A reality check for altitude tents and houses


As of September 2017, new Sweat Science columns are being published at www.outsideonline.com/sweatscience. Check out my bestselling new book on the science of endurance, ENDURE: Mind, Body, and the Curiously Elastic Limits of Human Performance, published in February 2018 with a foreword by Malcolm Gladwell.

- Alex Hutchinson (@sweatscience)


[UPDATE 11/30: Lots of great discussion of this post below and on Twitter. I’ve added a new post with some responses, more data, and further thoughts HERE.]

A recurring theme on this blog is that not all studies are created equal. The quality of the study design makes a huge difference in the amount of faith that we can place in the results. So it’s always a big pleasure to see awesomely painstaking studies like the new one in Journal of Applied Physiology by Carsten Lundby’s group in Zurich. The topic: the “live high, train low” (LHTL) paradigm used by endurance athletes, in which they spend as much time at high altitude as possible to stimulate adaptations to low oxygen, while descending to lower altitude each day for training so that their actual workout pace isn’t compromised by the lack of oxygen.

There have been a bunch of LHTL studies since the 1990s that found performance benefits — but it’s really difficult to exclude the possibility of placebo effect, since athletes know they’re supposed to get faster under the LHTL strategy (and, conversely, athletes who get stuck in the control group know they’re not supposed to get faster). But Lundby and his colleagues managed to put together a double-blinded, placebo-controlled study of LHTL. The main features:

  • 16 trained cyclists spent eight weeks at the Centre National de Ski Nordique in Premanon, France. For four of those weeks, they spent 16 hours a day confined to their altitude-controlled rooms. Ten of the subjects were kept at altitude (3,000 m), and six were at ambient (~1,000 m) altitude.
  • Neither the subjects nor the scientists taking the measurements knew which cyclists were “living high.” Questionnaires during and after the study showed that the subjects hadn’t been able to guess which group they were in.
  • On five occasions before, during and after the four weeks, the subjects underwent a whole series of performance and physiological tests.

So, after going to all this trouble, what were the results?

Hemoglobin mass, maximal O2-uptake in normoxia and at a simulated altitude of 2,500 m and mean power output in a simulated 26.15 km time-trial remained unchanged in both groups throughout the study. Exercise economy (i.e. O2-uptake measured at 200 Watt) did not change during the LHTL-intervention and was never significantly different between groups. In conclusion, four weeks of LHTL using 16 hours per day of normobaric hypoxia did not improve endurance performance or any of the measured associated physiological variables.

This is, frankly, a surprising result, and the paper goes into great detail discussing possible explanations and caveats — especially considering the study didn’t find the same physiological changes (like increased hemoglobin mass, which you’d expect would be placebo-proof) that previous studies have found. Two points worth noting:

(1) The subjects were very well-trained compared to previous studies, with VO2max around 70 ml/kg/min and high initial hemoglobin mass. It’s possible that the beneficial effects of LHTL show up only in less-trained subjects.

(2) There’s a difference between living at 3,000 m and living in a room or tent kept at oxygen levels comparable to 3,000 m: pressure. “Real-world” altitude has lower pressure as well as lower oxygen; this study lowered oxygen but not atmospheric pressure. Apparently a few recent studies have hinted at the possibility that pressure as well as oxygen could play a role in the body’s response to altitude, though this remains highly speculative.

As always, one new study doesn’t erase all previous studies, nor does it override the practical experience of elite athletes. But it suggests that we should think carefully about whether altitude really works the way we’ve been assuming it works. As the researchers conclude:

In summary, our study provides no indication for LHTL, using normobaric hypoxia, to improve time trial performance or VO2max of highly trained endurance cyclists more than conventional training. Given the considerable financial and logistic effort of performing a LHTL camp, this should be taken into consideration before recommending LHTL to elite endurance athletes.


