Study Finds Weight Loss Apps Don't Work

• 

  iridiumstem on 08/11/2015 - 02:21

  +1

  This is not exactly my area, but wow, the article title is over-generalising and you are over-generalising even further.

  They tested two strategies, at best the conclusion you can say with some certainty is "two strategies didn't work in this sample groups", going by the article's info since I have no access to the paper (Stupid maintanence). Everything else involves some sort of uncertainty. I could see that this is supporting the possibility of no effective intervention methods using smartphones. That said, that kind of generalised statements are really easy to refute and you cannot say those statements without uncertainty (since only thing it requires is one counter example).

  Quoting the article, "They called for more research to see if apps could be optimized or bundled into personally tailored and/or multi-pronged weight loss plans."

  I personally think it's far fetched since I could expect some kind of Hawthorne effect (the participants do better because they are in an experiment) on the control group (which I have to read the research paper to check whether this has been into account).

  Eh, I need the paper to say things for certain but I doubt that the researchers intention was to state such a sensationalistic statement.

  • 

    peterpeterpumpkin on 08/11/2015 - 15:42

    +1

    Yeah like most scientific reporting, the title has been "Oversimplified" (bada boom ch!) even more so than the article itself.

    And as usual, the general public or scientific community probably couldn't contest the full article without purchasing it or a journal subscription. The legacy of Aaron Swartz lives on…

    But I always find overconfident conclusions both confusing and irritating (scientific reporting is meant to be more introspective and rigorous than religion and conspiracy). Like saying "books no better at improving reading ability than a bar of soap" after two picture-only books were used for comparison.

    Like say if I create an app that monitors your movements and if you don't move it electrocutes you via a bluetooth controlled dog collar, threatens to injure members of your family via a hired crew of thugs, turns on your phone's camera to expose your lack of movements, transfers money from your bank account to groups you despise and other nefarious things - you might lose a kilo or two.

    • 

      iridiumstem on 08/11/2015 - 15:58

      +1

      The researchers don't make any conclusive statement like what OP and the article suggests they do. It simply says something like caution is needed if only the app is used as it may not be as effective as using them with other means or something like that.

      The article's title is misleading and the article doesn't even mention the title of the paper, but the title of the paper is rather plain and doesn't make any sensationalistic statement (it's boring but tells you more about what they will do in the research paper rather than coming up with rather misleading summary of the results)

      :P Yes, I skimmed through the paper from the article (one of the boons that come with being a student, I suppose). The paper looked to me as if it had some flaws with how they've done the control group (the researchers were expecting a weight increase, but the weight decreased for the control group). Then again, not exactly my field, none of my concern.

• 

  [Deactivated] on 08/11/2015 - 03:42

  Dieting still works. Eg. Eight weeks of healthy vegetables only. With very little else that is not essential to minimum dietary needs, and an overweight person will lose a lot of weight.
  Apps to lose weight, just sounds like a sidestep.

  • 

    c64 on 08/11/2015 - 08:04

    +1

    i dare say that any alteration to a typical diet, eg restriction of junk food, would result in weight loss. keeping it off long term is much harder.

    • 

      [Deactivated] on 08/11/2015 - 10:16

      +2

      Spot on. I think the longterm battle is not with food, it's with self, particularly discipline. Seriously true with smoking too.
      Our bodies will adjust to any healthy diet quickly and cravings for crap will cease. If you eat like a rabbit, the body will adjust and crave what you give it at the right time. If you swallow a greasy barbecue each night, and follow it down with a cow, you will put on weight.
      People go round and round with these things, but 'you are what you eat' applies with losing weight.
      Ancient Fitness guru Jack LaLanne coined the phrase: If it tastes good, spit it out. Never forgot that one.

      Weight loss, stop eating crap, consider typical ages old asian diets they are very smart. I'm cannuck.

• 

  eatwell365 on 08/11/2015 - 03:45

  +4

  The study simply means app alone does not solve obesity - not necessarily waste of time.
  App is a tool and does not say it is a solution.

