Think yourself thin…

August 23rd, 2008 by Ben Goldacre in bad science, placebo | 33 Comments »

Ben Goldacre
The Guardian,
Saturday August 23 2008

What I particularly enjoy is the spectacle of fat people – ideally drinking beer – watching television, while somewhere on the other side of the world citizens of all nations are getting some nice exercise in the Olympics (throwing javelins, jumping over metal bars, climbing lamp posts with banners, and running away from the water cannon). These are the people I imagine paying for gyms they never visit, while I am cheerfully cycling to work and carrying the shopping up the stairs.

But can obsessing over sport actually improve your health? Slightly, possibly, if you’ve got something to work with.

image Alia Crum and Ellen Langer from Harvard psychology department took 84 female hotel attendants in 7 hotels. They were cleaning an average of 15 rooms a day, each requiring half an hour of walking, bending, pushing, lifting, and carrying. These women were clearly getting a lot of good exercise, but they didn’t believe it: 66.6% of them reported not exercising regularly, and 36.8% said they didn’t get any exercise at all.

Their health, measured by things like weight, body fat, body mass index, waist-to-hip ratio and blood pressure, was related to their perceived amount of exercise, rather than the actual amount of exercise they got, and this, so far, isn’t very unusual. A classic study of 7,000 adults found that that perceived health is a better predictor of death than actual health, and another looking at elderly people found that those who perceive their health to be poor are 6 times more likely to die than those who perceive their health to be excellent, regardless of how healthy they actually are. Once again this goes to show the danger of relying on self-report data for health research.

But it gets better. Crum and Langer then divided the hotel workers into two groups (by hotel). One group got a one hour presentation on what a fabulous amount of exercise they were getting, how they were meeting and clearly exceeding the Surgeon General’s recommendations for an active lifestyle. They were given information sheets, in English and Spanish, showing the calorie burn for activities like vacuuming, or cleaning a bathroom, and the researchers even put notices up in communal areas explaining what excellently healthy exercise their work was.

The other group was left alone.

Four weeks later the researchers measured everything again. The group who had been tutored about the health benefits of their work now perceived that they did more exercise than before – unsurprisingly – while the group who were left alone didn’t change. Neither group had changed their actual levels of activity.

But amazingly, despite no change in actual exercise levels, in the intervention group, simply being told about the value of what they were already doing caused a significant change for the better on every single one of the objective health measures recorded: weight, body fat, body mass index, waist-to-hip ratio and blood pressure.

It’s an outrage. Maybe mindset alone can influence metabolism and the benefits of exercise: perhaps this experiment shows, essentially, the placebo benefits of exercise. Maybe the cleaners changed their behaviour, or their diets, in ways that the researchers didn’t pick up, perhaps they had more spring in their step, tipping the scales in their favour. And maybe it doesn’t actually matter what caused the change, as long as we can exploit it: because the links between body and mind are far more fascinating than any pill peddler would ever have you believe.

References:

Crum AJ Langer EJ. Mind-Set Matters: Exercise and the Placebo Effect. Psychological Science 18(2) 165-171, 2007.

dx.doi.org/10.1111/j.1467-9280.2007.01867.x

Kaplan, G.A., & Camacho, T. (1983). Perceived health and mortality: A nine-year follow-up of the Human Population Laboratory cohort. American Journal of Epidemiology, 177, 292–304.

Idler, E., & Kasl, S. (1991). Health perceptions and survival: Do global evaluations of health status predict mortality? Journals of Gerontology, 46 (2), S55–S65.


++++++++++++++++++++++++++++++++++++++++++
If you like what I do, and you want me to do more, you can: buy my books Bad Science and Bad Pharma, give them to your friends, put them on your reading list, employ me to do a talk, or tweet this article to your friends. Thanks! ++++++++++++++++++++++++++++++++++++++++++

33 Responses



  1. dw said,

    August 23, 2008 at 12:21 am

    “A classic study of 7,000 adults found that that perceived health is a better predictor of death than actual health, and another looking at elderly people found that those who perceive their health to be poor are 6 times more likely to die than those who perceive their health to be excellent, regardless of how healthy they actually are. Once again this goes to show the danger of relying on self-report data for health research.”

