Today the American Academy of Pediatrics issued new guidelines that call for cholesterol screening in children as early as age 2 and no later than age 10. According to the guidelines, treatment with cholesterol-lowering statins should be considered for children as young as 8 with certain risk factors. I don't know enough about the risks and benefits of statins in children to say whether this is a good idea or not, and it sounds like research on the question is meager. But I was struck by the way the New York Times story about the guidelines conflates cholesterol levels with weight:

Proponents [of the guidelines] say there is growing evidence that the first signs of heart disease show up in childhood, and with 30 percent of the nation's children overweight or obese, many doctors fear that a rash of early heart attacks and diabetes is on the horizon as these children grow up....

Previously, the academy had said cholesterol drugs should be considered in children older than 10 if they fail to lose weight after a 6- to 12-month effort.

Shouldn't the decision to treat a patient for high cholesterol hinge on how high his cholesterol is, as opposed to how much he weighs? Even if there's a correlation between weight and cholesterol, doesn't it make sense to focus on the risk factor that's the target of the treatment, instead of simply assuming that all fat kids have high cholesterol and that failing to lose weight is the same as failing to reduce cholesterol?

Another, apparently related puzzle: By way of justification for more-aggressive use of statins in  children, Jatinder Bhatia, a neonatologist who serves on the committee that revised the guidelines, says, "We are in an epidemic." But between 1988 and 2000, according to the guidelines, "triglyceride concentrations [measured by the National Health and Nutrition Examination Survey] decreased approximately 8.8 mg/dL in adolescents aged 12 to 17 years, and total cholesterol, LDL, and HDL concentrations remained relatively stable." Furthermore, a comparison of NHANES data from 1988-1994 with data from 1966-70 found "a decrease in mean total cholesterol concentration of approximately 7 mg/dL" among 4-to-19-year-olds. Evidently, then, Bhatia is not talking about an "epidemic" of high cholesterol levels. Probably he means an "epidemic" of obesity (although recent data indicate that the upward weight trend in children and teenagers seen in the '80s and '90s has leveled off in the last decade or so).

This slippery switching between cholesterol and weight disguises the paucity of evidence that extra weight per se is unhealthy. NHANES data indicate that people in the "overweight" (but not obese) category actually have lower mortality rates than people in the "healthy"/"normal" category. Even for the BMIs that are correlated with shorter life spans, it's not clear how much people should worry about fatness, as opposed to the poor diet and sedentary lifestyle associated with it. This is especially true in the case of heart disease. "As near as I can tell," Barry Glassner reports in his 2007 book The Gospel of Food, "not a single published study demonstrates that heart disease among the overweight and moderately obese results from their heft rather than from other factors that contribute to both obesity and heart disease." Other critics of the conventional wisdom about weight, including Paul CamposEric Oliver, and Gina Kolata, make similar points.