phase, intervention, uses the results of the analytical phase to make more targeted interventions to try to halt the outbreak and stop the spread. I will now dive in and provide a little more detail on what occurs during the 3 phases.

The first step of the descriptive phase (and of any outbreak investigation) is to determine if an outbreak is occurring.  Here, it is worth mentioning the definition of an outbreak. An outbreak or epidemic occurs when we see levels of a disease above what is expected. For example, we expect to have a certain number of flu cases each flu season. However, if we see higher levels of flu than expected (as with 2009-2010 H1N1), an outbreak is declared. In the case of the 2009-2010 H1N1 outbreak, because it was an epidemic or outbreak affecting the whole world, we refer to it as a pandemic. Right, so back to step 1. To understand an outbreak is occurring, we must know if the number of cases of a disease is higher than we expect. For foodborne disease, this is done through surveillance systems such as FoodNet and PulseNet. Each year, we expect a certain number of sporadic cases of the various foodborne pathogens (Salmonella, E. coli, Campylobacter, etc.) based on trends from previous years. If we ever see an unexpected spike in cases, the alarm is sounded, and an outbreak is declared. Outbreaks can also be declared when a specific strain of a pathogen is detected in several, seemingly unrelated people. This is done using PulseNet which sequences pathogens recovered from sick individuals. If the same sequence appears in multiple individuals, an outbreak can be declared.

The next step of the descriptive phase is to create a case definition and identify your outbreak cases. Setting an appropriate case definition is an important step as making the definition too inclusive may overload the system and overestimate the size of the outbreak. However, making the definition too exclusive will leave out cases and valuable data. Using simple, objective criteria for the case definition is the best way to ensure systematic and uniform inclusion of cases. The case definition can also evolve over time as more is learned about an outbreak. For example, if an outbreak of Salmonella were occurring, you might start with a case definition of any individual with lab-confirmed Salmonella matching the outbreak strain. Then as you learn risk-factors and potential source foods, you can target the case definition to be any individuals with lab-confirmed Salmonella matching the outbreak strain who ate certain foods or attended a specific event.

Once the cases are identified, you want to begin part 3 of the descriptive phase, describing the occurrence of disease in time and place. Here is where you will find the “epidemic curve,” a plot showing cases over time, as well as maps showing spatial trends in the outbreak. In a foodborne outbreak investigation, investigators will also conduct interviews with confirmed cases to obtain information on what they ate, how they prepared food, if they attended any events or restaurants, etc. These interviews are performed by local health departments and contain questions based on previous foodborne contamination events in order to avoid bias and better direct questions. The information gathered here will aid investigators in the analytical phase of the investigation.

Speaking of the analytical phase, that is the next stop in our outbreak investigation. This phase has two steps: hypothesis generation and hypothesis testing. Using the epidemiological data collected in phase 1, investigators try to answer the following questions:

  1. What is the cause of the outbreak? (what is the pathogen or infectious agent?)
  2. What is the source of the outbreak? (common point source or multiple sources?)
  3. How did the outbreak spread? (what is the transmission pathway?)
  4. What are the risk factors associated with the outbreak?

The goal behind answering these questions is to identify key components of the outbreak in order to inform the third and final phase of the outbreak investigation.

The last phase of an outbreak investigation is the intervention phase. Similar to the previous phase, there are 2 parts to the intervention phase: implementing control and prevention measures and communicating the findings. The goal of any outbreak investigation is to stop the outbreak from spreading and limit the number of illnesses and deaths. Information gathered during the previous two phases of the investigation informs the interventions of the third phase. For example, in foodborne disease outbreaks, once investigators learn the source of an outbreak (i.e. romaine lettuce from Yuma, AZ), they will advise people to avoid food from that source until corrective action is taken. In cases where there is a food processor or company behind the outbreak, investigators will gather enough information to convince the processor or company to make a voluntary recall in order to help prevent the spread.

Lastly, it is crucial to communicate the findings from any outbreak investigation. As humans, we can make mistakes while learning along the way. In an outbreak investigation, mistakes may get made and there may be things the investigators wished they had done differently. By communicating and reporting on the outbreak as it happened, future investigations can be aided by the knowledge gained during previous investigations. Once the final numbers are in and reported, the outbreak investigation can be closed. Hopefully now you have a better understanding of how outbreaks are investigated and the work that goes into completing such an investigation. If you would like to dive deeper into how outbreak investigations occur, I recommend starting at the CDC’s website at the following link: 

Drew BarkleyDrew Barkley

Graduate Research Associate




Posted In:
July 15, 2021 - 4:35pm --

Add new comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.