Sounds like this “study” (aka a self-reported, retrospective, epidemiological survey - which is a type of statistics that I think just confuses the public to call a study but whatever) needs a lot more work to say anything with certainty. The kicker in the article is this I think:

“…the different windows of time-restricted eating was determined on the basis of just two days of dietary intake.” Yikes. That, and it sounds like they didn’t control for any of the possible confounding variables such as nutrient intake, demographics, weight, stress, or basically any other risk factors or possible explanations. Its entirely possible that once they actually control for this stuff, the correlation could shrink to almost nothing or even reverse when we see that people who tried this diet were just baseline higher risk than who didn’t.