19 Replies to “A reality check for altitude tents and houses”

  1. @Audrey: Yeah, I was pretty blown away. Can you imagine the discussions they had while recruiting volunteers? “You want me to do WHAT?! For HOW LONG?!” (On the other hand, eight weeks all expenses paid in the French Alps… if they do a follow-up, maybe I’d volunteer!)

  2. As a cat 1,former “pro cyclist”, who has always wanted that extra edge to break through into the real pro ranks but could never afford an altitude tent….BAM! Even if this article isn’t 100%, I still happily read it!!!! Thanks for the insight. Sadly, I still kind of want an altitude tent minus the actual using it part!

  3. This study has a lot of holes in it, especially since some of the “best” physiologists state that LHTL works. One bogus study cannot change the work that guys like David Martin and the Australia of Sport (AIS) have performed.

  4. Do you think that being ‘confined’ to one’s room for 16 hours a day may have affected the results? That suggests fairly sedentary enforced behaviour during the time the volunteers weren’t out training. I know it would have driven me nuts.

  5. Alex – what were the individual responses as opposed to statistical average ? were there any individual improvements ? (don’t have access to the full paper myself)

  6. Hi. You hit on the difference between the conditions at altitude and the conditions in an altitude-controlled chamber – pressure. However, at altitude there is the *same* amount of oxygen, but the pressure is lower, which means that less oxygen gets into your system. And of course, all the other elements of air are reduced as well – nitrogen and argon and so on.


    So, what’s interesting is that to simulate altitude, the amount of oxygen is reduced, which isn’t the same thing, since (I would imagine) that means that the ratios of gases are being altered.

  7. Interesting stuff.

    Even though they said that athletes didn’t know if they were living high or low, I do find it very hard to believe that the athlete’s who were set at 3,000m were not aware that they were the ones up high. I used a tent before a 6 day stage race (running) in Colorado and I sure as heck knew when I was sleeping at 9,000ft or higher. Guess the bottom line in this study made this a bit irrelevant, but still.

    The other question I’d have about the study would be where the athlete’s had trained before the study or even what their previous time at altitude camps/tents had been like. Seems to be a little easier going back to altitude sleeping/training each time compared to the first.

  8. >>the subjects hadn’t been able to guess which group they were in.>>

    At 3000m??? That’s almost 10,000′. In my system I struggle to sleep at that altitude and am very aware I’m higher than 1000m. MOST people are the same.
    This is the first place I’d look for what they did wrong and why they were unable to get the sort of gains other studies and athletes have found.

  9. I am surprised that no one has touched upon the idea that perhaps measuring exercise economy at something as pathetically low as 200 watts could not have possibly elicited differences in measurement, especially among elite cyclists for whom 200 watts comes just by thinking about cycling.

  10. Thanks to all for the interesting comments and queries. I’ve tried to address some of the questions (and add further thoughts of my own) in a new post here. Some other quick thoughts:

    @David: Yeah, a little schadenfreude here too. I never managed to spring for a tent (or, more accurately, never got good enough that it seemed worth it!).

    @Hilary: re. the enforced sedentary confinement being a problem. Yes, this is something they considered and couldn’t rule out. The confinement times were 20:00-07:00, 08:00-10:00 and 16:00 to 19:00, which isn’t that crazy for an athlete training hard. But they note that it’s possible that the confinement could have led to a loss of plasma volume.

    @Derrick and @Shaun: Did the athletes REALLY not know whether they were sleeping at 10,000 feet? All I can do is report the results of the questionnaire. They asked the athletes to guess which group they were in on six separate occasions. At the end of four weeks, 25% guessed right, 25% guessed wrong, and 50% had no clue. So why was it so obvious to you guys? Well… no likes to believe that the placebo effect can affect THEM, so I’ll just say that I have no idea! 🙂

    (For what it’s worth, I’ve spent several weeks sleeping at 3000-5000m in the real world, and don’t remember being particularly aware of altitude — other than during exertion — until well over 4000m. Did you try sleeping in the tents at normal altitude, to see what effect the contraption itself had on your sleep?)

