If you could do *any* study, what would it be?

Anyone who has sat through a thesis defense has heard some variation on the following question: “If you had unlimited funds, what would you do for your next study?”. It’s a good question. Like a lot of researchers, I have various lists of study ideas floating around in notebooks, on my computer, and even on my cell phone. I spend a lot of time daydreaming about different studies – ones that would be feasible during my post doc, ones that will need to wait until I land a tenure-track job, and others that will only be possible if I befriend Mark Zuckerberg.

While I am admittedly more nerdy than the average person, I can’t be the only person who has ever listened to a presentation and thought to myself “Wow, if only we could do ________!”.

So I ask you, people of the internet: If you had unlimited resources (money, lab equipment, trained personnel, participants etc), what study would you run? The study needs to be ethically feasible, but after that you are limited only by your own imagination.

Personally, my ideal study stems from an idea I had after Bob Lustig spoke at one of our lab meetings in Ottawa at the start of my PhD (this is based on my recollection of the presentation, which I think is accurate, but may have morphed with time). If you don’t know him, Dr Lustig is an endocrinologist, and one of the leading “anti sugar” advocates. He argues that we should do all we can to cut sugar (and especially fructose) from our diets, and claims that their inclusion in our diets results in all manner of metabolic disturbance (insulin resistance, weight gain, etc).

Most interesting to me, Dr Lustig also claims that sugar intake can reduce physical activity – his basic argument (if I understand correctly) being that after the blood sugar spike, you get a sugar crash, and no one feels like exercising when they have low blood sugar. And so (the thinking goes) if people were to consume less sugar, they would spontaneously begin to exercise more.

That, to me, is a pretty cool idea. So my ideal study would be to take a bunch of inactive kids who also consume large amounts of sugar (especially in the form of pop, juice, etc). I would measure physical activity and sedentary behaviour levels of all kids at baseline. Then we would randomly reduce the sugar intake of half the kids, and see what happens to their physical activity levels over a period of several months. This sort of study could have really important implications for the way we view energy balance (e.g. can you out-train a bad diet, if the bad diet is actually causing you to train less?), and would be (in my opinion) very cool.

Now it’s your turn. If I gave you unlimited resources, what is the one study you would choose to execute and why?


Related Posts Plugin for WordPress, Blogger...
This entry was posted in Miscellaneous. Bookmark the permalink.

12 Responses to If you could do *any* study, what would it be?

  1. The sugar study you propose is a great idea. I would love to see more extensive testing about any of the anti-sugar ideas, but your idea in particular is cool! So many people, myself included, feel this “let-down” effect after a sugar rush. I presume the bio-chemical explanation would be that one’s insulin gets used up in the initial sugar rush, and for a brief time after the rush you don’t process energy efficiently? It would be great to quantify it. Do people with more insulin resistance experience more or less of this “let-down” effect than the average person? Could the “let-down” effect serve as a proxy to estimate a person’s insulin resistance? Definitely test it on different types of people. I think I’d rank your idea as the #2 idea I’d love to study with unlimited funds.

    My own #1 idea is just something that has crossed my own mind repeatedly when I discuss obesity with my friends — although I’ve never actually proposed this to anyone else, it’s just a cynical dark thought deep in the dungeons of my mind — so I’d like it answered just for my own piece of mind.

    I’d like to see if and how much bias exists when test subjects observe obese people eating, or maybe even exercising too. Secretly train a group of scientific collaborators, of various different body sizes, very carefully to put the exact same number of calories onto their plate at a buffet (close as you can get it). Then tell the test subjects that the test subjects are the scientific collaborators, ask them to help the experimenter neutrally and objectively estimate how much food people put on their plates at buffets. Don’t discuss obesity. My hypothesis is that your average layman will over-estimate how many calories an obese person is eating, even when skinny people select the exact same meal right next to the obese person. Conversely, you could do it with exercise. Use hidden meters to precisely measure the calories that various scientific collaborators expend in a gymnasium. Give the test subjects some guidelines about how to estimate caloric expenditure, do _not_ discuss obesity, and ask the test subjects to help the scientist estimate the caloric expenditure of everyone in the gym, in a neutral and unbiased manner. My hypothesis is that you would show a mathematical trend where laymen test subjects under-estimate how many calories that obese people are spending while exercising, and over-estimate the thin people.

    I’m not a scientist, I’m an engineer, so I’m not at all familiar with lab experiment ethical guidelines. Are those experiments ethical? Did the Milgram Experiment create an ethical guideline where you’re not supposed to deceive the test subject about who is being tested?

