Hacker News new | past | comments | ask | show | jobs | submit login
The Men Who Stare At Screens (why going to the gym might not be good enough) (well.blogs.nytimes.com)
206 points by kscaldef on July 14, 2010 | hide | past | web | favorite | 106 comments

This is a warning sign: "Men who spent more than 23 hours a week watching TV and sitting in their cars (as passengers or as drivers) had a 64 percent greater chance of dying from heart disease than those who sat for 11 hours a week or less."

23 hours per week vs 11 hours per week? I call this a warning sign because it screams of post-hoc analysis. Unless they then used these as pre-specified cutoffs for later studies, I would be very cautious not to overinterpret.

It's always wise to suspect that sort of data-mining, but if (as I think but am not sure) your point is that they look like arbitrary numbers, suggesting that they were chosen by the researchers to give a good-sounding result -- well, let's express them differently: "24 hours or more" versus "less than 12 hours". It sounds a lot more canonical that way :-).

... Except that, looking at the abstract (there's a link in the NYT article), I find this: "Men who reported >10 h/wk riding in a car or >23 h/wk of combined sedentary behavior had 82% and 64% greater risk of dying from CVD than those who reported <4 or <11 h/wk, respectively." [I've changed one unhelpful bit of notation.] So, looks like data mining after all.

On the other other hand, as well as those (possibly cherry-picked) thresholds there are statements like "After age adjustment, time riding in a car and combined time spent in these two sedentary behaviors were positively (P(trend) < 0.001) associated with CVD death." which suggests something less ad hoc.

From the NYTimes article:

What was unexpected was that many of the men who sat long hours and developed heart problems also exercised. Quite a few of them said they did so regularly and led active lifestyles. The men worked out, then sat in cars and in front of televisions for hours, and their risk of heart disease soared, despite the exercise. Their workouts did not counteract the ill effects of sitting.

From the journal abstract:

In addition, high levels of physical activity were related to notably lower rates of CVD death even in the presence of high levels of sedentary behavior.

If anybody wants to look at the article, I have access to the paper.

I agree with you on all points.

From the abstract of the study:

Participants were 7744 men (20-89 yr) initially free of CVD who returned a mail-back survey during 1982. Time spent watching TV and time spent riding in a car were reported. Mortality data were ascertained through the National Death Index until December 31, 2003.

They noted that they adjusted for age. I presume other factors like the economic status correlate also with the time they watched tv. Don't know if they've adjusted for that as well.

Right, but my point is not about whether or not they adjusted for the right factors. Whether or not they did that, it is clear that they took a retrospective peek at the data and said, "Aha! If we make cutpoints at 11 hours and 23 hours, we get interesting results." Adjustment for age, which should be done, has no bearing on whether this post hoc slicing of the data to give good results is proper. This is only OK so long as you're intentionally exploring, and then perhaps doing a follow-up study using those cutpoints in prespecified fashion using fresh data.

In other words, if you went back and looked at your data and found the best cutoffs to maximize difference in risk, is it any surprise that you found a big difference in risk? This is why they need to validate these cutpoints in a separate study.

I'm afraid you use the post hoc fallacy incorrectly. The post hoc fallacy is drawing conclusions based on statistically insignificant data sets, normally anecdotes of size 1, at least in every day life. Now assuming this study was done by half way competent statisticians they would have followed the tried and true methods of establishing statistical significance. Sure they could have made mistakes on the way, the entire study could be completely invalidated by a detail or two. But even in that case it isn't a post hoc ergo propter hoc.

Your thinking could stem from an affliction brought on by high schools teaching of science. They really only teach one of the scientific methods, that of hypothesis, experimentation, observation and conclusion in that order. However depending on the field of science and the subject at hand that isn't necessarily the order thing are done in. Science isn't a simple linear process like we are lead to believe in high school.

I am speaking in the plain language used by those of us who do science for a living. I'm using the term post hoc in the way that we in science usually do (see http://en.wikipedia.org/wiki/Post-hoc_analysis if you're not familiar). I'm not sure how high school science teachers fit into this; sounds like you had a bad experience and I'm sorry if so, but that really has nothing to do with this discussion.