    Oh, and if you’re wondering whether they forgot to turn the altitude pumps on or something, here’s an excerpt from the methods section:

    “Three days prior to the study start all rooms were controlled and calibrated with precision gases by the company who built the facilities (Fieldbrook Ltd, London, UK). Subjects were blinded towards the environmental condition they were exposed to in their rooms. Furthermore, all investigators except for the main investigator (who did not perform any measurements) were blinded towards the group assignment. All rooms had the required hypoxic equipment installed, and the air pumps were constantly turned on. O2 fraction in each room was continuously monitored from two independent O2 probes (T7OX-V, Oxygen CiTiceL®, City Technology Ltd, Portsmouth, UK) connected to a control panel located in a room with restricted access. O2 fraction was controlled by the main investigator five days per week (this person staying in the control room overnight) and by a qualified person from the centre, not involved in any measurement, during the two other days. In addition, O2 fraction in each room was controlled twice daily via a portable O2 sensor (VN2O2, Cambridge Sensotec Ltd., Cambridge, UK) by either the main investigator or the person from the centre.”

    Why did they go to all this trouble? Because the placebo effect is very, very powerful.

    @Eddie: Economy was measured during the warm-up for the VO2max test. Time-trial performance was the main performance outcome measure.

  11. @alex
    Yes, have actually noticed a substantial difference while sleeping at 4,000ft (to adapt) and then sleeping at 9-10,000ft. Including that my waking HR is usually at least 10% higher than normal when at 10,000ft.

    Also, I live at sea level and was on vacation in Colorado a couple of years ago (before getting the tent). We stayed in Boulder the first night (5,000ft), but then drove to almost 10,000ft the second day we were there. I almost felt impaired driving so really felt that the differences were really quite noticeable to me even when not exerting.

    (Alex…you need to get a ‘subscribe to comments’ button for your blog. I keep forgetting to check back)

  12. @Derrick: I gather that there’s quite a bit of individual variation in response to altitude, so you may on the “greater response” end of things. Perhaps the 25% of subjects who did guess right weren’t just random — they really did know which group they were in. Of course, the bottom line is that, if the subjects were able to guess their group, we should have seen a STRONGER effect rather than a weaker one, so this is a bit of an academic point (though certainly interesting).

    Re. “subscribe to comments” — yes, someone else suggested that a couple of days ago. There is a “subscribe to comments” RSS feed right at the bottom of each post (below where I’m typing right now), but I’ll have to look into getting an e-mail feed.

  13. Sherpas aren’t known for winning marathons. That always seemed like a glaring refute right there.

  14. I was part of the study. I’ll address a few things here and would be happy to answer any questions.

    How did they convince us to go? Uhh, 8 weeks in the Jura mountains, all expenses paid where all I have to do is ride my bike and give some blood and muscle tissue?? Piece of cake decision.

    I think for the most part, prior to the study, most of were living at fairly relaxed altitudes. As in pretty close to sea-level. And I’m very certain that none of us had trained at altitude or had ever had access to tents.

    While in the rooms, there was no way to tell that you were at altitude. To this day, I don’t know if I was placebo or not. For me, I tended to be parched quite alot, whether in the room or not, but this could also have been due to the fact that normoxia during the study was 3000ft which is about 2500 more than my normal life was.

    Being in the room 16hrs was fairly easy. 8-10hrs was sleepy time which leaves another 8-10. It was pretty easy to get up, have a big brekkie, chill for a couple of hours letting it warm up outside and hit the road at noon then come home, eat, nap, watch tv, have dinner and then watch more TV. It’s really no different than what pros do during the Tour.

    All in all, great experience. I’d do it again. For sure.

  15. @Andrew: Awesome, thanks for the inside look! And if they decide to do a follow-up that includes runners, give me a shout — I’d be up for it in a heartbeat! 😉

Comments are closed.