self experimentation in health and fitness

Large scale scientific studies tell you less than you think they do.

Unless you were a participant in a study you do not know how the studied input will affect you. The result, while not meaningless, should not be taken as gospel. This is especially true when we talk about complex topics like diet, supplements, or exercise, where confounding variables, genetics, and environmental factors can radically affect results.

The only way to find out how a new diet, exercise, or supplement is going to affect you is to run your own N of 1 study.

Consider this completely made up study:

Research question: Will this magic pill make people taller?

Research approach: 

• 100 people were recruited, their height was measured and they were split into 2 groups of 50.

• Group 1 was given a magic pill with breakfast everyday for a year and group 2 was given a sugar pill.

• After 1 year everyone’s height was measured again.

Results: “On average, the magic pill made people 15cm taller and our statistics dork told us he’s pretty sure the magic pill will make you somewhere between 10cm and 20cm taller than a sugar pill will.”

Averages and ranges are helpful on large scales, such as public health, and when studying safety profiles because they give us general guidelines, like effective doses, and provide obvious red flags, like toxicity. However, when you consider results in terms of averages you lose a certain amount of nuance.

-

I recently read a tweet I loved that is related to this idea from @mmay3r:

“The richness of anecdote must be stripped to be integrated into data”.

-

For example, unless you saw the magic pill study’s raw data you would have missed the fact that 20 people in the magic pill group didn’t grow any taller and one dude grew 50cm in a year.

Until you go and eat a magic pill with breakfast every day for a year you won’t know whether you’ll grow 0cm or 50cm in a year. Turns out this study told you very little about how the magic pill is going to specifically affect you.

Large scale studies do provide valuable information, such as the safety profile of a new supplement or the injury risks of a new exercise. They can also help you set expectations for your N of 1 study. For example, in our made-up study nobody died from eating a magic pill everyday, so you can comfortably run your N of 1 study on the magic pill without fearing the reaper. In addition, you know you can probably expect that you won’t grow much more than 50cm in a year.

My point is that to truly know how a diet, supplement, or exercise will affect you, you must try it. You can and should use larger scale research to find obvious red flags and guide things like dosage, but beyond that the only way to get a true result for yourself is to conduct an N of 1 study. 

Send it, silly!

If you’re interested in personal training at home, at the gym, or online please visit our contact page or click the ‘sign up for training button’.