  It is like selling home gym gear in the middle of the night through TV infommercial shopping channel.
  Does it mean the home gym gear does not work? Exercise works as part of the healthy formula together with eating well. It is selective bias that this group normally does not exercise, they dream to lose weight and have 6 packs abs once paying for the machine without using them/change a thing to their lifestyle.

  Anyway, face to face weight loss consultant may have financial interest/support to the study. You know who??!! It's all about marketing. Yeah..Chocolate is good for you, according to studies.

• 

  Ughhh on 08/11/2015 - 14:47

  +2

  So where the participants 100% following the directions/recommendations from the app?

  If not then, how can one expect a mobile app (or anything really) to help you if you don't even help yourself?

  • 

    iridiumstem on 09/11/2015 - 22:11

    I think that is what they "measured" (I am using quotation mark, because I think that's what they think they thought they did) with number of times apps were opened per week. It's probably what they wanted to find out (the interaction of the apps they've developed with the motivation + existance of some human trainers and the interaction of that and the motivation).

    • 

      Ughhh on 10/11/2015 - 00:00

      +1

      If thats the point of that study, then its a very useless one.

      The same thing can be questioned about gyms, doctors, personal trainers, dietitians etc. You can't expect to lose weight/get healthy by just buying a gym membership/seeing a doctor/hiring a PT etc without putting the effort in yourself to follow the plan or advice.

      People rely on others too much these days, and when it doesn't work they're quick to blame others, everything else except them are useless.

      • 

        iridiumstem on 10/11/2015 - 00:12

        In many many many cases, it's almost impossible to tease out the effect of something without looking at the interaction effect. The study that they've conducted looked at two different scenario. One that uses app alone and another that uses apps + personal trainer.

        Let's say this, looking at a simple example, spending $10 on an app would probably give you more motivation to use it than let's say a free app that you downloaded because it was free. It probably would give a significant (stastically speaking, which means "something that probably happened not as a result of chance (usually 95%)") difference. This is based on a theory that people's elasticity of usage changes as the cost to attain it changes. It's a job for scientists to look at those factors, if see whether these factors gives you a significant difference. You'd look at simple effect, interaction effect and other possible explanations using statistics.

        If an app gives you more motivations to exercise than let's say not having an app, then it's worth studying because then we would have insights on mechanisms of how these actually improve the chance of people losing weight.

        I personally think methology needed bit of tweaking, but then again, they had two psychiatrists who probably know better than I do.

      • 

        Ughhh on 10/11/2015 - 00:42

        Problem is though, $10 is not really a lot. Lots of people buy gym memberships with the similar purpose, hoping that by paying $100s and getting themselves into a contract would force them to attend the gym, thus get fit. But how often does it follow through? Gym companies know and use this 'motivation' to lure you in. Making the app $100 would be ridiculous, no one would buy it because its not worth that much, especially when the app market is flooded with free apps.

        The app is really only a tool and perhaps a minor motivation (as it would be easier to keep track via the app than via paper), rather than the major motivation.

      • 

        iridiumstem on 10/11/2015 - 00:52

        @Ughhh: As I said, the theory I've mentioned dictates that what I said should happen. It's a job for scientists to look at those factors, if see whether these factors gives you a significant difference. If it doesn't work, then we need to look at why and how. If it does, then same thing. Science is all about learning about those.

        What people think is right means nothing in science (well other than possibly being a starting point for a research). Usually psychology and human behaviour related science is under evalutated by people; "It's common sense" or "Everyone knows that". In science, you have to back everything up with numbers, results, etc etc. Everything has to have some form of support so that those things can be refuted or it simply has no value in science (other than potential starting point for a research possibly).

        What you are saying sounds reasonable, I don't disagree with you on that. Can you back it up with some evidence that's controlled for biases and can be extrapolated to the general population without potential biases? Can you infer any causal effect without possible biases? Could you use what you've said into making a policy? That's the reason for the empirical studies. I am sure there are more reasons for researches, I probably missed some.