    If anything, I should have thought that it would show the opposite.

  2. ChrisP said,

    August 23, 2008 at 8:10 am

    Fascinating.

    Could it all be in the mind, or at least the brain? The brain uses about 25% of our total metabolic energy. An active brain more than a depressed brain. Could the improvements be the result of one group being mentally stimulated by all the positive attention?

    For this reason, I’m not very impressed by the choice of control group! They should have had a third control group who were given lectures on how good their work was, or something equally stimulating but unrelated to exercise.

    Chris

  3. drunkenoaf said,

    August 23, 2008 at 8:41 am

    It would be awesome if there was a LTFU– 4 weeks is fine, but my mental kick for losing weight– a new year’s resolution lasted until April. Then I stopped jogging again. It would have been far more informative to have tracked the duration of the effect, dagnammit!

  4. Galen said,

    August 23, 2008 at 10:31 am

    Oh, come on now, Doctor.

    Neither group had changed their actual levels of activity.

    Are we sure? How do we know? Did anything else change apart from levels of activity? Diet for example.
    One group had an intensive presentation on health and then had reminders posted up around the workplace. Could this have had no effect; in decisions to use the stairs, rather than the lifts or to choose fruit instead of ice cream, for example?
    How was the presentation understood (not all the audience were English speakers)? Did a subtle message go out that the hotel prided itself in the fitness of its staff, and that any cleaner who did not lose weight would be sacked? More to the point, do we know how the presentation was received?

    despite no change in actual exercise levels, in the intervention group, simply being told about the value of what they were already doing caused a significant change for the better on every single one of the objective health measures recorded: weight, body fat, body mass index, waist-to-hip ratio and blood pressure.

    So something must have changed, mustn’t it? Something very significant too, to be measurable within four weeks.

  5. projektleiterin said,

    August 23, 2008 at 1:10 pm

    In just four weeks *significant* changes in weight, body fat, body mass index, waist-to-hip ratio and blood pressure??

    I joined a gym in May and since then have gone exercising on average every second or third day and while I noticed I’m a bit fitter it’s not what I had hoped for. I’d also like to get some of the placebo effect, looks like a lot less work to me!

  6. Irene Adler said,

    August 23, 2008 at 2:48 pm

    Interesting. I bet they changed their diets slightly. There’s nothing like feeling that you’re getting a decent amount of exercise to give you pure thoughts, resulting in choosing an apple over a chocolate bar.

  7. muscleman said,

    August 23, 2008 at 11:38 pm

    Comfort eating, population of women. Need I really say any more?

    Thinking about it after watching my wife and daughters obsess about, agonise over then finally make and consume a chocolate brownie (with mixed nuts and white chocolate bits). This was two days after I made the youngest a shepherd’s pie as comfort food when she was feeling poorly.

    No change in metabolic activity is really necessary, though I’m sure happy people are more incedentally active than depressed people.

  8. muscleman said,

    August 23, 2008 at 11:46 pm

    projektleiterin, you and I are clearly different. I have two distinctly different modes: couch potato and athlete. The former is characterised by lassitude, snacking. backache, digestive upset. motivation problems. After two runs (gyms? pah!) with two days in between my waking pulse will have dropped 10bpm, I will be more alert and find myself wondering what I can do and my appetite reduces. I also have a greater taste for tea (black) and my coffee consumption decreases.

    I have done this experiment numerous times and it is perfectly reproducible. It is like a switch is thrown and I am literally in a different mode. Most of the changes are fairly subjective, except for the pulse, that one is the freakiest. I only need run every 4th day on average to maintain this state.

  9. surdin said,

    August 24, 2008 at 8:28 am

    My grandfather, used to really worry about his blood pressure and was taking a lot of medicine, and still it didn’t stabilize normally, etc.(though earlier in his life he worked in nature, and expeditions, and was of perfect health), until they found a tumor in his belly, then he had health problems related to that(infection after surgery and so on), but his blood pressure staid at 120/80 all the time.

