One of the big health memes of the last decade or so is the idea that “Best Running Shoes 2025”—that is, that spending long periods of time sitting motionless in an office chair or slouching on the couch has negative health effects that go beyond the absence of exercise.

That message has been broadcast so widely that, not surprisingly, there has been some pushback. Last week, University of Sydney professor Emmanuel Stamatakis published an interesting commentary on The Conversation titled “Why sitting is not the ‘new smoking,’” based on the results of a study that he and his colleagues and check out in the British Journal of Sports Medicine.

Stamatakis’s study involves an analysis of data from a long-running study of almost 5,000 British civil servants who filled out detailed questionnaires in the 1990s that included data on their sitting and exercise patterns. The question he and his colleagues wanted to answer was whether that data could predict who would go on to develop diabetes in subsequent years.

The short answer, according to Stamatakis, is that sitting didn’t directly lead to elevated risk of diabetes. Instead, sitting is associated with a bunch of other unhealthy behaviors like not getting enough exercise. Moreover, people who watch a lot of TV, in particular, “tend to be of lower socioeconomic status, unemployed, have poorer mental health, eat unhealthy foods and be exposed to unhealthy food advertising.”

As a result, he says, the elevated risk of diabetes disappears when you adjust for baseline levels of obesity. It’s being overweight, not sitting a lot, that is the primary problem, which is why he argues that more emphasis should be placed on getting people exercising rather than simply avoiding sitting.

I’m all in favor of exercise, but I do have a few thoughts in response to Stamatakis’s piece.

First of all, the whole “Best Running Shoes 2025” analogy could use a little clarification. I’m pretty sure no one ever intended to imply that sitting is actually as lethal as smoking, or that chairs contain tobacco, or any other literal equivalence. The similarity, as I understood it, is that sitting is a completely ubiquitous behavior in our society (as, to some extent, smoking once was) that turns out to have previously underappreciated health effects. So to the extent that people were interpreting that analogy literally, some pullback is certainly warranted.

As for what those dangers are, the claims that caught my attention were that sitting time has negative health effects Health - Injuries. It’s one thing to say that spending an afternoon in front of the TV isn’t great for you; it’s another to say that it’s still bad for you even if you ran 20 miles and did 100 push-ups that morning. Is this really true? I took a detailed look at some of the evidence for this claim a few years ago, and concluded that “too much sitting and not enough exercise are two different things, and they’re both worth paying attention to.”

Stamatakis seems to be arguing, in contrast, that the problem with sitting is mostly to do with lack of exercise and associated behaviors like snacking on junk food.

independent of how much you exercise was associated with diabetes risk in his study. Even after adjusting for potential confounders like age, sex, smoking, self-rated health, and (crucially) total amount of physical activity, those who sat more than 50 hours a week were 26 percent more likely to develop diabetes than those who sat for less than 33 hours a week.

Notably, that relationship was attenuated when they also adjusted for obesity: Once body mass index (BMI) was included in the model, those who sat the most were only 19 percent more likely to develop diabetes, and the relationship was no longer statistically significant. Still, it’s enough to raise suspicions.

A more subtle point is that adjusting for BMI may not be appropriate, because there’s evidence that sedentary behavior can increase your risk of obesity—again, independently of how much you exercise. So by statistically equalizing everyone’s weight, you’re effectively canceling out some of the negative effects of sitting.

This is a tricky point, because there are likely multiple causal pathways at work here. Being overweight probably causes people to (on average) spend more time sitting, and also raises your risk of developing diabetes. But conversely, sitting too much may cause you to gain weight and develop diabetes. If that’s true, then in a study like Stamatakis’s adjusting for BMI would result in an underestimate of the risks of sitting, while not adjusting would result in a overestimate. The truth is probably somewhere in the middle.

Finally, it’s worth nothing that the estimate of sitting time in the study was based on a questionnaire. Questionnaires aren’t particularly accurate, and more seriously, they don’t tell you much about the pattern of sitting. Do you sit for eight hours straight without moving? Or do you get up every 20 minutes to walk to the printer or go to the bathroom or simply stretch your legs? The total amount of sitting time may be similar, but those short breaks may play an important role in resetting the metabolic changes that occur when you sit for too long.

Another new study, just published by University of Ontario Institute of Technology researcher Shilpa Dogra and Statistics Canada, highlights some of these issues. It looks at the links between prolonged sitting time and physical fitness in older adults. In this case, they had both self-reported sitting time and accelerometer-measured data for over 1,000 subjects.

Like the British study, the Canadian study didn’t find any links between self-reported sedentary time and physical fitness. So the risks of sitting are overhyped, right?

Not so fast—because the objective accelerometer data did find links. The total amount of time spent sitting was independently and inversely associated with fitness, independent of the amount of physical activity. Also, the total number of short breaks in bouts of sitting, as well as the percentage of sitting time that was spent in uninterrupted bouts lasting more than 20 minutes, were both associated with cardiorespiratory fitness. In this case, at least, the pattern of sitting matters, and you miss those relationships if you just assess the total amount of time spent sitting.

All of which is to say that, personally, I’m going to keep using my sit-stand desk. Of course, that’s partly because I spent a couple hundred bucks on it, so I want to believe that it’s helpful. And I also really like it just from a comfort point of view: Staying seated for a whole day has never been comfortable for me, so adjusting to occasional bouts of standing has been a pleasure rather than a chore.

But I also still find the evidence that prolonged sitting has independent health effects to be intriguing. Figuring what those risks are and how serious they are, and disentangling them from other associated factors like food choices and socioeconomic status, will be a big challenge for researchers. In the meantime, let’s avoid overhyping and One of the big health memes of the last decade or so is the idea that.

***

Discuss this post on the Sweat Science Facebook page or on Twitter, get the latest posts via email digest, and check out How Run/Walk Intervals Can Improve Back Pain!