"Was he looking for 'those toe shoes'?"

"Yeah. He's off."

"Did you not tell him about Newton's third law?"

I work in a running shop in Edinburgh. We don't sell any of the more extreme "barefoot" running shoes, so we get a fair few disappointed customers leaving the shop more or less as soon as they've walked through the door. Seasoned members of staff can normally categorise runners at first glance, in much the same way as you can tell the difference between a door-to-door salesman and a Jehovah's witness. To begin with, the people looking for Vibrams are usually already wearing Vibrams, which is something of a giveaway. They also tend to look disgusted when you tell them that the shop doesn't stock them, as though the sale of more traditional running shoes was some sort of conspiracy.

What started as a trend sparked by Christopher McDougall's book Born to Run has blossomed into a fully fledged religion for some of its converts, but I think the logic of barefoot running breaks down fairly quickly. This is a paraphrased version of a conversation I had with one (particularly extreme) barefoot runner:

"I've managed to teach myself to run in Vibrams. It's taken about eight months, and I still can't run as far as I used to, but it feels much more natural."

"That's good. What sort of surface do you run on?"

"Pavements, sometimes the canal towpath."

"I see."

"I've also started the Paleo diet."

"Is that where you only eat what you can kill or find?"

"No, I still go to the shops. I just only eat raw food."

"Oh."

"We didn't evolve to run in big heavy shoes, you know."

You can't really argue with the final statement here. The problem is that we also didn't evolve to run on roads. Or if we did, the evolutionary process that built the roads was the same one that designed the cushioned running shoe. When people were following the "Paleo diet" from necessity rather than whimsy, their life expectancy was probably in the 20s, if they were lucky. The people who were loping over the savannah in the mind's eye of the barefoot brigade probably didn't weight 14 stone.

Proponents of barefoot running are fond of pointing out that Roger Bannister ran a sub-four minute mile while training in plimsolls. The very fact that Bannister was the first to break this milestone should be evidence enough that he was quite exceptionally physically gifted, though, and he probably would have run quite a lot faster with modern training methods and equipment. For those who are planning to spend £100 or more on barefoot shoes to emulate Bannister, it should be noted that you can buy plimsolls for £4 on eBay.

Others point to the east African athletes currently dominating the sport for evidence that barefoot running is best. When I was in Ethiopia, I saw people running in all manner of different shoes. It would appear, however, that only Eddie Izzard runs barefoot.

When I see people out running in barefoot shoes, they never look as if they're having a particularly good time. You don't even really need to look at their feet, just the awkward shuffle that they are forced to adopt to protect their knees. Most people who use barefoot shoes admit to having to run less because of the pressure on their joints. For someone who enjoys running long distances and the feeling of having a bit of a spring in their step, the choice of barefoot shoes is therefore a bit baffling. I, for one, wouldn't be able to do those things in a pair of glorified socks.

The evidence that barefoot running is more "natural" and less likely to lead to injury is largely anecdotal, and given the short time since the trend started, any studies that have been published are unlikely to reveal long-term damage to participants. A recent study published in Medicine and Science in Sports and Exercise, however, supports what I've learned from talking to customers who have experimented with barefoot shoes.

The study took 36 experienced runners with no problems in their feet or lower limbs according to baseline MRI scans. Half of them began running in Vibram FiveFingers shoes gradually, following the guidelines set out on Vibram's website, while the control group continued to run as normal. After 10 weeks, the control group had remained the same, but 10 out of the 19 runners who had transitioned to Vibram FiveFingers showed raised levels of bone marrow edema (a buildup of fluid similar to bruising) in their feet.

The worry is that the barefoot running trend is a fairly recent one, and most of the study's runners with raised levels of bone marrow edema wouldn't actually have noticed an injury yet – many of the injuries caused by barefoot shoes may be lurking just around the corner. Two of the 19, however, had developed full-on stress fractures. Almost all participants were running less than at the beginning of the study.

Given how much runners usually dread injury, it is strange how readily many have accepted that it is a good idea to go out running in barefoot shoes. Most running shops will now offer gait analysis, with the aim of selecting the type of shoes most suitable to the individual's running style, body type and the surface they are planning to run on. Some people are designed to run in barefoot shoes, with excellent biomechanics and access to soft surfaces to run on. Others, however, need a bit more help from their shoes.

It seems a good idea for as many people as possible to be able to run regularly and comfortably. "Foot-in-a-shoe" shoes may be a better way to achieve this than barefoot shoes for most people.

• Michael Crawley blogs at acceptableintheighties.wordpress.com.