The online science publication Ars Technica came out with a piece today entitled, “The new study suggesting sitting will kill you is kind of a raging dumpster fire.” The title alone fills me with glee! Debunking science! Using “kind of” in the title! And, it’s OK to sit again!!!!
In truth, the study today’s article alludes to completely freaked me out. It had me googling “treadmill desks” and envisioning doing squats during nighttime TV watching. But the sun has come out today — the Ars Technica authors poke enough holes in this study to put me back on good terms with our couch.
The particulars on the beef the authors have with the original study, which appeared in the Annals of Internal Medicine:
- The study was funded in part by Coca-Cola. The authors find fault with this saying that rather than focus on unhealthy consumption, Coca-Cola funds studies like these to shift the conversation to focusing on unhealthy behaviors. Personally, this doesn’t bother me that much. I feel like most people interested in heath — i.e., the people who are interested in these studies — already know drinking sugared sodas isn’t a great choice. The “conversation shift” certainly won’t lead me to begin chugging sugared soda after a workout.
- The study didn’t actually measure sitting. OK, getting warmer. Researchers did not rely on subjects’ reports of sitting times, as these are notoriously error prone. So, they used something called an Actical accelerometer, mounted to one’s hip, which tracks the amount of oxygen a person uses in various activities. Unfortunately, it cannot distinguish between sitting and standing. Or lying down and light movement. So, in this study, working at a standup desk is lumped into regular old sitting. As the authors point out, nuance is severely lacking here.
- The study demographics are problematic. A bunch of people dropped out early, leaving the more sedentary participants to complete the study which lasted a week only. Also, the more sedentary groups in the study were on average ten years older than the least sedentary groups.
- Statistical challenges. Once the data were in, the researchers broke the participants into four groups, those who:
- rarely “sat” (or did something else of low intensity), and when they did, “sat” in short bursts
- rarely “sat”, and when they did, “sat” for long durations
- “sat” frequently, in short bursts
- “sat” frequently, for long durations
The numbers of participants in the middle groups, as well as mortality rates, were not high enough to permit any statistically significant conclusions on the effects of sitting in short or long bouts. The researchers themselves cap off the study with a statement on the difficulty of drawing any conclusions due to issues with statistical significance.
The problem is that while this disclaimer appears in the actual research paper, how many of us read that? Rather, we tend to read the news summaries, which are often based off of summaries like this, which appears in Science Daily. And according to this summary,
- “A new study founds that sitting around for 12 or more hours per day, particularly if accumulated during 60- to 90-minute periods, increased the risk of early death — even in those who exercised.”
With no mention of the above limitations. So a big thank you to Ars Technica for probing deeper!