There is nothing wrong with "my thinking" but you appear to think so because your understanding is incorrect. Post hoc ergo propter hoc is the fallacy of thinking that, because something comes before another, that first something causes the second. I'm not talking about that at all. Post hoc analysis is something totally different - viz, it's performing non-prespecified analyses. It's more closely coupled to multiple testing than it is to the other fallacy you are referring to. It certainly is not limited to small sample sizes or sample sizes of one, in contrast to what you state. You can do science improperly with samples of any size.

I could be wrong, but at least in my little corner of science, this particular kind of post-hoc analysis, where you're moving thresholds on a linear scale for reporting purposes, seems not to be viewed with great suspicion, because it doesn't really produce arbitrary degrees of freedom. The worse kind of post-hoc analysis is where you can end up claiming things like "men who watched TV between 5.5 and 9 hours a week [whatever] more than men who watched TV either 1.0-2.0 or 11.5-12.5 hours/week", retroactively discovering a million probably random correlations in your data. Though even there there are ways of reducing the risk of doing so; for example, cross-validation is commonly used to test whether discovered correlations are artifacts of sampling or not, and has some pretty solid statistical theory behind it.

But in this case the general hypothesis is pre-chosen and fixed, basically "men who watch TV more hours [whatever] more than men who watch TV fewer hours", and the choice of threshholds is more of a reporting step, pulling out some binned tails at either extreme for emphasis and ease of discussion, while the real correlation is full-curve-to-full-curve.

I largely agree with you. I view moving thresholds with suspicion when you're trying to report effect sizes, but sure, I mean, I don't doubt the association. I doubt the magnitude, not the association per se.

And I absolutely agree that a sort of nonparametric post hoc approach that you poo-pooed would be even more highly suspect. But I don't see the choice of thresholds as simply a reporting step that can be manipulated ad lib without adjustment of your belief in the association and magnitude of effect. (I don't mean to imply that you think that you can adjust cutoffs willy-nilly - just stating my opinion clearly here.)

So they didn't take time spent sitting in an office into account at all? Seems odd. I spend at least two hours sitting at a desk for every hour I spend sitting on a couch or in a car seat.

It's interesting indeed.

Either they controlled for occupation (doubtful) or sitting in an office didn't matter or the thing whole is just off-base.

One way that sitting in an office is different from watching TV involves complete passivity whereas in an office setting, your actively engaged in some task.

But driving a car is probably closer to sitting in an office than watching TV. So the whole things seem problematic.

23 hours per week vs 11 hours per week? I call this a warning sign because it screams of post-hoc analysis. Unless they then used these as pre-specified cutoffs for later studies, I would be very cautious not to overinterpret.

What exactly are you saying? I honestly can't decipher the meaning of your words.

Sorry, I'm speaking a DSL I guess. I'm talking about the difference between a prespecified analysis (one that is planned in advance) and an analysis that was not planned in advance.

One of the problem with analyses that are not planned in advance is that they suffer from the problems inherent in multiple testing.

If I plan in advance to compare people weighing over 150 pounds to people weighing under 150 pounds, then I am conducting one test on the data. If I, instead, collect a bunch of data and run it through the ringer, torturing it until it spits out an answer that looks good to me, then I can find all kinds of spurious associations. They are more useful for hypothesis generation (i.e., generating a hypothesis that can be tested with other data).

The Wikipedia article does a better job of explaining this than I have:


I tend to expect that this sort of language is a bit of simplifying in the article, not necessarily an issue in the original study.

It'd be interesting to see how much gym workout is equivalent in benefits to the frequent light-intensity activities recommended in the article.

There are some pretty simple ways to add extra light physical activity to your daily routine. I tend to drink a lot of water, which means I need to go fill my bottle a few times a day and walk to the washroom more often than I would otherwise.

There was also a TED talk about communities in Italy and Japan with long lifespans and lifestyles that involved lots of minor activities

For me it brought to mind the discussions a few weeks ago about standing desks. Would simply working standing be sufficient?

Hmmm, except it's "common" knowledge in the waitress industry that being busy enough to have to walk around and do things frequently is way better than slow days when there's lots of standing around. standing around wreaks havoc on joints and stuff.

so....if you have a standing desk maybe the point is that you walk around more, whereas when one sits one is less inclined to get up for a minute walk.