      • 

        Ughhh on 10/11/2015 - 01:34

        Perhaps slightly off topic, but to be honest, I don't consider or put Psychology under science, at least not with Chemistry, Physics and biology.

        Anyway, my point is- you can't help someone who doesn't help themselves. You can develop the best app in the world or best magical machine to help people lose weight, but if they don't use it, then there's nothing else you can do except maybe put a gun to their head and force them. I'm not gonna bother finding evidence for that, it's not my problem.

        It reminds me, last time I went out to eat at a Pub, a family of 4 was sitting at a table near by. The entire family, including the 7-10 daughter and son, were pretty big… horizontally. The kids were just eating deep fried chips and calamari for dinner, no greens…well no wonder! They're not exactly helping themselves and no app is going to stop them.

      • 

        iridiumstem on 10/11/2015 - 02:16

        @Ughhh: Your point and my point are looking at slightly different things. What you are saying is not exactly what the research was looking at. The research is looking at whether the app they made makes difference, they failed to find one between the app group, app with trainer and control group.

        If you want to say, "apps don't matter, your motivations do", that's fine. That's a reasonable argument. That said, that's not the scope of the research. The research was looking at the effect of the app. Extraneous variables (anything that's not accounted by the research conditions they've put out) were "controlled", at least statistically speaking. What they wanted to find was "given everything is equal, using apps increases chance of losing weight by blah" and that everything equal includes baseline motivation. If having apps work by affecting motivation, it still means there is an effect from having an app.

        As for their results not showing any significant differences and their methodology, I think I have an explanation that's plausible. I think they wanted to use the control as the baseline, but their control group was not representitive of the control condition. I believe there was Hawthorne effect at play and the control group was not controlled properly and therefore it probably had an upward performance bias in control group. I personally think control group might've not been homogenous (they kinda refute this idea, but I thought it was still arguable) and I personally think that there are other factors in play, that said, they've given their reasoning behind why they believe that the composition of the samples were all not significantly different.

        In terms of psychology as a science, many many many universities that puts psychology under science. I'd argue that that's a general consensus in Australia at least, simply because at least Go8 puts psychology under science. I am sure many other universities do the same, it's just that I cannot be bothered to look up the regional ones.

        Psychology studies behaviours and mind using scientific methods. This means that anything that affects human behaviours and mind are looked at with scientific methods. Cognition, brain (hormones, neurotransimitter to structure of brain etc etc), human anatomy and functionality of organs, drugs and medicine, basic immunology, etc etc. I've come across those during my degree in science majoring in psychology.

        If you think of psychology as what Freud have put out, his ideas are not exactly prodominant any more (since 50s I'd argue). What I've learnt about him was from a history side of psychology. He did have influence and the prodominance of that differs between the fields, but at best, the influence stops at subconscious processing exists and he popularised the field of "psychology".

      • 

        iridiumstem on 10/11/2015 - 06:34

        Now I think about it, I mistook your statement as motivations are the medium via which the apps work, therefore apps are useless. That was my misinterpretation when I wrote all that before the one before this one.

        Anyways, yeah, everything I've stated there shouldn't be too unreasonable (unless my caffeine fueled nights caused me to say something that I didn't really meant to state). I've deliberately tried to not use any jargons and not ramble, but I tend to do so especially at nights.

        Think of this as a TL;DR.

        1. The research's looking at the effects of the app (well one strategy that researchers've adopted), so as a result, they would've looked for a difference given that everything else in constant between the groups. Motivation might be a mean in which apps operate under, however, baseline motivation probably was controlled between sample groups given how the difference in samples were not statistically significant.

        2. The research didn't find any significant (in terms of statistics, it means the condition causes differences that is not likelty to be an out-of-chance (95%) result) differences, however, I believe they had flaws with how they've set the control group that they've compared the app and app + PT group to causing a upward bias in the control group's performance.

        3. Psychology is taken as a science by many universities, it uses scientific approaches in studying behaviours and mind.

• 

  teardrops21 on 09/11/2015 - 21:41

  Study finds that most studies are a waste of time.