  10. cannon said,

    August 24, 2008 at 10:48 am

    “…who perceive their health to be poor are 6 times more likely to die than those who perceive their health to be excellent,…”

    I’m no scientist but I’d have thought the chance in both cases was identical – 100%

  11. Getonyerbike said,

    August 24, 2008 at 8:42 pm

    “And maybe it doesn’t actually matter what caused the change, as long as we can exploit it: because the links between body and mind are far more fascinating than any pill peddler would ever have you believe.”

    Precisely. Well almost, obviously it does matter, we need a far more sophisticated understanding of what’s going on. In Ben’s case above, informing the cleaners about what they were already achieving gave them hope in a situation that they percieved to be hopeless and hence triggered other changes.

    I’m looking forward to the development of the placebo discussions and research to find ways of making modern medicine more effective than ‘simple’ NICE approved, evidence based prescribing

  12. DSThomas said,

    August 24, 2008 at 9:35 pm

    I think that its likely there was a change in diet that wasnt picked up here.

    To give an example i do alot of cardio based exercise (i.e. the stuff that makes u lose weight, running swimming etc) yet generally eat a crappy diet so my weight stays constant. However if i get it in my mind to try n eat healthy n ditch the pizza’s for a week or so i know from experiance that i can drop around half a stone in weight without too much effort. this will generally happen after someone points out im looking less fat than normal :p I know i wont keep the weight off for more then a month or two but the push is enough to make me do it and thats what i think is happening here

    I will admit i was too tired to read the actual article before posting this so if im missing the point ignore me

  13. underblog said,

    August 24, 2008 at 9:49 pm

    “A classic study of 7,000 adults found that that perceived health is a better predictor of death than actual health, and another looking at elderly people found that those who perceive their health to be poor are 6 times more likely to die than those who perceive their health to be excellent.”

    It’s not just that people are better at determining their own level of health than scientists and doctors? How can you control for this?

  14. Robert Carnegie said,

    August 25, 2008 at 1:29 am

    Since Ben liked this paper enough to write about it, I’m going to trust in his authority and assume that it is not a lot of bollocks with gaping holes iin the method, although strictly this is the wrong thing to do. Of course it could be entirely made up, peer review doesn’t test for that; it’s hard to tell but apparently not impossible. But if it’s not that, then something changed, even if they didn’t detect a physical cause; perhaps thinking of yourself as an athlete has physiological effects. Thinking of yourself as someone who has just missed a bus because it left two minutes early, does have short-term physiological effects!

    For that matter, thinking that you’re about to undertake a day of paid aerobics may set you up to get the benefit of it, or to avoid being damaged by it. I suppose our ancestors had behaviours of leiisurely gathering of fruit and nuts, and running like heck when the sabre-tooth tger wants to join the party. No doubt each has its own metabolic mode, but perhaps actually takes its cue from the mind.

  15. Ben Goldacre said,

    August 25, 2008 at 1:52 am

    evening all. i did mention the possibility that people might have changed their health behaviour in a way that wasn’t picked up by the researchers, but there is actually another flaw which someone emailed in and which i didn’t spot, which is that the researchers seem not to have corrected for clustering in their data. i explain what that means in this older column here on cocaine use “doubling” in the playground:

    www.badscience.net/?p=230

    to my late sunday skim this looks like a pretty valid criticism, and might reduce the statistical significance of the findings, without the raw data nobody can tell by how much. i’ll be interested to see if this gets taken up in the journal with the authors.

  16. rogerhyam said,

    August 25, 2008 at 11:49 am

    What makes this and the placebo stuff interesting is our oversimplification. We draw an imaginary line through a complex system to divide body from mind then we are amazed that things occurring on one side of that line appear to be related to things occurring on the other side of the line.

    The amazement is caused by drawing the line not the observable facts.

    I am naive in this area but I can’t think of a falsifiable experiment to demonstrate the existence of mind in a subject so perhaps we should never, within science, talk of a relationship between mind and body any more than we should talk of a relationship between body and soul or soul and God.