I'm finding with my standing desk that having good shoe support and a good soft carpet/pad to stand on is quite important for that very reason. It does help me to walk around more. I tend to move around quite a bit more than when I was just sitting.

Have you tried using your desk barefooted?

Or wearing Vibram FiveFingers. That's what I do.

Might be hard in an office environment.

Depends on your company. I am alternating going barefooted and using the Vibram Fivefingers where I work.

(FYI, you can edit/delete comments if it has been less than two hours since you first posted them. Look for the edit and delete links on the right side of the comment's header line.)

I'm finding with my standing desk that having good shoe support and a good soft carpet/pad to stand on is quite important for that very reason. It does help me to walk around more. I tend to move around quite a bit more than when I was just standing.

Probably the same reason why men don't like shopping - too much standing around and not enough movement.

I tried a self-made standing desk for a couple days but went back to sitting because I couldn't take the standing for long periods of time.

I'm finding with my standing desk that having good shoe support and a good soft carpet/pad to stand on is quite important.

If you want to be extra sure, you could go for one of those treadmill desks


I would imagine the constant bobbing up and down would be even worse for your eyes...

I sit on a Swiss Ball, constantly bouncing up and down. It doesn't seem to have any impact on my typing or ability to read code.

Hunter-gatherers would be running around, tracking prey in the distance. Maybe the brain is wired to handle that well.

Tracking prey a hundred yards away is different from tracking source code 3 feet away.

Clearly the solution is to get a thin monitor and tape it to your cat or dog.

... and patent application filed. thanks!

I bought trekdesk (http://www.trekdesk.com) and I use it walking about 1.3 mph.

As a side note, I'm surprised that that site's FAQ appears to be advocating computer use on the treadmill but makes no mention of static discharge. In the winter, our treadmill becomes a giant Van de Graaff generator, unpleasant to use if you don't constantly ground yourself on a handrail. I gather this has not been a problem for you?

I just started using it. Maybe I can try the wrist straps used by electronics workers to protect against static buildup. We'll see.

I got a static wrist strap which helps. (Every day, not just in winter!)

Here's my desk: http://rwmj.wordpress.com/2010/01/25/treadmill-desk-part-6-h...

Exactly what occured to me in this context. I've been thinking about an ergonomic sit/stand desk for a long time. A friend of mine, a teacher, has bought one and praised it immensely.

Have a look at these products http://www.leuwico.com/web_e/products/office_furniture/ergod...

The height can be adjusted either manually (hydraulic system) or with a motor. It works better than it sounds. You can adjust the height even with a loaded desk easily.

The only question is: Would I remember to change my posture regularly?

Link to mentioned TED video: http://www.ted.com/talks/view/id/727

Summary: To find the path to long life and health, Dan Buettner and team study the world's "Blue Zones," communities whose elders live with vim and vigor to record-setting age. At TEDxTC, he shares the 9 common diet and lifestyle habits that keep them spry past age 100.

Short work outs will produce the same result if they are performed within same time span day after day. You can mix the hours up for each work out but let's be honest... that would be too much of a hassle.

I don't think article is very scientific; 'hitting the gym' can mean a lot of different things to a lot of different people. They can sit in the gym, do cardio, do weights, do stretches, etc. Too many variables. Obviously if subject goes to the gym and does 20 curls before calling it quits there will be absolutely no benefit.

If you hit the gym during the lunch break or before/after work and throw cardio/stretching in (not just weights) you will be fine. Or simply do some good stretching here and there; problem solved.

Sitting for long periods of time has an effect you can easily see in a lot of Americans. A fat ass, from all the blood pooling in their mid section.

There is one really large potential problem with this study: it is just a big survey. Often there is a discrepancy between what people say and what people actually do. What if many of the individuals polled simply lied, told half truths, or were mistaken about their times? It's already been proven that men lie about their incomes, why doesn't this extend to an active lifestyle as well?

For me to really take this seriously, these people need to be observed in some non-intrusive way. We do have the technology now, or so I think.

While I'm not an expert on these type of studies, I am married to one. I often see these types of criticisms bantered about, but I'm fairly certain that these researchers are MUCH more sophisticated in their controls than we give them credit for.

