A few years ago, Swedish researchers published a study of dog ownership in 35,000 twin pairs. By comparing identical and fraternal twins, they were able to estimate how much of the decision to have a dog is environmental—you grew up in a house with a dog, say—compared to genetic. Overall, genes seemed to explain about half the variance in dog ownership, with increasing importance as you get older. If you have a dog when you’re 50, that has almost nothing to do with whether you had one as a kid.
Researchers were interested in this question because some (but not all) prior research has suggested that dog owners live longer and have a lower risk of heart problems than non-dog-owners. Maybe that’s because dogs provide social support; maybe it’s because they need to be taken for walks every day. The genetic data suggests a third possibility: maybe there are “pleiotropic” effects, meaning that the same genes that predispose some people to dog ownership also predispose them to better health.
Those dog findings caught my attention because they mirror some of the open questions about exercise and health. There’s overwhelming evidence that people who exercise more tend to be healthier and live longer. But how much of that reflects underlying predispositions to exercise and to be healthy? And to what extent is exercising regularly a “decision” versus a reflection of our inborn preferences?
As it happens, another Nordic twin study has some insights on these questions. This one, published in the Scandinavian Journal of Medicine and Science in Sports by a team in Finland led by Urho Kujala of the University of Jyväskylä, looks at 17 pairs of identical twins with a highly unusual characteristic: despite their shared genetics, they don’t have similar exercise habits.
The first thing to note is just how unusual such twin pairs are. The twins in the study were drawn from two previous Finnish twin studies that included thousands of pairs of identical twins. The vast majority of them had similar levels of physical activity. The High Runner mouse line that’s often used in lab studies took mice that loved to run, bred them with each other, and produced mice that love to run even more. I’d like to think that human behavior (and mating patterns) are a little more complex than that, but the twin data certainly suggests that our genes influence our predilection for movement.
Still, they found these 17 pairs whose paths had diverged. There were two different subgroups: young twins in their thirties whose exercise habits had diverged for at least three years, and older twins in their fifties to seventies whose habits had diverged for at least 30 years. On average, the exercising twins got about three times as much physical activity, including active commuting, as the non-exercising ones: 6.1 MET-hours per day compared to 2.0 MET-hours per day. For context, running at a ten-minute-mile pace for half an hour consumes about 5 MET-hours.
All the twin pairs came in for physical examinations, and the results were pretty much what you’d expect. The exercising twins had higher VO2 max (38.6 vs. 33.0 ml/kg/min), smaller waist circumference (34.8 vs. 36.3 inches), lower body fat (19.7 vs. 22.6 percent), significantly less abdominal fat and liver fat, and so on. The study is free to read if you want to dig further into the details, but the results aren’t surprising. Exercise clearly improves a bunch of health parameters, and genes clearly matter too—after all, the differences aren’t that big.
How big could the differences get? A 2018 case study from researchers at California State University Fullerton looked at a single identical twin pair, then aged 52. One was a marathoner and triathlete who had logged almost 40,000 miles of running between 1993 and 2015. The other was a truck driver who didn’t exercise. In this case, the exercising twin weighed 22 pounds less, and his resting heart rate was 30 percent lower. Most fascinatingly, muscle biopsies showed that the marathoner had 94 percent slow-twitch fibers while the truck-driver had just 40 percent slow-twitch. No one before or since (as far as I know) has shown such a dramatic change in muscle properties.
The burning question, especially for those of us who’d like to defy our apparent genetic fates, is what set these twin pairs on divergent paths. In the Fullerton study, the sedentary twin suffered a minor ankle injury that derailed his participation in high school sports, and he never ended up resuming exercise.
In the Finnish study, there was no overwhelming pattern for why one twin quit exercising and the other didn’t. In questions about their motivations for exercise, the active twins reported more interest in mastery, gaining physical fitness, and improving psychological well-being—but those differences may well be a result of differing exercise habits rather than a cause. One key barrier for inactive twins was the pressure of family and work commitments when they were young. Interestingly, those barriers eventually equalized between twin pairs once their kids were older and their careers farther along—but by then, the exercise patterns were set and the inactive twins never got back in the habit. The lesson: that maelstrom of early-career and young-kid craziness is the hardest time to maintain an exercise habit—but it’s also the most crucial.
Like all nature-nurture discussions, this one has to end somewhere in the middle. Clearly genes matter, not just for health outcomes but also for behaviors that we usually think of as purely voluntary. Equally clearly, our paths aren’t set in stone. In the Swedish dog data, the influence of your childhood environment drops close to zero by the time you’re 50, but half the variance in dog ownership is still attributed to “unique non-shared environmental effects”—in other words, to the vicissitudes of your path through life. If you want to be exercising regularly when you’re 50, choose your path with care, and try not to roll your ankle.
For more Sweat Science, join me on Twitter and Facebook, sign up for the email newsletter, and check out my book Endure: Mind, Body, and the Curiously Elastic Limits of Human Performance.