  17. curlyfries said,

    August 25, 2008 at 3:25 pm

    I don’t have a copy of the paper yet but I should at least point out that Ellen Langer believes that all health conditions, including AIDs and Cancer, are caused entirely by our belief in them. She got into trouble for reusing needles in an experiment:

    www.thecrimson.com/article.aspx?ref=505277

    To her, this wasn’t an issue as blood diseases don’t really exist. She is generally regarded as a complete crackpot at Harvard.

  18. Jamie Horder said,

    August 25, 2008 at 9:45 pm

    curlyfries : Do you have a reference for her beliefs on AIDS and cancer etc.?

    I found this article :
    psychologytoday.com/articles/index.php?term=20000701-000028&page=1

    In which she says that our beliefs about disease, such as breast cancer, can (directly or indirectly) affect our health. But she never says that disease is caused by our belief in it. It looks like fairly standard “Think positive!” advice to me.

  19. Twm said,

    August 26, 2008 at 3:52 am

    This is a really interesting story, thanks Ben.

    I have seen the behavioural pattern below in family and friends:

    After doing exercise, many tend to stay off the sugary foods/deserts for fear of undoing the good work of exercise.
    But perversely, at times when they feel that they are not getting enough exercise are more likely to indulge perhaps in the belief that their health drive will start the following week.

    Convincing workers that they are exercising vigorously is likely to influence self esteem and dietary choices(small changes such as putting less food on the plate might be hard to detect).

    The other factor is how it changes the workers’ attitude to their job.
    A worker who sees an extra benefit from the job in the form of exercise may ‘work out’ more effectively – modify how they do their job (e.g hoover more vigorously).

    And finally, a worker who feels knackered and sore after a shift, instead of feeling achy and in need of a sofa and comfort food, may when informed feel a sense of reward from the aches and take better care of their diet and maybe take other forms of exercise (such as walking to shop rather than taking the car).

  20. sam_chew75 said,

    August 26, 2008 at 1:31 pm

    “and another looking at elderly people found that those who perceive their health to be poor are 6 times more likely to die than those who perceive their health to be excellent.”

    Quick pedantic point: isn’t everyone equally likely to die?

  21. cannon said,

    August 26, 2008 at 2:25 pm

    Comment 26:

    “Quick pedantic point: isn’t everyone equally likely to die?”

    Check comment 13. Wonder what comment 39 will be about.

    And no response from the guru about bad use of numbers

  22. Raelor said,

    August 26, 2008 at 4:40 pm

    Comment 27:

    Firstly, you were beaten to your hi-fucking-larious point by eyalbd at number 5.

    Secondly, it’s pretty clear that Ben meant ‘likely to die during the period studied’.

    Perhaps your crusade for accurate writing would be better directed against those who deliberately fudge numbers in order to mislead people and not against the author of a web blog which is, well… on your side.

  23. chltx said,

    August 27, 2008 at 7:10 pm

    “If anything, I should have thought that it would show the opposite.” -dw

    All of the interesting bits of science start out with, “hmmm… that’s strange…”

  24. chltx said,

    August 27, 2008 at 7:24 pm

    Other observations:

    1) Dietary recall is notoriously poor. So much so that unless you have detailed contemporaneous diet logs, dietary recall is essentially worthless. Even contemporaneous diet logs are only reliable if they are kept by somebody *watching* the subject, and not by the subject. If you are not weighing and measuring, you are guessing, and your guess is no better than anyone else’s.

    2) Most of the ‘studies’ I have read on nutrition would not get a passing grade in a rigorous science course. My favorite along that line was a study (in a peer-reviewed journal, no less) in which the ‘researchers’ removed the sugar from the diet of one group, and concluded from the observed health improvements that animal protein was bad for diabetics. I kid thee not. Modern nutritional education is not far enough removed from religion, and generally suffers from massive confirmation bias.

