The physiology of the cold: Why might women out-‘survive’ men?

19 Apr 2018 Posted by

So this is a brief post, and it’s a written “mull”, in the sense that I wanted to share with you some thoughts.  Not answers, just theories and thoughts about that absolutely fascinating 2018 Boston Marathon.  As you know, it was a cold, wet and windy day in Boston, and it made for huge excitement, unpredictability and two of the most memorable marathons I can recall watching.

I wasn’t going to post anything about the cold, but then on Wednesday night I was asked by a New York Times Journalist to offer some thoughts as to why men had a higher DNF rate than women.  The stats show that 5% of men dropped out of this year’s race compared to 3.8% of women.  In 2017, for comparison, 2.8% of men dropped out, and 3.3% of women did not finish.

So, in the video below, I started to explore that question, specifically by looking at some of the physiology of cold exposure, and the factors that might predict who succeeds and thrives, as opposed to who merely survives, in extreme cold.


This video introduces two concepts – insulation from higher body fat and delayed and reduced shivering, which may partly explain who was less affected.  One of those – insulation, by virtue of the fact that women tend to have high body fat percentages than men – might apply to this men vs women question.  But there are others, and so that’s what this post is about.

So here are five possible explanations, offered as a conversation starter, and which are by no means proven or necessarily true!


First things first – it may be random.  Or nothing

To begin with, we have a dropout/DNF rate of 5% among men, 3.8% among women.  That’s a substantial increase in the men, from 2017’s 2.8%, and only a small increase in women, from 2017’s 3.3% to 3.8%.

It may mean nothing. It may be within normal variation and the fact that more men dropped out than women in 2018 might be a great big red herring without any physiological significance!  Perhaps everyone is reading too much into it!  It would be helpful to know the data going back 20 years!

However, if we lived that way all the time, we’d have nothing to speak about, so it’s still interesting to speculate that IF the women perform better in the cold conditions, there might be physiological and psychological reasons, and those are fun to discuss.  So here are five possibilities I came up with.  I’m sure there are more.


1.  A clothing confounder?

There’s a potential confounder here, and I have no reason to suggest it other than what I observed in the elite races, and it concerns selection of clothing.  I saw more elite men running in nothing more than a singlet and shorts than I saw women.  My bald, unverified observation was that more men were thus underdressed for the conditions.  And so I wondered whether there may be a general trend for wiser clothing choices among the women than the men?  At the very least, this is a possible confounder that you’d have to control for if you were really trying to answer this question.


2.  A pacing difference that affects heat production later in the race

When exercising in extreme cold, and this is true for land and water, the key is to generate huge amounts of heat.  You’re going to lose heat, especially when it is wet, and so the only way you stay warm is if you generate or produce heat that offsets the heat you are going to lose.  In mountain survival, the concept is that as long as you keep moving, you stay alive.  If you stop, and lie down or rest, then your body temperature can really plummet.  Next stop hypothermia, and a bad day out.

So with respects to performance in the cold, one of the prerequisites for survival is that you have to run hard enough to generate the heat needed to stay warmer.  This, incidentally, may partly explain why getting on the front of the race in the elite races and facing that headwind was not as bad as it would be on a warm day – perhaps the cold matters more than the headwind, and leading allowed heat production that offset any wind effects.

As you see in the video above, shivering will kick in if it has to, but ideally you want to delay that, and that means producing heat through running.  Heat production when running is proportional to velocity, so run fast.

Problem is, you can’t simply go faster.  You’re limited by other things.  So, with this in mind, is it possible that women are better able to maintain heat production in the latter stages of the race because of differences in how they pace themselves?  Could it be that they are better at managing the race, and conserving energy for the second half, when they can speed up or maintain their pace, whereas the men might more often go off too fast, then be forced to slow down after halfway?  And of course, once you slow down, you’re in free-fall as far as temperature goes.
We know, from other studies and some fun observations that I’ve seen, that women do tend pace themselves more evenly than men.  The concept would be that if you have someone running a 3h20 marathon, a woman will do it with two halves of 1h40, whereas men may be relatively more likely to run 1h35 then 1h45.
So, it is possible that if the drop outs were happening in the second half of the race, they were secondary to a loss of heat production and the fall in body temperature it produces.  It would be interesting to track the 5km splits of the DNF athletes and see if this happened (though telling apart cause and effect is difficult this way).



