Hacker News Reader

Comment by jackschultz | Hacker News Reader

I'm always very skeptical with studies like this.

> [R]esearchers recruited a pool of twenty sedentary commuters... These commuters were then asked to shift their transport to work to the e-Bike fleet

> After a month, volunteers revisited the lab and had the very same tests conducted again. Notable improvements to the 20 subjects’ cardiovascular health were evident, with improved aerobic capacity and blood sugar control noted in all cases.

Twenty subjects and a month of testing? They don't mention how notable the improvements are other than "improved". They don't necessarily take into account that them being studied for health benefits might have made them more active overall not just in commuting. There are so many other reasons that could have changed them into having more healthy results. With all those possibilities, I cannot see how twenty participants and one month of changes can be considered confirmed.

View on HN

I don't think the month is a problem. If you're sedentary and you switch to any kind of regular daily exercise, you will see significant improvements in about a month.

Riding a pedelec 5 miles feels to me about the same as walking 2 miles. It's less exercise than riding a normal bike, but it's far more than sitting on your ass. I'm not at all surprised that someone replacing a 10 mile round-trip commute experienced health improvements. It's the equivalent of walking ~4 miles a day.

I agree it's a reasonable result, but that doesn't necessarily make it correct. I think jackschultz's skepticism is warranted: this is not a proper experiment with controls. A better designed experiment would have two control groups: one group that rides normal bikes, and one group that does nothing different. The latter group could try to control for people-being-studied-behaving-different effect. And the normal bike riding group could give us a better handle on what is being "missed" by using electric bikes. We could also get super-fancy and add a fourth group, based on what you said, who don't change their commute, but instead walk 4 miles a day.

If you actually read the study referred to by the article, you'll see that they aren't making the claim you and jackshultz are arguing against. Nor does the article define "cheating" the way you seem to.

They measured some basic indicators of metabolic health, and found that e-biking is moderately effective cardiovascular exercise. The metabolic response they saw is well outside the noise, which is typical for these experiments: Sedentary humans respond remarkably well to exercise. (If that's not good enough for you, they also analyzed heart rate and GPS logs and found that their e-bikers were averaging about 5 METS for 40 minutes a day, 3 days a week. This is roughly what a doctor will recommend to an office workers who wants to lose some weight.)

Thus, you are not 'cheating' by pretending to exercise. If you want to know whether e-biking is "cheating" compared to regular biking, you can look up similar studies on the health benefits of biking and compare the numbers.

Controls are necessary when you're looking for more subtle effects, but the health benefits of replacing no exercise with moderate exercise are not at all subtle. All this experiment had to do was establish that e-biking elevates one's heart rate. Which is a gigantic 'duh'.

So what exactly are you hoping to accomplish by kvetching in the comments about their lack of scientific rigor?

I am not objecting to the "cheating" claim. I disregarded it immediately and only considered this as an exercise experiment. My point is that single studies without control groups are not going to tell you much. Are their conclusions reasonable? Of course. Did we learn much from the study above what we already knew about moderate exercise? Not really.

"I'm always very skeptical with studies like this."

The study was large enough to disprove the hypothesis. It failed to disprove the hypothesis. Science has advanced.

The real problem with this is sedentary. There are so many studies that try to prove "X improves endurance" and run a study with sedentary people.

The problem is that when you have sedentary people do any kind of activity that increases heart rate regularly, they will come out of it with improved markers. The learning curve here is pretty steep. It tells us nothing about the actual activity you were trying to test.

Given that "electric bikes" is a major section on this website, I wouldn't be surprised if this were a carefully placed advertisement. They even dropped the names of a couple of local [electric] bike shops that donated funding to the study.

Reply on news.ycombinator.com