There is an entire body of research dedicated to ascertaining the quality and reliability of surveys, how surveys are delivered, and the conclusions that can be drawn from them. In doing these studies those methodologies are generally well understood (and clearly cited) as part of the paper itself.

While I know those of us who where educated in the 'hard' sciences (I'm a physicist by training) find it hard to buy into these... I've been quite impressed by how rigorous they really are.

I haven't read this particular study, so I may be way off base... but the fact that the fundamentals of the study involved a survey do not make it bad or invalid science (as long as the study was conducted properly that is).

This isn't about how sophisticated researchers are. The limits to research into real world human behavior are pretty much inherent in the complexity and interrelatedness of said human behavior. Doing statistics is extremely hard because behaviors and conditions correlate with a many things at once, making the causal factor very hard to isolate.

There have been a number of clever and rather specific ways in which research psychologists have gotten around these problems. But being an expert doesn't make you superman or give you unlimited bag of tricks dealing with this inherent difficulty.

One obvious way that the study might be flawed could be that the respondents were more honest about how much TV they watched than about how much working-out they did, so TV-watched actually correlated better with exercise level than claimed level of exercise (and that's just one of hundreds possible problems).

IE, until we know some specifics that researchers might have gotten around untruthness in research based on only survey data, we're entirely justified in being somewhat skeptical. "They're expert, they can do it" isn't enough.

I can understand the use of surveys when it's not realistic to actually measure this stuff, but in the 2-3 decades since this survey was done we now have technology like this:


There's no reason why this study shouldn't be updated with real data.

I've been thinking I'd like to get a standing workstation as it might encourage me to move around more, and I believe that even just to be standing most of the time is supposed to impact positively on heart health. Something like this would be awesome (though I imagine, pricey): http://www.amesystem.com.au/activdesk/

I love mine and highly recommend doing it. You can do it VERY cheaply too. I went with 2 Stanley FatMax telescopic sawhorses and a door.

Non-affiliate link: http://www.amazon.com/Stanley-11031-Telescopic-Plastic-Sawho...

Horribly outdated but accurate pic: http://icecreamforeveryone.files.wordpress.com/2009/03/img_0...

I feel more creative, have more energy throughout the day and my back feels stronger.

I did it much cheaper than you.

I used 4 stacks of paper for keyboard, an empty cardboard box for my mouse, and a pencil holder to tilt my monitor when I'm standing.

When my legs get tired, it takes 10 seconds to remove all of that and sit down again.

I stuck an old kitchen table on some cinder blocks back in May. It's great.


(Other than being a happy customer, I'm not affiliated with Indoboard.)

Don't you bump your head on that low ceiling all the time?

I think the roof is at a significant angle, standing 2 feel back from the screen it's probably 1 to 2 feet above head height.

It's an outdated pic, I quickly moved downstairs and took over our master bedroom. We sleep in the attic now.

I'm surprised that sleeping is still a group activity for you after that.

I love mine. I stacked two IKEA tables on top of each other. It turned out to be perfect height for me. YMMV.

I love love love standing while working on the computer. There was a transition period of 2 weeks for me of leg and foot soreness. Best of all, my pre-existing knee pain went away!!

I don't have a pic yet. But I will post it to my software blog sooner than later: http://bp321.com/code/

The IKEA Fredrik desk is adjustable and can be made into a great stand-up desk, claim many blogs that I researched when I first became interested in stand-up desks. Currently as I recently moved to a spacious apartment my computer is stacked on boxes. The reason I tried a stand-up desk was to get that intense focus, as I am now not easily distracted from a task when I work.

I like to remind myself, when I read about these kind of studies, that everybody dies from something.

I know some people, who haven't died, yet.

Why do these studies always avoid to talk about the intensity of the excercise for each subject. There's plenty of people at my gym that I've never seen sweating.

I wonder if this is because sitting and watching TV tends to correlate with poor diet (lots of snacking on unhealthy things while on the couch), whereas it's quite difficult to eat bad stuff while moving around all the time.

When I was a kid I used to skateboard around town about 8 hours a day, moving around all the time. I ate nothing but junk food during while doing this. What was I going to do, steam some brussell sprouts in the parking lot? It's more like existing in America tends to correlate with poor diet.