  25. thegrouchybeast said,

    August 28, 2008 at 12:35 pm

    another looking at elderly people found that those who perceive their health to be poor are 6 times more likely to die than those who perceive their health to be excellent, regardless of how healthy they actually are. Once again this goes to show the danger of relying on self-report data for health research.

    Soooo…

    People’s perceptions of their own state of health are a better prediction of their risk of death than whatever collection of surrogate markers are being used by the study to determine ‘health’. And yet, somehow it’s the people who are wrong, not the markers? I know that when a model fails to match reality, normal practise is for scientists to blame reality, but mostly they manage to be a little less blatant about it.

    (And, in passing, I tend to regard any study which uses a lower BMI as a marker for ‘health’ as automatically suspect. Overweight/obese BMI categories are consistently associated with a lower risk of death. I hope that Bad Science knows better than to buy into the trend for aesthetics-based medicine.)

  26. fifecircle said,

    August 28, 2008 at 4:39 pm

    It would be interesting to find out if moves were taken to eliminate the Hawthorne Effect – (i.e. management takes an interest so moral improves) or was the control group just ignored.

    After all the Hawthorne Effect generally results in an improvement in productivity which in this case would be more thorough and hence more vigorous cleaning.

    e.g. 1) advertising healthy eating and the other factors which would ideally be eliminated – in both environments.
    2) pointing out the wonderful safety is in the hotel
    3) (to go for the classic Hawthorne observation) taking equal care to explain a wonderous lighting levels initiative.

  27. Jamie Horder said,

    August 28, 2008 at 5:05 pm

    thegrouchybeast : Re: BMI, it’s a bit more complicated than that

    www.ncbi.nlm.nih.gov/pubmed/16996880?

  28. eponymous85 said,

    August 29, 2008 at 9:49 am

    As an aspiring psychologist, the last sentence of that article made me shed a little tear of joy. There’s hope yet. Thank you!

  29. used to be jdc said,

    August 29, 2008 at 4:37 pm

    @Diotima -
    Ooh, ooh, I’ve heard of those pink patch things. A blogger got their ads taken off myspace. Apparently, they’ve been banned from facebook too.

    TBtAM

    and here

    I’ve also seen a page advertising a dietary supplement as a cancer drug on myspace.

  30. BryanKitts1 said,

    August 30, 2008 at 3:44 pm

    I’m a bit surprised there was no improvement in the control group: if I was weighed, quizzed about my poor diet/alcohol consumption/lack of exercise, and told I would be weighed again in a month, I’d probably try to cut down my fried mars bars (if only for that month).

    I also don’t understand how the initial observation is “not unusual”. Isn’t it effectively saying either that they’re burning more calories than they’re eating, or that the measurements or model are seriously flawed?

  31. max.dobberstein said,

    September 4, 2008 at 7:40 pm

    I don’t watch much telly. When I do I don’t watch sport. I don’t care for beer. I don’t subscribe to a gym. I would like to cycle — or better yet, walk — to work but I expect I would be run over as there are no bike paths around here.

    Does that mean I am not fat or that I am not a person?

  32. psychgradmum said,

    August 28, 2009 at 8:36 pm

    Maybe the ladies in question gave it more ‘gusto’ knowing that they were doing themselves some good by just working. Maybe perhaps? I know if I’m in a better mood I tend to do the housework etc with a bit more energy.

  33. laptopbatteriesshop said,

    April 3, 2010 at 3:59 am

    We don’t just offer the typical cheap laptop batteries that you may find from other e-retailers, we pride ourselves with providing our customers with the most cost effective solution towards laptop battery replacement without sacrificing quality. All laptop batteries and AC adapters will meet or exceed OEM (Original Equipment Manufacturer) specifications. Every that is available on our website is also guaranteed to look, fit, and perform just like your original laptop battery (and usually better). The only real difference is the price. You don’t have to pay a ridiculously high price just for a laptop battery. We produce over 95% of the laptop batteries, laptop chargers, laptop chargers and other products that we sell. There’s also no middle man for us to pay. Lower cost for us translates into lower prices for you.

You must be logged in to post a comment.