For years, epidemiological studies have turned up an odd, inconvenient finding: people who eat more ice cream tend to have lower rates of type 2 diabetes and heart disease than people who eat less — even though ice cream checks every box of what nutrition science tells us to avoid (ultra-processed, high in sugar and saturated fat). The Atlantic digs into why researchers have been reluctant to highlight this result, and what it reveals about the limits of the "ultra-processed food is inherently bad" framework that's come to dominate public health messaging. It's a good read for anyone skeptical of tidy nutrition narratives — a reminder that food categories like "UPF" can obscure as much as they explain, and that correlation in diet studies is a minefield.

Read more here!