But on the other hand, you spent much more time skating around than eating, versus time on the couch is 100% available for mindless snacking.

I found just having access to high-quality ingredients at a reasonable price to be difficult in the US. More difficult than in Europe at least & waaay more difficult than in Asia (ok, one datapoint: China).

Having lived in the US, Europe and Japan and currently living China, I'd say that yes it's very difficult in the US, but in China it's rather difficult too... For example, I find it very hard to find good quality fish to cook (most of the fish you get here have a strange test I find)...

It's easy to find organic produce here but it's not that cheap (at least compared to Europe), and I'm not exactly sure if I trust it as much as I trust organic produce in Europe or Japan...

That said from my 6 month experience living in Rochester, US is much worse... (and don't talk to me about university "food" there)

I hear this all the time but when I lived in Japan and Germany I thought the food was expensive and the produce was of poor quality. Japan had cheap stuff like noodles, but vegetables and fruits were expensive. I've found the produce in the USA incredibly cheap and good most of the time.

IMO it is way easier to eat healthy in NYC than in any other place on the planet (and i've been to several).

NYC is technically not "America" though, so i'm not sure if I've raised a point of contention or simply bragged about how my city is teh awesum.

One of my first jobs was working at McDonald's. For a while I worked full time, 40 hours a week or more, all of it on my feet. I ate a LOT of their food, and this was back when they still cooked the french fries in beef lard.

During that time I never had any issues with weight gain, cholesterol, or blood pressure.

That depends. Sometimes you get stuck with 7 eleven food as a result.

I work at Ergotron, and we are building solutions to this very problem, affordable sit/stand workstations. There are numerous studies that appear to indicate our bodies are not built to sit for prolonged periods of time. Sitting may in fact be killing us early. Check out http://juststand.org

Do I get bonus METs if my legs never, ever, ever stop bouncing the entire time I'm at my desk?

I do remember reading that being twitchy can help you stay healthier, and lose weight. It all burns calories.

I recall another study that figured out people who are the gym who flex in the mirror all the time (hate those guys!) get more fit. That probably applies to desk jockeys as well.

Awesome. I'll pick up a mirror tonight, and start flexing into it while I work, shirtless.

I'm sure the coworkers won't mind.

Show them this study and tell them the solution is to flex in a mirror while you work. The goal is to get all of your coworkers doing this. Then you'll really have something.

I believe so. For me, I'm always squashing stress balls (not because I'm stressed), twirling pens around my fingers, or playing with rubber bands (in addition to bouncing legs).

I've always been on the thin side which people put down to high metabolism, but I suspect a lot of it is also to the fact I am in fact almost always in motion (even if only a small part of me).

> twirling pens around my fingers

Man, your fingers must be pumped!


It's satisfying to know that I've avoided wasting my time at the gym.

Instead, I go through life accepting the fact that I'll be dead by 80. I find it both motivating and relaxing. Instead of literally wasting 20% of my time at the gym, I can instead waste it in much more interesting ways: on games, programming, TV shows. I believe this will result in an overall healthier mental lifestyle. And since my body maintains a constant weight of ~215 pounds, and since I have no use for physical stamina or strong muscles, I literally have no use for the gym.

Try sitting on an exercise ball. That gets you plenty of muscle contractions just balancing on it.

A Knee* Chair is another cool way to sit and always be "in movement". Due to their design, you're forced to reposition yourself every few minutes or so.

*Technically you don't sit on your knees. You sit on your bottom (bearing most of the weight), and use your shins for balance.

What is it about improving health that brings out the most ridiculous concepts in fitness tool designs? This was advertised next to the article: http://i.imgur.com/PXsxO.png

What makes people want to spend money on tools like this when the greatest fitness tool is the human body itself?

I'm fairly sure that 90% of those wacky "exercise" ideas - like the hip-swivel chair - were created on a whim, or as a bet between two businessmen.

"Alright, we each have a month to come up with, manufacture, and advertise the most retarded exercise tool we can come up with. After six months of sales, we'll determine the winner by units shipped. Loser has to do a line of blow off the winner's genitals."

Well, that explains the Shake Weight..

That's the #1 reason I want me and at least one good friend to be awash in FU money.

