July 20, 2012
The Wall Street Journal
Sitting in Judgment of Sitting
By: Carl Bialik
Read the story below or view it on The Wall Street Journal
My print column this week examined a recent study of the dangers posed by sitting. The study was covered widely, including in The Wall Street Journal. But reports that excessive sitting could subtract two years from one’s lifespan — or, conversely, that sitting less could extend life by two years — may have been premature. The study didn’t establish definitively that sitting causes early death, only that the two are linked.
“We are just starting down the road of causality with sedentary behavior,” said Peter Katzmarzyk, lead author of the study, published last week by the online medical journal BMJ Open, and associate executive director for population science at the Pennington Biomedical Research Center in Baton Rouge, La. “Much more work needs to be done.”
This topic is gaining added attention this month with the start of this year’s Summer Olympics, a quadrennial event tied to public-health efforts to boost physical activity outside the Games. The Lancet, a U.K. medical journal, this week published a series of papers examining exercise and inactivity world-wide.
Ulf Ekelund, group leader of the physical activity epidemiology program at the MRC Epidemiology Unit in Cambridge, U.K., and co-author of one study used in Katzmarzyk’s meta-analysis, also thought other factors might explain sitting’s apparent dangers. “Many of the observed associations between sedentary time and health outcomes may be explained by poorly measured or unmeasured confounders,” Ekelund said.
Among these possible confounders are occupation, climate and health awareness, said Jessica Utts, a statistician at the University of California, Irvine. “As for potential confounding variables, there are probably so many that it would be impossible to measure and account for them all,” Utts said.
Part of the challenge with establishing that sitting causes death, said Donald Berry, a biostatistician at the University of Texas MD Anderson Cancer Center in Houston, is that a “a randomized trial of lifestyle is not possible. One cannot randomize Mr. X to spend much of his life sitting while Mr. Y has to get off his butt. The same problem exists in attempting to attribute poor health conditions to smoking. But we have extensive and highly credible biological evidence associated with the negative effects of smoking. With sitting all the arguments are ‘seat of the pants, ” Berry said. This critique comes despite Berry’s avowed bias against sitting, as when he isn’t traveling he rides his bicycle 20 miles a day — standing up.
Lutz E. Kraushaar, who blogged skeptically about the study, elaborated on why a controlled trial of sitting wouldn’t work. “You can’t blind the subjects, and you can’t get groups of people to adhere to a limited or extended daily sitting protocol for any meaningful period of time,” said Kraushaar, co-founder and chief scientific officer of the disease-prevention company adiphea GmbH in Baden-Baden, Germany. “But you would need to do this in order to compare, after many years, the effects of the ‘sitting’ intervention.”
“I don’t want to get nihilistic, but I also don’t like the idea of coming to definitive scientific conclusions based on poorly controlled observational studies,” added Charles B. Hall, professor of epidemiology and population health at the Albert Einstein College of Medicine in Bronx, N.Y.
However, Lance A. Waller, a biostatistician at Emory University in Atlanta, said it was instructive that many different studies came to similar findings about the health risks of excessive sitting. “Sitting for fewer hours per day over the course of a lifetime (accompanied by more exercise instead) likely will improve one’s chances of a healthier lifestyle but is not a guarantee that any particular individual will live longer,” Waller wrote in an email. The study’s reported cut of two years to lifespan was to average adult life expectancy in the U.S., not to any single American.
Skepticism about the latest sitting study shouldn’t be taken as a call for less exercise and more sedentary behavior while waiting for further results to come in. But caution is advised in changing how to spend time when not in motion, particularly because some studies of sedentary behavior lump sitting in with lying down, or count only television viewing and not other sedentary time. Without more definitive evidence about the health effects of different “doses” of breaks from sitting, public-health advice in the area “will remain general and tentative,” said David Dunstan, head of physical activity at the Baker IDI Heart and Diabetes Institute in Melbourne, Australia.
Sitting in Judgment of Sitting