Confession. One of my favorite parts of applied epidemiology is trying to find the source of an outbreak. It is, after all, what the Father of Epi, John Snow, did to stop the cholera outbreak.

Identifying the *source* of an outbreak of a respiratory disease is more difficult. On an individual case level, it’s rare that we can, unless we do molecular epidemiology. We can, however, identify clusters.

There are two parts to Disease Investigation:

Prospective Contact Tracing

This is when we talk to a case, find out with whom they’ve been in contact (for COVID, we ask for 48 hours prior to testing or symptom onset) until the day you call and we tell you to isolate. And then you call all those contacts, ask about symptoms, housing, quarantine, testing, etc. And you follow up to make sure they’ve gotten tested, are doing ok, have access to food and other resources.

This is how you stop the spread of disease moving forward.

Retrospective Contact Tracing

Because the incubation period for COVID-19 can be 14 days, the investigator also asks where you’ve been/with whom you’ve been for 14 days prior to symptom onset (or day of test). The epi teams review the data regularly to see if any patterns are emerging. It’s very much like being a detective.

For example, upon interviewing five separate people, the epidemiologist discovers all had been at the same restaurant on the same day. Bingo. The restaurant is the source! Wait, no. You don’t know that yet.

You look at the contacts of each — did the five individuals know each other (which would be determined by “with whom you’ve been”). Were those 5 folks at the same restaurant because they’re friends and were sitting together?

And if they went to a restaurant together, where else did they go together?

So you dig deeper — did they mention other gatherings? Five friends at a restaurant together may also have been in the same car, or at one’s house. And sure enough, they mention getting together for drinks at one’s house. That person had a housemate who was also sick.

They may have been at the restaurant together, but the restaurant likely had nothing to do with the transmission. Or rather, the staff and other patrons did not. The restaurant is not the source of this particular cluster.

On the other hand, the epi finds that these five people don’t know each other, were each sitting at different tables, had no interaction. So the epi checks out the restaurant — how is their seating set up? What are their capacity limits? What is the waiting situation? Have any staff at the restaurant been ill, even mildly? And you find some red flags.

In this case, the cluster IS likely linked to the restaurant, and you work with them to improve their protocols so it doesn’t happen again.

Restaurants and bars are assumed to be higher risk for transmission because people don’t wear masks when eating and drinking and they’re more relaxed. But the same applies to private homes. And TBAC has said that liquor sales — i.e. people buying for home, not at bars — are up dramatically.

Schools, weddings, churches, summer camps, family parties have all had such clusters traced to them. And yes, some bars and restaurants too.

But the goal is not to find the source of blame. It’s to identify potential clusters — or hotspots — so you can intervene to prevent further spread.

I love this part of epidemiology. It’s like fitting together the pieces of a puzzle.

Image for post
Image for post
An example of a cluster traced to a family gathering.

Dr. Rohr-Allegrini is an epidemiologist and tropical disease scientist currently working to prevent diseases through immunizations.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store