Happy Birthday! I Got You COVID

F. Perry Wilson, MD, MSCE


June 23, 2021

Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

This transcript has been edited for clarity.

Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I'm Dr F. Perry Wilson of the Yale School of Medicine.

On March 6, 2020, I attended a birthday party for my friend Doug.

Though we didn't know it at the time, this would be the last time we'd gather for a birthday for more than a year. The next week, with the first COVID cases appearing in Connecticut, the world shut down. It's only now starting up again.

The folks who were at that party all remember it the same way — as a sort of gauzy past before the realities of the present came crashing down. If we only knew.

Fortunately, no one at that party got COVID. Well, at least not from the party. But small gatherings like Doug's party are a potential important source of transmission, though this has been really hard to measure — at least unless you get clever, which is what a team led by Anupam Jena did in this article appearing in JAMA Internal Medicine.

Birthdays are the key, it turns out. See, birthdays are pretty random.


But when a birthday happens, the likelihood that a small gathering happens goes up; we all want to celebrate the Dougs of the world.


As such, birthdays might be a good proxy for small gatherings. And since birthdays are more or less random, we can use birthdays to test whether, and how much, small gatherings contribute to COVID infections. This is known in the biz as an instrumental variable analysis. Quick aside: If you ever meet an epidemiologist, ask them what their favorite instrumental variable is. You'll get an earful.

Jena and his team leveraged a commercial insurance database containing information on more than 6 million people and 2.9 million households. They used the birthdates present in that data to ask a simple question: Did the rate of COVID noticeably increase among people in the 2 weeks following a birthday of someone in their household?


Okay, I said it's a simple question, but it's not, really. They had to control for a number of factors, the most important being the local prevalence of COVID. If there is no COVID around, you wouldn't expect birthday parties to increase risk much. If there's lots of COVID around, the risk might be substantial.

In any case, what they found was a pretty strong link between birthdays and new COVID infections.


There was even a dose-response effect between the birthdays and the COVID rate based on the background prevalence of COVID in the community, just as you might expect. In fact, during times when COVID prevalence was highest, households where someone had a birthday had a 30% increased risk for a new COVID infection compared with households where no one had a birthday.

And it's worth noting that lots of people, myself included, just skipped the birthday parties during the highpoints of the pandemic curve. That would tend to dilute out these effects, suggesting that, in reality, small gatherings are quite a bit worse than this.

Now, there are a lot of assumptions built into this analysis, and the authors did a nice job trying to disprove their own results. For one thing, they performed a falsification analysis by flipping the temporal relationship between the exposure and outcome. The idea here is that if birthday gatherings are driving COVID infections, that only works forward in time; you shouldn't be at any higher risk in the weeks before your birthday, unless there is some other bias that links birthdays with COVID infections that has nothing to do with the gatherings that go along with birthdays.


No reverse link was found, further supporting the primary results. Sure, getting COVID might make you cancel your birthday party, but remember, that's not what's being studied here. Getting COVID can't change your actual birth date.

Not all of the analyses panned out, though. They specifically evaluated milestone birthdays — 40th, 50th, 60th — in addition to just any old birthday. The thought here was that you're more likely to have a party on those big birthdays, COVID be damned. But there was no dose-response effect noted there. That may be because the strongest effects emerged when kids in the household had birthdays, not adults.

In the places with the highest community prevalence of COVID, households with kids' birthdays had a threefold higher risk of having a new COVID case than households with adults' birthdays. It makes sense, honestly. It's way easier to skip your own birthday than to deny that to your child.

The world is a bit different now than when this study was conducted. Vaccines make birthday parties way safer than they used to be. But COVID continues to spread, particularly among the unvaccinated, and studies like this remind us that we probably need to worry just as much about the small, intimate gatherings as the bars, restaurants, and clubs that get most of our attention.

F. Perry Wilson, MD, MSCE, is an associate professor of medicine and director of Yale's Clinical and Translational Research Accelerator. His science communication work can be found in the Huffington Post, on NPR, and here on Medscape. He tweets @fperrywilson and hosts a repository of his communication work at

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: