Sitting for too long can kill you, even if you exercise | Hacker News Reader

Sitting for too long can kill you, even if you exercise

cbc.ca

67 pointsmalingo45 comments

This is an observational study, and therefore has the usual "correlation does not imply causation" shortcoming. Maybe people with poor heart function or prediabetes don't enjoy standing as much, and tend to do it less (i.e. reverse causality from what the article suggests). Or maybe impoverished people don't have jobs that allow them to stand or move at regular intervals, and also don't have easy access to nutritious foods, causing poor health (i.e. a lurking variable).

I agree - there is a tremendously high risk of residual confounding in this study, even after adjusting for known/available risk factors.

One of the findings, for example, is increased risk of death from cancer. Since this is an observational meta-analysis, I am inclined to wonder whether an equally true assessment is "as people near death from cancer, they tend to be more sedentary".

I haven't yet logged in through the academic VPN required to gain access to this article and the articles from which its meta-analysis is derived, so I'll have to reserve judgment until I can do so. For now, it's a fascinating association but I'd be cautious about the interpretation of the direction of the arrow of causality.

Any further reading recommendations for concept of "correlation does not imply causation"? I always look for a concise way to describe this (there are just too many reports concluded in this way, I hope I could let others understand this point sooner to avoid being trapped in the details for too long if not forever) but have not been successful. Your words seems the best I can find.

I really like this video. He goes over the difference between clinical and observational studies.

https://www.youtube.com/watch?v=y1RXvBveht0

For a humorous look at what happens if we assume correlation causes causation there's Spurious Correlations: http://www.tylervigen.com/

My favorite: "Number people who drowned by falling into a swimming-pool correlates with Number of films Nicolas Cage appeared in"

With enough variables and noise you'll find things that correlate if you look hard enough. Without experiments where you control one thing and observe the other it's hard to demonstrate causation.

Every time I bring this concept up on HN, I get massive numbers of downvotes....I'm shocked that the parent comment is currently the top comment. Maybe it's me. Anyway, here's one of my favorite applications of the concept:

http://www.investopedia.com/terms/s/superbowlindicator.asp

Hm, this and some of the other comments being mentioned are the much less interesting instances of "correlation != causation". Specifically, these examples (including the Superbowl indicator) are assumed to be cases of coincidence (with enough data, certain slices of noise will look like signal).

By contrast, the much more interesting and relevant examples are things like hidden variables: a third factor that has causal relationships (in whatever direction) with both of the correlated variables being described, even though the two correlated variables have no such causal relationship.

For an extremely trivial example, the data likely supports the conclusion that you're more likely to get into car accidents when you have an umbrella with you. This obviously doesn't imply that having an umbrella in your car causes more car accidents: but rather that the hidden variable of rainy weather causes both "people have umbrellas with them" and "people get into more car accidents". Controlling for rainy weather (i.e. examining the correlation during sunny days only or rainy days only), you'd likely find that the correlation vanishes.

Thank you all for sharing your recommendations.

I guess the main reason you got downvotes with this concept is that, in most cases, you have different conclusion/insight from majority's. For this report, it's easier for us to come to similar ones. As an explanation for the possible issue for this report, users tend to upvote the parent comment as most of us accept it.

I've read quite a bit on standing desks and I'm curious if there is any study/research that shows, experimentally, that it's actually [standing] that leads to some of the proposed benefits. I only ask because I wonder if simply exercising while sitting [1] or in general [2] lead to the same benefits or if the very sedentary nature of the circumstances surrounding sitting that lead to these poor outcomes (which then are solved by standing).

[1] http://www.amazon.com/FitDesk-v2-0-Desk-Exercise-Massage/dp/...

[2] http://www.amazon.com/dp/B00B1VDNQA/ref=psdc_3407781_t3_B00C...

EDIT: Forgot to mention, when I say exercising, I mean during the same time where you would be sitting (e.g. during work, continuously)

Using a standing desk drastically decreased ambient pain in my hands, neck and shoulders. Highly recommend.

Definitely. If it's to be taken at face value, where is one dead person? "Died from sitting." Seriously?

I haven't seen a single shred of hard evidence of these FUD claims by the standing desk "cool kids" that are telling us "you're holding it wrong" and that we should do it their way "or it will kill you." C'mon. Trendiness maximus.

> Definitely. If it's to be taken at face value, where is one dead person? "Died from sitting." Seriously?

People don't die from any single cigarette, either.

(Your point might be true, but your argument doesn't pass muster.)

More people die in their beds than in their chairs, so clearly lying down is more dangerous than sitting.

People don't die from a single nights sleep either but in the long run it gets them.

Yes, exactly the point I was trying to make. Reductio ad absurdum.

I find that since I stopped sitting, I walk around much more. There may well be few benefits to standing per se, but it's certainly an excellent catalyst for movement and exercise.

I waste a lot less time at my desk when I'm standing. I've also experienced a reduction of neck and upper body issues.

But I definitely have to wear more comfortable shoes.

Get a standing mat off of Amazon. You will thank me.

So what are good measures for people that need to work in front of a computer, like to read a lot, etc. to avoid/minimize that risks?

So far I know of:

- Getting a standing desk (which is no alternative suitable for everyone)

- Walking around/exercise every ~30 minutes for a short while

When I was in an office, I made a habit of constantly drinking water, from a fairly small cup. Every time I needed to ponder something I'd reflexively wander off to get more water, then stare out the window a bit (apparently focusing on something far away once per hour or so reduces eye strain).

Having just started working at home, after a few months I'm feeling the loss of this habit.

My strategy is constantly drinking water from a large cup. Being adequately hydrated is only one benefit. The frequent walks to the restroom nay be even more important.

Ahh... the good old staring out the window with a drink in hand. My father invented that.

Thank heavens he didn't patent it.

....did he?

Don't worry.... the patent's expired.

Source article is behind a paywall [0]

Sedentary Time and Its Association With Risk for Disease Incidence, Mortality, and Hospitalization in Adults: A Systematic Review and Meta-analysis

Conclusion: Prolonged sedentary time was independently associated with deleterious health outcomes regardless of physical activity.

[0] http://annals.org/article.aspx?articleid=2091327

Some past items on this topic for consideration:

Sitting Time and All-Cause Mortality Risk in 222 497 Australian Adult [1]

Curb sitting time to protect aging DNA and possibly extend lifespan [2]

Television Viewing and Time Spent Sedentary in Relation to Cancer Risk: A Meta-Analysis [3]

Take a stand and be active to reduce chronic disease, make aging easier, research finds [4]

[1]: http://archinte.jamanetwork.com/article.aspx?articleid=11088...

[2]: http://www.eurekalert.org/pub_releases/2014-09/bmj-cst090114...

[3]: http://dx.doi.org/10.1093/jnci/dju098

[4]: http://www.k-state.edu/media/newsreleases/jan14/rosenkranz11...

Nowadays, the science news are more and more becoming this: http://www.phdcomics.com/comics.php?n=1174 or maybe this http://xkcd.com/882/ :P

Whatever doesn't make you immortal literally can kill you, even if you do other things.

I would like to see a study done that factors in the type of exercise. "Moderate to rigorous exercise" is a very broad definition. That could be anything from pumping iron for an hour to running six miles.

Standing is a weight bearing activity which has a great effect on bone density and muscle mass. These in turn effect one's long term health. Weightlifting can do more to compensate for the loss of lean body mass due to sedentary living compared to cardio. I'd be curious to see whether weightlifting is more effective than cardio in offsetting the effects of sitting most of the work day.

How about lying down, or even sleeping? The article mentions "sedentary time" and not just sitting in an upright position - presumably that includes sleeping too...

Moderation in all things including moderation excluding ricin, botulinum, radioactive polonium and militant extremism.

If you can't remember to get up from coding, there are timer apps that you can get to remind you to walk around. Besides, it is good for creativity and debugging to get away from the screen and go somewhere else, even if it is only to another part of the office or to stare out a window. In fact, staring out the window also relaxes the eye muscles which helps prevent vision problems and/or migraines.

While there is definite correlation with sitting and not being super healthy, going to other extreme, using standing desk all the time would also give you varicose veins.

Some healthy balance of being active and also being knowledge worker and smart Alec would definitely be best.

Dangit I need my drawing table back. Loved being able to stand or sit at my desk as needed.

relative risk increase != absolute risk increase, and I hate when news articles say "20% higher chance of X" without saying the absolute risk.

Correlation doesn't imply causation

I should be dead

"a higher risk of ... death"

is there really a higher risk of death? i think my chances of dying are already pretty close to 100%.

If you think brain uploading and/or gerontology will possibly produce a cure for death, your odds of dying may be somewhat less than 100% over some arbitrary finite but very long time span.

Even if you disagree with that, you can still parse the statement the way it was intended.

they didn't say a higher risk of "eventual death". At every moment in your life, you have a certain risk of death. When you reach 114 years of age, the risk is pretty high from moment to moment. Right now, fingers crossed, your risk is pretty low.

Too much sitting increases the average of the risks of death for some of the remaining moments of your life. Ha! Pedant your way out of that one!

yup, and skydiving without a parachute doesn't change your risk of death either since it is 100% anyway!

"higher risk of death" means on average people who engage in the behavior die faster.

Actually not 100%: http://www.greenharbor.com/fffolder/ffallers.html

I understand that the use of aircrafts in WW2 was incredibly high, but I would have thought maybe 1 person survived a free fall and lived, not 6(recorded on that site). Thanks for sharing that.

By my troth, I care not; a man can die but once; we owe God a death and let it go which way it will, he that dies this year is quit for the next.

Don't be a smartass, read the study.

They measure risk of death as 'your risk of dying in the next 12 months'

And if you really want to be a smartass, note that about 5% of all people have never died.

Reply on news.ycombinator.com