Our most reliable pandemic number is losing meaning
Daily pandemic data are, and have always been, riddled with quirks. They’re subject to holiday lags (remember Labor Day?), weekend lags, general lags, inconsistencies, and all kinds of denominator problems. They’ve become a language few speak fluently, while the rest of us are bumbling around like tourists asking for directions.
This spring, we shared with you five expert tips for reading COVID data like a pro. That was before the Delta variant changed everything. Consider today’s newsletter a metaphorical booster shot for your data fluency.
A reliable pandemic metric is maybe not as stable as was once thought. Although cases don’t always capture who is sick—especially as Delta has prompted more asymptomatic and mild cases in vaccinated people—hospitalizations are supposed to paint a more accurate picture. But still-early research suggests that even that metric may come with caveats.
Breakthrough case counts sometimes include asymptomatic infections. “That means breakthroughs writ large aren’t the most relevant metric to use when we’re evaluating vaccines,” my colleague Katherine J. Wu wrote back in July.
No one knows what “booster bandits” mean for vaccination data. Even the CDC says it can’t tell how many people got an illicit third shot, but that behavior could cause “public-health authorities [to] think more people have gotten their first or second shots than is actually the case,” our senior associate editor Rachel Gutman warns.
We’re in the messiest phase of the pandemic yet. The ways the virus is changing are forcing us to rethink what we do and don’t know, Alexis C. Madrigal, a contributing writer and a co-founder of the COVID Tracking Project at The Atlantic, wrote last month.