Just another way to exploit the human weakness for easy answers. If you're fat, losing weight and getting in shape is going to take work. Either discipline to eat less, or to do real exercise (3 minutes a day using some gadget is going to do NOTHING), most likely both. TINSTAAFL.

Preferably the body of someone else.

(And to be honest, the ads for pure resistance workouts are even douchier.)

Huh, maybe I should just go into the tour guide business. I'd get lots of walking around here, probably eat better, and spend less time dicking around with bullshit code. I wonder what sort of unpleasant grief annoying tourists might cause... :-)

I have no idea, but I imagine it this way. Your users are incentivised to talk to you on a regular basis. You are never going to see a tourist again. If you hate users, you will LOATHE tourists.

I like talking with people, actually. I could definitely see walking around talking for half a day - it'd be much healthier than sitting on my ass staring at crappy code.

The study mentions nothing of eating habits; are we to believe each man (in 1982 no less) was eating a very healthy diet and all other factors were taken into consideration with regard to their lifestyle? For all we know they sat in front of the TV with a big bowl of ice cream, or ate a heavy meal in the evening before sitting down to relax. What about stressors, work or otherwise? Meh.

I wonder if this applies to people who have nervous energy and move around when at their desks... I've read about studies that correlate your amount of nervous energy (tapping your toes, etc.) to your likelyhood of being obese.

Maybe they're talking about movement on a different scale.

This smacks of the kind of news you see on The Today Show. I wouldn't be alarmed by this, we're a pretty adaptable species. Just stay active in your free time and you'll be fine. Sheesh.

Hmm... I wonder if this has any implications when it comes to school?

Absolutely it does...

1. The sit and watch - lecture model needs to die. 2. Students need to be given the option of sitting or standing in class and be allowed to alternate. 3. Punishments like sitting in detention, losing recess, or timeouts, should be replaced with activities like squats, sports, running laps, or simply walking.

The wisdom of taking a hyperactive kid's recess away has always baffled me.

Yes because people have sat in desks in school only since the 1990s and we do not know whether they die earlier or not, so we must trust some study.

They have been doing it for centuries and everybody is fine!.

"Everybody is fine", huh?

I'm not sure you can make that claim so casually.

Regular workout sessions do not appear to fully undo the effects of prolonged sitting

Well there go my health insurance rates (if I could afford insurance).

So when are some hacker news hacker going to make an iPhone app that uses to the gyro to track your movements through the day?

Sitting all day is probably bad for you. But everything productive I do is on the computer. How do I avoid sitting?

Stand: http://www.osha.gov/SLTC/etools/computerworkstations/positio...

It takes a little while to get used to it, but once you're there, standing all day is just fine.

Sit AND stand: No position for a long time is ever good for you.

You can stand and work around. You do not have to stand in the same position all the time.

Oops, "walk around".

Wonder if drinking green tea helps to bring you back to 'normal'..

Who knows what to make of this.. It's not easy hearing that the profession you've chosen puts you at greater risk for heart disease.

If this is true I bet it could be balanced by good diet. I also have a dog which forces me to walk for at least 1-2 hours a day.

This code is hardly rocket science. It's looks more like hard manual labour. But then again, so is also probably lot's of the code at bing and google to.

  my %holidays = (
    'mlk day' => undef,
    'martin luther king day' => undef,
    'martin luther kings birthday' => undef,
    'presidents day' => undef,
    'washingtons birthday' => undef,
    'memorial day' => undef,
    'independence day' => undef,
    'labor day' => undef,
    'columbus day' => undef,
    'veteranss day' => undef,
    'thanksgiving' => undef,
    'fathers day' => undef,
    'mothers day' => undef,
    'inauguration day' => undef,
    'ash wednesday' => undef,
    'shrove tuesday' => undef,
    'groundhog day' => undef,
    'saint patricks day' => undef,
    'easter' => undef,
    'earth day' => undef,
    'arbor day' => undef,
    'cinco de mayo' => undef,
    'st patricks day' => undef,
    'flag day' => undef,
    'rosh hashanah' => undef,
    'yom kippur' => undef,
    'halloween' => undef,
    'election day' => undef,
    'hanukkah' => undef,

Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact