Genes and BMI Part 1: Between-group variation and within-group variation are different things!

The Atlantic recently published a piece that is overall worth reading (and which I hopefully will write a separate blog post about soon), Rethinking the Calorie. It’s a relatively comprehensive look at the reasons why measuring “calories in” is not straightforward at all; it spends a lot of time on gut microbes in particular. One thing that they get wrong, though they are hardly alone in getting it wrong, is this:

Until recently, the idea that genetics plays a significant role in obesity had some traction: Researchers hypothesized that evolutionary pressures may have favored genes that predisposed some people to hold on to more calories in the form of added fat. Today, however, most scientists believe we can’t blame DNA for making us overweight. “The prevalence of obesity started to rise quite sharply in the 1980s,” says Nestle. “Genetics did not change in that 10- or 20-year period. So genetics can only account for part of it.”

The change in obesity rates between population does not imply that there isn’t a significant role for genetics in who gets fat. The amount of within-population variation is an order of magnitude different than the change in weight of the population between the 1980s and now.

Americans have gained about 30 lbs, along with about an inch of height, since the 1960s. But you can also find Americans with a three hundred lb difference in weight–100 lb people and 400 lb people both exist, and neither are all *that* uncommon. Speaking of that one-inch increase in height (or the ~4-inch increase in height that many countries have experienced over the last hundred years), that also doesn’t disprove that genetics play a significant role in height, and no one would seriously argue that it did, because differences between populations and the differences within populations don’t necessarily have the same causes.

For that matter, the differences in gut microbes and their association with BMI is all very interesting (sincerely!), but since they’re looking at differences between individuals in the same population and not differences between the gut microbes of, say, 1960s Americans and now, they’ve done nothing to show that gut microbes have anything to do with the significant changes in BMI at the population level known as the “obesity epidemic”.

This isn’t the first time that the difference between within-group variation and between-group variation has come up in a political context. It’s also come up in the context of race and intelligence, or more accurately, IQ tests, since we can’t directly measure actual intelligence. Performance on IQ tests has a significant heritable component; some people have tried to argue that this means that if black people as a group have lower IQ scores, it must be for genetic reasons. Those people are making a similar error, but in reverse–they are assuming that the cause of between-group and within-group differences must be the same, but they are taking a big component of within-group differences and assuming it must apply to between-group differences; in the case of BMI, people are also assuming that between-group and within-group differences must have the same cause(s), but they’re assuming that the “real” cause is the between-group cause.

An analogy I’ve seen used in the context of race and IQ is also useful here. Imagine two gardens with many varieties of corn growing in them. Within each garden, environmental conditions are the same. Each part of the garden is getting equal amounts of light, water, fertilizer, etc. The only reason the plants vary within each garden is because of their genes. But one garden gets optimal light, water, fertilizer, etc. and the other does not get the right amount of those resources. The difference between the gardens is not because of the plants’ genes, it’s because of their environment.

The concept is a bit counterintuitive, but it’s something I would expect scientists and experts on nutrition and weight to understand. And yet I see a surprisingly large number who don’t, even among those who are a bit smarter than to go along with “everyone knows” in other cases–for example, Barbara Berkeley. But definitely plenty from a certain food & nutrition writer who seems to be a very popular source of quotes and commentary for journalists.

This entry was posted in Uncategorized and tagged , , . Bookmark the permalink.

One Response to Genes and BMI Part 1: Between-group variation and within-group variation are different things!

  1. Pingback: Genes and BMI Part 2: Evidence for a genetic influence on weight in a post-‘obesity epidemic’ world | closetpuritan

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s