For the video version, click here.

Childhood obesity is becoming a major problem in this country, and, like you, I am ready to blame anything except our sedentary lifestyle and carbohydrate-rich diet.  But according to a study appearing in the journal Pediatrics, it looks like day-care is off the hook when it comes to our children's waistlines.

Several studies in the past have documented a link between daycare and subsequent overweight and obesity in children:

The problem with these studies is that they were observational – and it is always possible that unmeasured factors, or confounders as the epidemiologists call them, are creating a link where none really exists.

Parents with less free time to cook healthy meals may also be more likely to put their children in day care, for instance.  A lot of these factors can be hard to measure and thus hard to control for.  So the question was open.

But this latest study, appearing in the journal Pediatrics, used some clever statistical tools to get to the bottom of the mystery, and that's what I'd like to discuss today.

Researchers used data from the Early Childhood Longitudinal Birth Study Cohort, a study that followed around 10,000 infants through the first 6 years of their life. About half of these kids ended up in non-parental childcare at some point in this period.

Just like prior studies, the researchers found that kids in non-parental child care had higher BMIs than those cared for by parents at pretty much every time point:

By kindergarten, 37% of kids who spent time outside of their parents’ watchful gaze were overweight and obese, compared to 33% of those who remained firmly under-wing.

But here's where it gets clever. The researchers then did something called a fixed-effects analysis.  This type of analysis uses each kid as his own control.  Many kids went in and out of daycare, so the researchers could see if, on average, the kid was heavier when he was in daycare than when he was not in daycare.  Think about weighing yourself before and after a Club Med vacation. This analysis accounts for a lot of confounders – provided one assumes that the other exposures (like mom and dad's cooking habits) were unchanged over time. And looked at this way, daycare had no effect on BMI.

The authors also performed an instrumental variable analysis which showed similar results. I'll pass on the details here, as that analysis rested on the assumption that proximity to family members is not related to body mass index and as someone who lives close to many family members I can 100% confirm that this is not the case. 

But regardless, this study makes a very compelling case that preschool is not a cause of obesity in this country. But who can we scapegoat now before we blame our own activity levels and diets? Well, there's always congress.