Bans in the early days of the pandemic saved millions of lives and prevented tens of millions of infections. But in most places, quarantine measures have never been perfectly implemented. While governments have the power to shut down shops and other places where people gather in public, the U.S. mandates did not prevent citizens from meeting friends and family at home.
Anecdotally, informal social gatherings such as vacation get-togethers, parties, and weddings appear to have played an important role in spreading the virus. But there is no easy way to rigorously test this hypothesis on a large scale. A new study published in JAMA internal medicine avoids these difficulties by analyzing the correlation between birthdays and COVID-19 infection rates. According to the results, households with a recent birthday – and thus a greater likelihood of throwing a party – in counties with many cases had an approximately 30 percent higher risk of infection in the following two weeks compared to households that did not have a birthday . The authors found that those celebrating a child’s birthday were at the highest risk of all.
“Everyone in the study was surprised at how big the risk increase that 30 percent seemed,” says lead author Christopher Whaley, a health policy researcher at Rand Corporation, a nonprofit research group based in Santa Monica, Calif. That really speaks to where many infections come from came last year. “
While a handful of previous studies examined the role of informal gatherings in novel coronavirus infections, they mostly focused on one-off events like a single Maine wedding reception. Whaley and colleagues realized that birthdays could be a particularly useful tool in uncovering national COVID-19 transmission patterns for a number of reasons, including the fact that every person has one and they occur year round.
The authors obtained nationwide representative data for 2.9 million US households from Castlight Health, a company that helps people find their way around the health system. The data, which spanned the first 10 months of 2020, included the birthdays of family members as well as any positive COVID-19 test results they received based on insurance claims. While the researchers don’t know if the people included in their study actually had a birthday party, they hypothesized that birthdays could act as a proxy for the likelihood of a social gathering occurring.
The results showed a significant correlation between birthdays and an increased risk of COVID-19 transmission. In the counties with the highest prevalence of the disease (the top decile), there were an average of 8.6 more cases per 10,000 inhabitants in households with a birthday than in other households without birthdays. In areas with low virus prevalence, birthday-related infections were also low. Significantly, the results were not influenced by political bias, as measured by how the counties voted in the 2016 elections. And they were not affected by quarantine orders or the weather. Taken together, this suggests a more universal pattern of behavior than something influenced by extenuating circumstances, such as rain bringing a party into the house or household ideology. “When you hang out with family and friends informally, the more likely you are going to become less alert or not wearing a mask,” says Whaley. “Psychologically, you may not think that family and friends have the same level of risk as the general public.”
Whaley and his colleagues were particularly surprised to see the potent effect that children’s birthdays appeared to have on the risk of transmission of COVID-19. In the counties with the highest prevalence of the disease, a child’s birthday is more than double the risk of infection for an adult, with an increase of 15.8 additional COVID-19 cases per 10,000 residents compared to the numbers for people who do not have a birthday had . While the researchers don’t know what is responsible for the added risk, Whaley hypothesizes that, perhaps even in the midst of a pandemic, parents may be more reluctant to go without a party for a child or that party goers on a children’s birthday party may be less disciplined about it social distancing are.
For both children and adults, however, the results only reflect the average impact of a birthday on the risk of COVID-19, which means they are almost certainly an underestimate for those who have gathered for celebrations over the pandemic. If Whaley and his colleagues had been able to distinguish between people who were having a party and those who were not – or between people who are having a party, who wore masks and social distancing at that gathering or not – , the results would likely have been even more pronounced.
Still, the new study clearly confirms that “it’s okay to be a partygoer during a pandemic,” says Donald Redelmeier, a medical professor at the University of Toronto who was not involved in the research. “In retrospect, it’s a great affirmation of the gains among those who skipped all kinds of social gatherings – weddings, birthday parties, whatever.”