CDC's first binge-eating question finds 40% of high schoolers report it over 30 days

A 30-day window catches every student a three-day window does, and more, so the 40% ^[1] and the 17% ^[3] do not conflict. The subgroup figures need more care. STAT reports 19% of female and 16% of male students ^[5]. Those bracket the three-day rate and sit 21 and 24 points below the past-month rate, so they appear to be three-day figures ^[20]. The race figures are also far below 40% ^[6]. We read them as the same window, but the article does not label it.

Ariel Beccia, a researcher in the adolescent and children's medicine division of Boston Children's Hospital, said of the results: "It was extremely alarming seeing how high those numbers were" ^[4].

If all of these are three-day figures, Black students sit 4 points above white students and Hispanic and multiracial students 5 ^[15]. Beccia said the pattern counters the longstanding stereotype that eating disorders and disordered eating primarily affect white girls and young women ^[7]. Eating disorders affect everyone, she said, and "they affect marginalized youth more, actually" ^[8]. For binge eating, the data lean that way. The ranking is not tidy, though. Middle Eastern or North African students report the lowest rate of any group listed, 5 points below white students ^[16], and female students still run 3 points above male students ^[17].

The coalition that pushed for the item wanted to look at inequities partly because, Beccia said, youth facing food insecurity and economic strain are "more likely than their more advantaged peers to engage in binge eating" ^[12]. The article does not give group sample sizes, margins of error or rates adjusted for economic strain, so a 4- or 5-point gap cannot be assigned to race or to economic strain.

The item asked on how many of the past 30 days a student ate an unusually large amount of food in a short period and felt a loss of control over how much they were eating ^[10]. That is a behavior reported in one survey question. The thing this doesn't tell you is how many of these students would meet a clinical diagnosis.

The links to social media use, mental health and suicidality ^[9] come from a questionnaire answered in one sitting. In our view that design can show these things travel together and cannot show which came first.

The food-environment argument carries its own caution. Lindsey Smith-Taillie, a nutrition epidemiologist at UNC Gillings School of Global Public Health, said by email that "Many of these foods are engineered for overconsumption" ^[18]. Ashley Gearhardt, who co-authored two recent papers on ultra-processed food and addiction, pointed out that binge eating came to be recognized as a clinical problem in the same period that highly processed foods took over the food supply, but said the overlap does not establish causation ^[19].