    • Travis says:

      Thanks Thomas, those are both excellent ideas! Obesity-related bias is always interesting, and I don’t think anyone has done those particular studies. I completely agree with both of your hypotheses. I also like that they are both very intuitive – I like studies that you can understand without having an advanced degree on the topic!

      Deception is generally ok, I think. It needs to be justified, and you need to debrief people afterwards to let them know about the deception (and also ask if they figured out the deception). But I’ve personally participated in a bunch of psych experiments as an undergrad that involved deception, so I think that it’s generally permissible, given that it meets certain criteria.

  2. Test different models of peer review and scientific evaluation as if they were new drugs coming to the market.

  3. Andy says:

    I would like to pluck people from their stressful lives and allow them to live as near to stress-free as possible for at least two years and see if that alone leads to weight loss. The environment would probably have to be something akin to living in the country in a welcoming, crime-free community, staying with an understanding family and having a close relationship with a mentor (perhaps the family itself). The mentoring would be simply helping pay their way by helping out on the farm and around the house in any small way they can, working for the community, possibly as a teaching assistant and probably some art therapy of their choice such as writing that novel, or maybe learning an instrument or a skill.

    This wouldn’t suit everyone and it would be impossible for parents to leave their kids for two years. So it would apply to young, single people who need to turn themselves around to avoid a life of obesity.

    To ease the stress still further, they would be paid a wage for being an experimental subject. That way they wouldn’t be stressing in the same way they might do at ‘fat camp’ where they know it’s only for a few weeks, it’s costing money and they have to go back to their stressful life and triggers and money problems in a short while.

    It’s true that living with a rural community, good cooking and eating habits would be instilled anyway along with an uptick in daily energy expenditure. That may confound the results as to whether stress reduction was the key. But if you had a control group back in the city who were given once-per-week classes in all of the above and left on their own to implement that same knowledge, I doubt they would fair as well. And that would tease out the key role of stress reduction.

  4. Jameson Voss says:

    Travis, great idea to test these ideas with science. For what it’s worth, this new study suggests higher sugar content increases exercise intensity and fatigue over the short term. Perhaps it could reduce motivation to initiate activity even if improving performance during activity. Either way, it’s great to test rhetoric on any side with cold hard science.


    • Travis says:

      Thanks for that, Jameson! This fits with my own experience – I would take a gel at the start of my marathons for the same reason. I’m wondering if regular consumption of a high GI diet might be different? Or maybe responders vs non-responders?

  5. Zach Ferraro says:

    Great post Travis! Always an examiner favourite. As for me, this is not what I said during my PhD defense, but here’s what I’d love to see done with unlimited funds and no logistical complexities (what a utopia!).

    Since my work and primary interests lie in the developmental origins of adult disease I would use the money to build an extension of the Canadian Health Measures Survey (CHMS). As of now it’s an representative observational study of Canadians and it does not include pregnancy or pregnant women, nor does it follow newborns into adulthood. I will use the same methods of assessment for fitness, anthropometry, dietary intake and energy expenditure but tweak the design slightly if a woman consents to participate once she determines she’s pregnant. At this point I would randomize women (and families) to comprehensive lifestyle intervention (both advanced medical and behavioural support including diet, PA and psychology) or standard care. The primary outcome would be offspring body composition at birth and sequentially every 2 years thereafter. However, the intervention would focus on maternal health, appropriate kcal intake, adequate PA and adherence to the gestational weight gain guidelines. Therefore, the CHMS would become a randomized control trial within a longitudinal birth cohort that follows all subjects in both arms of the trial (mom, dad, offspring) throughout their life-course until death.

    Doing this study, with the precise direct-measurements employed in the CHMS, would help answer some vital questions still lingering in science and medicine concerning the genetic and environmental contributors of obesity, diabetes and CVD. There are many birth cohorts that show strong associations between X variable in mom and Y outcome in child however this would be the first of it’s kind to attribute some causality. Inherent biases aside, my hypothesis would be that comprehensive, accessible, holistic/multidisciplinary care will show tremendous health and economic benefits over what we are currently doing on the front line.

    With such a complex design I realize I there are many fine details to work out and that this study will take years and years before solid data is available however in a perfect world this all makes sense (to me at least)!
    – Zach

  6. WRG says:

    I really want to know more about gut bacteria. Apparently, there’s a great difference between the gut bacteria of thin and heavy people. I can’t believe there isn’t more research going on about this fundamental difference that has nothing to do with how “good” or “bad” you are.