3.  Body fat percentage?

This is covered in the video, so I won’t labour the point, but related to heat production is reduced heat loss.  Could body fat percentage play a role?  In cold water swimmers, if you don’t have a high body fat percentage, your chances of swimming the English channel are pretty much zero.  It doesn’t matter if you are an Olympic level swimmer, without some additional insulation, you cannot keep your temperature up.  There is research showing that otherwise healthy, strong swimmers fail in cold water, whereas those with body fat percentages above 20% are able to defend temperature and swimming ability much more easily.

It’s never been proven in runners, as far as I know, but the same concept may exist.  Part of the problem is that once muscle is being supplied with blood, it loses a lot of its insulating capabilities.  In other words, non-perfused muscle is an insulator, but perfused muscle is not!
Now, what is the situation during exercise?  The muscle is perfused, and so it cannot insulate.
The only options left, then, are clothing (see Point 1 above), and body fat.
And women do have, generally, speaking, higher body fat percentages than men.  So it’s possible that they are simply more able to cope with the cold, and this means that they delay shivering, they don’t develop the peripheral nerve and muscle problems that are caused by falling temperatures, and the end result of all this is that their performance is not compromised as much as for men.



4.  A difference in goal-setting?

This is also a bald assertion, but I suppose that it is possible that more men are uncompromising around their performance expectations, in the sense that they race Boston with a less compromising attitude towards running a PB or getting a place.  If you look at the weather and you came into that race with ambitions to break 3 hours, or to run 3:15, and you see that wind and that cold, your plan is gone.  Is it possible that men opt out of even trying, because if they can’t have a viable shot, they won’t take a shot at all?

I don’t know. Perhaps not. Perhaps it’s the other way around.  But if you really wanted to answer this question, you’d look to control for this possibility by evaluating the reasons for drop out.  Did the runner drop out because their pre-race expectations had disappeared?  Or did they drop out because they had to?  You might find an increased likelihood of one or the other.


5.  Toughness and resilience

And finally, there is the possibility that women are just tougher or more resilient than men!  That’s certainly where a lot of the speculation goes when the stats are first seen.

I’ll tell you why I’m not a huge fan of that.  I think that any marathon is tough.  And so if toughness and the ability to endure tough conditions is responsible for that swing in DNF numbers in Boston 2018 compared to 2017, then I would expect, at the very least, for men and women to have equal DNF rates every year.

Now, I only know the 2017 numbers – 2.8% of men and 3.3% of women – but that’s not clear enough, in my mind, anyway, to suggest that women are more resilient and able to ‘tough it out’.  If there’s a general trend, year after year, for a small proportion of women to drop out, then maybe.  The marathon is always tough, so it should always expose this characteristic, if it exists.  But if anything, we see the opposite.  More likely, we see nothing, and this is too easy a default explanation.

That doesn’t mean it’s not possible. Maybe it takes really tough conditions, like Boston 2018, to expose this difference.  Can’t rule it out.  But the physiologist in me is biased in favour of the previously mentioned points (please don’t send me hate mail!)

Finally, one thing that would be very interesting in this debate, in order to sort through these possibilities, is to know when the drop-outs happened?  If they are happening in the first 5 to 10km, then it would support my 4th and 5th point above.  That is, it’s a decision to not spend energy if they can’t run the goal time, or it’s a resilience thing.  Simply knowing a PB won’t be run, or feeling so unpleasant and horrible in the first 30 to 45 minutes is enough to cause that split in DNF rates.
If the drop outs are happening after say 25km, then I’m leaning more towards a possible physiological limitation that exists in men, and not women.  That’s because it doesn’t matter how cold it is, the adult body has too much heat to become hypothermic within the first 45 minutes, especially when running.  So if the DNF is happening before that, it’s being cased in part by psychology, and in part by those peripheral changes – weak muscles and peripherial paralysis – that I explained in the video.

The actual hypothermia, the lowering of body temperature, those effects would be seen much later.  So a second half DNF might also be because “this sucks and I’m cold and I’m going home”, or it might be a direct effect of actually being too cool (rather than feeling too cold, even though you aren’t!)


Anyway, fun topic.
Do NOT follow this link or you will be banned from the site!
%d bloggers like this: