Altitude Training for Sea-Level Competition 

A Baker and W G Hopkins

Summary · Background · Live High, Train High · Live High, Train Low · Practical Issues · Ethics · References

Affiliation: A Baker, MD, San Diego, California, USA; WG Hopkins, PhD, Physiology and Physical Education, University of Otago, Dunedin NZ.
Acknowledgments: Allan Hahn (reviewer), Don McKenzie (reviewer), Mary Ann Wallace (editing)
Correspondence: (Will Hopkins)
Reference: Baker, A. & Hopkins, W.G. (1998). Altitude training for sea-level competition In: Sportscience Training & Technology. Internet Society for Sport Science.

Summary. Training near sea level while living at an altitude of 2500 m (8000 ft) for a month enhances subsequent endurance performance, probably by increasing the oxygen-carrying capacity of the blood through an increase in production of red blood cells. A small proportion of athletes shows no improvement or even reduced performance with this "live-high train-low" strategy, but the enhancement for the average athlete is 2-3%. The extra red blood cells and the enhancement of performance are probably lost by 2-3 months after return from altitude. Living and training at altitude is less effective than living at altitude and training near sea level, because the lack of oxygen at altitude results in detraining through reduction in intensity of training. Loss of heat acclimatization from training in cooler temperatures at altitude can also reduce the benefit of altitude exposure. Sprinters may benefit from living or training at altitude, but less is known about the magnitude, duration, and mechanism of the effect. Athletes residing at altitude can get the benefit of training at sea level by performing high-intensity training on ergometers while breathing oxygen-enriched air. Athletes residing at sea level can simulate altitude exposure by resting in a chamber at reduced air pressure or in a "nitrogen house" or tent flushed with air containing less oxygen. Infusions of blood or injections of the hormone erythropoietin also simulate effects of altitude exposure.  All of these practices can increase the fraction of red cells in the blood toward a dangerous level, so they need to be monitored and controlled by blood testing.
Reviewers' comments


The higher you go in the atmosphere, the thinner the air. Thinner air means less air resistance, so athletes who sprint, jump, or cycle will perform better at high-altitude venues. But thinner air also means less oxygen, so the pace of hard endurance training and competition--which depends on high rates of oxygen consumption--gets slower at altitude.

If you live at altitude for several weeks, your body adapts to the shortage of oxygen. The most important adaptation for the endurance athlete is an increase in the number of red blood cells, which are produced in response to greater release of the hormone erythropoietin (EPO) by the kidneys. Red cells carry oxygen from your lungs to your muscles. More red cells means your blood can carry more oxygen, which partly makes up for the shortage of oxygen in the air. So to compete in an endurance event at altitude, you should live at altitude for several weeks before the event.

But what about when you come back to sea level? Will the extra blood cells supercharge your muscles with oxygen and push you along faster than ever? That's what should happen, but there are problems. When you first move to altitude, the shortage of oxygen makes it difficult to train intensely, and you may also suffer from altitude sickness. If you don't adapt well to altitude, you may overtrain or lose muscle mass. Even if you do adapt well, you still can't train with the same intensity as at sea level. The result? You detrain. When you come back down to sea level, you may do better or worse than before, depending on the balance between adaptation and detraining.

A potential solution to this problem is to live on a mountain, but avoid detraining by coming down each day to train--in short, to live high and train low. In this article we'll show that this strategy is effective, and we'll compare it with the less effective traditional altitude training. There aren't many places in the world where you can live high and train low, so we'll also look at how to get the train-low effect without coming down the mountain, and how to get the live-high effect without living on a mountain. We'll end with the ethics of these approaches to performance enhancement, and we'll explain how a blood test can help keep their use within safe bounds.

Research on altitude exposure for sea-level performance has moved ahead rapidly in the last few years. If the work has not appeared in peer-reviewed journals, we have used published summaries (abstracts) of papers presented at conferences. We checked for recent publications by searching the Web of Science for "altitude and training". In all we have reviewed 10 summaries and 15 full papers directly related to the effects of altitude exposure on sea-level performance, and we have cited 12 supporting papers and three reviews. Our conclusions differ from those of some recent reviewers, who took a more cautious view on the basis of then available evidence (e.g. Wolski et al., 1996).

We have not dealt with strategies of altitude exposure and altitude training for competition at altitude. That topic may be the subject of another review at this site.

Live High, Train High

Many athletes and coaches have generally accepted the idea that traditional altitude training--living and training high--benefits sea-level endurance performance. In a round-table discussion between four experts on altitude training for athletics, the average best altitude and best duration at altitude were 2200 m for 4 weeks. These coaches also thought that the effects of altitude training were optimal 2 to 3 weeks after return from altitude (Baumann et al., 1994).

What does research show about living and training high? In Table 1 we've summarized the 17 studies we could find that used athletes as subjects and that included tests of athletic performance. We've also included studies in which maximum oxygen consumption was measured, because a change in maximum oxygen consumption should produce a similar change in endurance performance.  Five studies showed a definite positive effect (unlikely to be zero), whereas only one showed a definite negative effect. Of the remainder, two were clearly positive, four were clearly negative, and five showed a minimal effect; but there were too few athletes in these studies to be sure what was really going on. In summary, there is a tendency towards a benefit from altitude training, but there is a large variation in the outcome between studies. Some of the variability could be a result of different altitudes (1300 to 4000 meters) and different periods of training (12 to 63 days). In the six studies with no control group, the changes in performance could also have been the result of factors unrelated to living and training at altitude.

In only two studies were the athletes tested more than two weeks after return from altitude, when coaches think the benefits peak. The performance test in one of these studies was not particularly sport specific (an incremental test to exhaustion for runners on a cycle ergometer), and the period at altitude was the shortest of all the studies. But the maximum oxygen consumption recorded in that test showed a much bigger increase at 16 days after the period at altitude than at 3 days (Burtscher et al., 1996).  Researchers in the other study addressed most of the problems of previous studies. We'll now analyze their results.

Levine and Stray-Gundersen (1997) randomized 39 runners to three kinds of training for four weeks: live high, train high; live high, train low; and live low, train low (the control group). There was a lead-in period of six weeks at sea level with two performance tests, and a follow-up period of three weeks at sea level with a further three tests. Training was periodized and tapered before the tests. The runners were strictly club-level: their average 5000-m pace was around 80% of world record. But on the basis of more recent work by this group (Chapman et al., 1998), it's safe to assume that the findings apply to top athletes. The data are summarized in Figure 1.

Figure 1: Percent change in 5000-m time from baseline performance (at 6 weeks) in three training groups.  


Recalculated and redrawn from data of Levine & Stray-Gundersen, 1997. The bar labeled likely range of true change is our estimate of the 95% confidence interval for points other than the baseline. 

We'll focus on the high-high and low-low groups first. The difference in performance between these two groups after altitude training averaged 2.5%.  But it's important to note that the altitude group showed only a marginal improvement in performance, and only at three weeks after the return from altitude.  In contrast, the control group performed worse and had not recovered fully by the end of the study.  Training camps, especially of the high-quality that was organized for the control athletes, are supposed to make athletes better, not worse!  To explain this paradoxical result, the authors pointed out that the athletes in the control and altitude groups performed the sea-level tests in a hot, humid environment (Dallas, Texas); whereas they all trained in cool, dry conditions. Apparently, athletes in all groups lost adaptation to heat during the camps, but athletes in the control group performed worse after their camp because they did not have the advantage of altitude exposure.  Athletes in the high-high group showed an unintended but substantial increase in duration of training relative to the other groups during the four weeks of altitude, but whether this increase could account for the difference in performance between the high-high and control groups is not clear.  It's also possible that the athletes in the control group did not have the same motivation to perform well in the post-tests, because they should have readapted to the heat and returned at least to baseline performance after three weeks back at sea level.

Was reduction in training intensity an issue for the high-high group?  Training runs of moderate intensity weren't much slower: the high-high group performed them at 76% of 5000-m race pace, compared with 82% for the low-low group. But intervals of high-speed running, which the low-low group performed at 111% of 5000-m pace, averaged only 96% in the high-high group. And athletes who did not improve in the first test after the period at altitude had performed these intervals at lower intensity than athletes who showed a marked improvement (Chapman et al., 1998). The simplest interpretation of these data is that the reduction in training intensity led to detraining that was enough to offset the beneficial effect of altitude exposure. Can athletes retrain in time to reap the benefit of altitude?  In Figure 1 notice that the high-high group was improving three weeks after returning from altitude. Only part of this improvement could have been due to retraining, because the control group showed a similar trend that was attributed to heat re-acclimatization. The trend towards improved performance may have continued beyond three weeks; but until more research is done, we won't know when performance peaks following a month of traditional altitude training.

Most studies of altitude training have focused on endurance performance, for the likely reason that adapting to shortage of oxygen should enhance performance in events that are limited by ability to consume oxygen.  But there are indications that sprint performance, where the energy is provided mainly by anaerobic processes, might benefit from altitude training.  In a controlled study that has been published only in abstract form, Martino et al. (1995) found major improvements in a 100-m swimming sprint and in peak and mean power of the upper body in a short all-out test. Levine and Stray-Gundersen (1997) also noted an improvement in performance of a 2.5-min run to exhaustion only in their high-high group, but they did not provide data for the other groups. A study by Rusko et al. (1996) appeared to show a negative effect of altitude training on anaerobic power, but the measure of power was an indirect estimate based on measurement of oxygen consumption.  Mizuno et al. (1990) discovered a possible mechanism for an enhancement of sprint enhancement: after altitude training, the muscles of their subjects had an increased ability to absorb (buffer) an increase in acidity.  A build-up of acidity is probably a limiting factor in sprinting, so more buffering should mean better sprint performance. These researchers did not test sprint performance, but they made another interesting discovery:  a very high correlation between change in buffering and change in performance of a run to exhaustion lasting about 5 minutes.  It follows that altitude training could enhance endurance performance not only by increasing oxygen supply to muscle but also by reducing acidity in muscle. The role of muscle and blood buffers in sprint and endurance performance following altitude exposure needs more research.

Live High, Train Low

We found eight studies of the effect of living high and training low on subsequent sea-level performance (Table 2). One group of researchers studied athletes who lived and trained at altitude but breathed oxygen-enriched air during hard training to simulate training low.  Five studies involved athletes living on a mountain at 2500 m and descending to 1250 m on most days to train.  In the other two studies, the athletes trained at sea level but got altitude exposure equivalent to 2200-3000 m by spending most of the time in a "nitrogen house" flushed with air containing more nitrogen and less oxygen than normal.

The average athlete in almost all of these studies showed an improvement in endurance performance within the first week of return from altitude, and in most studies the improvement was definite. The only researchers to look beyond a week are Levine and Stray-Gundersen (1997), with a group of athletes who were part of the study shown in Figure 1. After several weeks, the athletes in the high-low group showed a trend towards further improvement, but the athletes in the other groups showed similar trends. As noted already, these trends may have been the result of re-acclimatization to the hot conditions of the testing venue. It seems reasonable to conclude that the benefits of living high and training low are available immediately, but athletes may need several weeks to readapt to sea-level temperatures if they train in cooler conditions. By then, the average improvement relative to performance before altitude exposure is probably 2-3%.

The average athlete can expect an enhancement of performance of a few percent from living high and training low, but it is now clear that some athletes get an even bigger boost, while others may get no benefit at all. Chapman et al. (1998) have analyzed these differences between athletes, using data for sub-elite runners from several previous studies as well as data from a new group of elite runners. (See Table 2 for details of the design for this group of athletes.) They classified the sub-elites as non-responders (no improvement in performance of a 5000-m run 3 d after return from altitude) or high responders (better than the average improvement of 1.4%). Of 26 sub-elites who lived high and did at least their high-intensity training at a lower altitude, 31% were non-responders and 54% were high responders. The new group of 22 elite runners, who did their high-intensity training low but otherwise lived and trained high, had a similar average improvement (1.2%) and comparable proportions of non-responders (23%) and high responders (41%). In contrast, of 13 athletes who lived and trained high, 54% were non-responders and only 23% were high responders. These data reinforce the advantage of living high and training low over the traditional high-high training. What's more, the true differences between the proportions of non-responders in each group are likely to be somewhat greater for two reasons.  First, they are based on a test performed within a few days of return from the altitude camps, when the athletes had either not re-acclimatized to the Dallas heat or had not recovered from the detraining effect of reduced training intensity.  Second, the usual 1-2% variation in an athlete's performance between tests (WGH, unpublished observations) will tend to smear out the true differences in proportions of responders and non-responders.

What accounts for the individual differences in the response to altitude exposure?  There's always been a concern that better athletes might respond less because they might have less headroom for improvement, but that's clearly not the case here.  Previous work by the Dallas group had identified inadequate iron stores as a contributing factor (Stray-Gundersen et al., 1992): extra iron is needed for the increase in production of red cells stimulated by exposure to altitude.  But in their more recent work, all athletes had been given iron supplements to offset any iron deficiency. The authors could not identify any other blood test, lab test, or physical characteristic that would help predict which athletes were more likely to benefit from an altitude camp. There were clear differences after the camp: the high responders had a greater and more sustained increase in the concentration of erythropoietin, and they also ended up with a substantial increase in volume of red blood cells and in maximum oxygen uptake.

These differences between responders and non-responders show that the mechanism of enhancement of endurance performance is probably an increase in the capacity to transport oxygen to the muscles.  The differences also provide a strong argument against the possibility that the enhancement in performance is due entirely to a placebo effect, whereby athletes are motivated to perform better through knowing they have had a special treatment that is supposed to work.  Differences in performance due entirely to a placebo effect are most unlikely to be associated with changes in a physiological variable, especially when the athlete isn't aware of the changes.

Practical Issues

We now try to answer important questions for athletes and coaches wanting to use living high and training low:  who should get altitude exposure, how high should you go, how long do you need at altitude, and how long does the effect last?  We also examine practical aspects of the different ways of achieving a live-high train-low effect.

Who Should Get Altitude Exposure?

Researchers have focused on tests of performance lasting 10-20 minutes, but it's safe to assume that living high and training low will enhance any competitive performance limited by the ability of the athlete to consume oxygen.  That means high-intensity events lasting between a minute or two and several hours.  Ultra-endurance athletes would probably also benefit, although it's not clear whether their performance is limited directly by oxygen consumption.  More research is needed to determine whether sprint athletes benefit from living or training high.

One of the effects of altitude exposure is to increase the concentration of red cells in the blood.  Sports have begun setting limits on this concentration, so athletes close to the limit might have to opt out of altitude exposure. The limit will not be a problem for the average athlete. For example, the International Union of Cyclists has set the limit to 50% for males (Martin et al., 1997), but even after exposure to altitude, the club-level athletes in the study of Chapman et al. (1998) had an average of only 42% (typical variation, ±3%). The athletes who do need to worry are those taking the banned drug erythropoietin, which apparently can increase the concentration of red cells to well over 50% (Sawka et al., 1996).

Ideally, only responders to altitude exposure should get altitude exposure, but the only way to identify responders is to expose everyone.  Even then, there are problems deciding whether a given athlete has responded.  Changes in performance in an individual's time trial or test of maximum oxygen uptake may be too unreliable to make judgments about the response of that individual, because the effect of altitude on performance is similar to the normal variation between tests. What's more, performance might not peak for several weeks after exposure, owing to heat de-acclimatization, altitude sickness, or reduction in training intensity (if the low-altitude training can't be close to sea level).  Some athletes might even go easy in their pretests, to make sure they get an improvement in the post-test. They might also show a placebo-related improvement in the post-test.

A blood test would be a more objective way to assess response to altitude exposure, but changes in something simple like the proportion of red cells or concentration of hemoglobin (the protein that carries oxygen inside red cells) did not correlate well with changes in performance in the study of Chapman et al. (1998).  Changes in red-cell mass appeared to be more useful in that study, but red-cell mass is technically difficult to measure; and it may not be reliable enough to assess the response of individuals.  Monitoring the change in concentration of erythropoietin in the blood during the period of altitude exposure may be an effective way to identify non-responders, but the reliability of this method needs further study.  Meanwhile, keep in mind that the proportion of true non-responders to a typical high-low camp could be less than 10%, so there should be little or no harm in continuing to expose the whole team, including those athletes who apparently did not respond previously.

How High Should You Go?

In highly trained athletes, four weeks of training at an altitude of 1740 m produced no change in mass of hemoglobin in the blood and only a small increase in maximum oxygen consumption (Gore et al., 1997).  Although this study did not include a train-low component, we conclude that higher altitudes are needed to stimulate red cell production in athletes living high and training low.

The higher the altitude, the greater the stimulus to produce extra red cells.  But the upper limit of altitude is set by the side effects of altitude exposure.  A short-term effect is altitude sickness (Coote, 1995).  The symptoms--headaches, loss of appetite, sleeplessness, and feeling sick--usually last only a few days at altitudes of around 3000 m; but at higher altitudes they can be severe enough or last long enough to interfere with training. (See Bovard et al., 1995, for advice on how to minimize altitude sickness, including use of the drug acetazolamide.)

Although the evidence is anecdotal, overtraining is probably another side effect of living too high for some athletes (Rusko, 1996).  The shortage of oxygen at higher altitudes appears to be a stressor that these athletes cannot adapt to.  The result is inability to sustain previous training loads and a gradual loss of fitness.  A substantial increase in heart rate when the athlete stands up may be an early warning of the onset of such overtraining and a sign that the athlete should return to sea level (Rusko, 1996).  A related sign of living too high may be wasting of muscles.  The effect occurs in mountaineers at high altitudes (MacDougall et al., 1991), but it's not clear whether it's a risk for athletes living at altitudes of about 3000 m and training low, or whether it could be detected by suitable measurement of muscle mass. Athletes living at 3000 m or higher would be wise to include some form of monitoring for overtraining: feelings of fatigue during training, reduction in performance of criterion tests, and possibly changes in heart rate, body mass, or more sophisticated measures of muscle mass.

What about the optimum altitude for training low?  For endurance athletes, it's likely that closer to sea level is better, because high-intensity training, even at 1250 m, has to be reduced in intensity relative to such training at sea level (Levine and Stray-Gundersen, 1997).  Do you have to come down the mountain for all your training?  Apparently not: a group of athletes who did only their high intensity training at low altitude (a "high-high-low" strategy) got the same enhancement of endurance performance as a group who did all their training low (Stray-Gundersen and Levine, 1997).  For sprint athletes, the optimum altitude for training is an open question.

How Long Do You Need?

The best duration of stay at altitude is uncertain, but there are strong clues from changes in the concentration of erythropoietin in blood during altitude exposure.  The concentration rises in the first day at an altitude of 2500 m.  After two weeks it is still high, but declining.  After four weeks it is back to baseline (Chapman et al., 1998).  Conclusion: three or four weeks is long enough for one stay.  It's possible that shorter exposures repeated every few weeks will produce an overall greater release of erythropoietin and therefore a greater production of red cells. The only study of intermittent short exposures is the uncontrolled high-high study by Daniels and Oldridge (1970), which appeared to produce substantial enhancements in performance.  More research is needed here.

How Long Does The Effect Last?

Three weeks post-altitude is the longest time that researchers have looked, so we've made an estimate based on our understanding of the physiology.  If we assume the increase in mass of red blood cells is the main factor accounting for increased performance and that production of red cells returns to normal soon after return from altitude, the effect should last the lifetime of the cells. The lifetime is about four months in the average adult, but apparently only 2-3 months in athletes training hard (reviewed by Szygula, 1990, but he does not cite accessible references). So we expect the benefits to begin to wane by the end of the second month after altitude exposure, and to have disappeared completely after three or four months. Studies in which athletes have received infusions of extra red cells provide indirect evidence for and against this estimate. In one study, the infusions produced an enhancement of endurance performance that had returned to baseline after 16 weeks, although maximum oxygen uptake remained high (Buick et al., 1980).  In another study, published only as an abstract, maximum oxygen uptake and performance had returned to normal after four weeks (Goforth et al., 1982).

When the effect of altitude exposure has waned, further exposure should stimulate production of red blood cells again and restore performance. Whether the effect is re-established more quickly or augmented is not known.

What's The Best Way To Get Altitude Exposure?

Living up a mountain and coming down to train is just one of six or seven ways to get the effect of altitude exposure while maintaining training intensity. We summarize here the various strategies.

Use a Mountain and Valley. Living at altitude and descending to do high-intensity training three or four days a week is the literal way to live high and train low. The main problem is a shortage of suitable mountains, so for most athletes this option means the expense and stress of international travel and of living away from home for up to a month. Loss of heat acclimatization may also be a problem if the high and low training venues are too cool.

Stay High and Train Hard with Oxygen.  If your mountain does not offer the possibility of training low, you may be able to do your hard training on a sport-specific ergometer while you breathe oxygen-enriched air through a face mask. Although this option gives you a wider choice of mountains, you may still have problems with international travel and time away from home, and it doesn't suit all sports (e.g. swimming). There has also been little research on its effectiveness.

Live in a Nitrogen House. The Finns have pioneered use of dwellings that can be flushed with air diluted with nitrogen (Rusko, 1996). Altitudes of around 2500 m can be simulated by reducing the oxygen content from the normal 21% to around 15%. A nitrogen house can be sited almost anywhere as a fixed or mobile facility and is probably the most cost-effective way to deal with teams of athletes. But athletes still have to suffer a dormitory lifestyle away from home.

Rest and Sleep in a Nitrogen Tent. A mini version of a nitrogen house, in the form of a tent, has recently appeared on the market. (See the website.) It simulates altitudes of up to 2700 m (9000 ft) and can be modified to simulate up to 4000 m (14,000 ft). The tent is set up on a bed or on the floor. The advantages are substantial:  it is truly portable; it can be used with little or no disruption of family life, study, or work; and it is easily the best way to establish the altitude and program of exposure that suits the individual. The units are moderately expensive (US$5500), but comparable to the cost of a trip to a mountain and similar in price to other equipment used by top athletes.

Breathe Through a Nitrogen Mask Intermittently. We have heard that Russian researchers are investigating the effects of breathing oxygen-depleted air through a face mask for an hour or two, several times a day. The air has an oxygen content of 10-12%, equivalent to ~5000 m (~17,000 ft). Such treatments can increase the production of erythropoietin (Knaupp et al., 1992), so there is likely to be an increase in red-cell production and an enhancement of endurance performance. Research is needed to determine the effectiveness of the treatment relative to longer exposures in nitrogen houses or tents.

Live in a Large Barometric Chamber. Large steel chambers that can be evacuated to simulate high-altitude flying may be available at a national air-force establishment. This method of achieving altitude exposure has not been used for live-high train-low research yet, but there is little doubt that it would enhance performance. There are no specific advantages, and there are several disadvantages: high running costs, national travel to the chamber, daily travel between the chamber and training venue, a month's absence from home and work, and a severely cramped dormitory lifestyle with little privacy.

Rest and Sleep in a Personal Barometric Chamber. This device consists of a rigid cylinder little bigger than a person, with windows at each end and a vacuum pump attached. It has been available commercially for several years. (See the website.) Like the nitrogen tent, it can be used at home, but it's too cramped to accommodate a partner. It's also twice the price of a nitrogen tent, less easy to use, and less transportable. It may also be more noisy and uncomfortably warm.

Use Erythropoietin or Blood Doping. There is little doubt that some top athletes have been taking injections of erythropoietin to get the increase in red blood-cell mass that normally accompanies altitude exposure. There are no published scientific reports of its effectiveness with athletes, but non-athletes experienced an enhancement in peak running speed of 17% (Ekblom and Berglund, 1991). Intravenous infusion of extra red cells (blood doping) has a similar effect (Sawka et al., 1996). With excessive use, both strategies are dangerous: the blood becomes so thick that there is a risk of sudden death from blood clotting. So for reasons of fair play and health, their use is banned by the International Olympic Committee. Ironically, altitude exposure may be more effective anyway, if the increased buffering capacity of muscles that seems to occur with altitude exposure contributes to the enhancement of performance. (In his review of this article, Allan Hahn questioned whether altitude training could be more effective. Our reply: if erythropoietin injections and blood doping are more effective, it may be only because they increase the red-cell mass to a dangerously high level. We predict an athlete will get more performance enhancement by using altitude exposure to reach the permissible limit of red cells in the blood than by using erythropoietin or blood doping to reach the same limit.)

Ethics of Altitude Exposure

There are two good reasons for banning a practice that enhances performance: either it causes illness or injury, or it gives the athlete a technological advantage that is too expensive or too new for most other competitors to use. Let's see whether the different methods of altitude exposure should be banned.

Living on a mountain is obviously ethically acceptable--in fact, it probably seems romantic to the public, who ultimately determine the ethical stance of sports bodies. Frequent trips down to the valley are also unlikely to be considered unsporting. But aside from the temporary altitude sickness, is altitude exposure damaging to health? In a small proportion of the population, continuous exposure to altitude leads to accumulation of fluid (edema) in the lungs and brain, which can be fatal (Krasney, 1994). In another small proportion, excessive production of red blood cells increases the risk of sudden death through blood clotting or a heart attack (Coote, 1995). The average athlete who spends a few weeks at a moderate altitude will not have these problems.

Altitude chambers, nitrogen houses and nitrogen tents would be dangerous if the simulated altitude was high enough and long enough to raise the thickness of blood to an unsafe level.  An athlete using a personal altitude chamber or tent might well overdo it, but so far no one has made a public case for banning these devices on the grounds of health or safety.  It also seems unlikely they will be banned as an expensive innovation, because they are no more expensive than the high-tech equipment used in training or performance by many Olympic athletes.  If they aren't unethical, are they unsporting?  Perhaps.  A government-sponsored altitude facility is reminiscent of the Eastern Bloc's approach to national Olympic glory.  Somehow it's less objectionable if the individual athlete pursues this avenue of performance enhancement via a personal altitude chamber or tent. Still, it will be a sad day when all endurance athletes have to spend weeks of their lives in such apparatus to keep up with other competitors.  Can they be banned?   No, because you can't ban normal altitude training, so it's unfair to ban a safe practice that makes it easier or cheaper for athletes to achieve the same effect.

The practices are safe only if the blood doesn't get too thick. Partly for this reason, some sports have set limits either on the proportion of blood cells (the hematocrit) in a blood sample (Martin et al., 1997), or on the concentration of the red-cell protein, hemoglobin (Seiler, 1997). Research currently in progress will determine which of these tests is more effective, where to set a fair limit, and how to deal with athletes' attempts to cheat the test.  The resulting blood test may be a simple and highly effective way to keep use of real and simulated altitude exposure within safe limits.

Table 1: Studies of living and training high 



Time post-altitude, performance test, and outcome (+ = better, - = worse) 

Controlled Studiesa 

Gore et al., 1997 

13+8 runners, 28 d at 1740 m 




3.2-km run 


8+8 runners, 28 d at 1300 m 




3.2-km run 


Levine & Stray-Gundersen, 1997 

13 + 13 runners, 28 d at 2500 m 

4 d 



3-21 d 

5-km run 


Burtscher et al., 1996 

10+12 runners, 12 d at 2300 m 

3 d & 16 d 

VO2max (cycle) 

+1.4% & +8.7%* 

Rusko et al., 1996 

14+7 skiers, 18-28 d at 1600-1800 m 

<8 d 



max anaerobic power 


Telford et al., 1996 

9 + 9 runners, 28 d mainly at 1800 m 

<8 d 



~3-min run to exhaustion 

-0.6% b 

3.2-km run 


Martino et al., 1995 

20+13 swimmers, 21 d at 2800 m 


100-m swim 


anaerobic tests 


Jensen et al., 1993 

9 + 9 elite rowers (non-random assignment), 21 d at 1800 m 




6-min row 


Levine & Stray-Gundersen, 1992 

9 + 10 runners, 28 d at 2500-3000 m 




5-km run 


Karvonen et al., 1986 

3 + 4-6 sprinters, 21 d at 1850 m 




peak run speed 


~1-min run to exhaustion 


30-m run 



-6.1% to +8.8% 

Rahkila & Rusko, 1982c 

6 skiers + 8 skiers and runners, 11 d at 2600 m 



"small reduction" 

stair climb 

"not significant" 

1-min cycle 

"not significant" 

Adams et al., 1975 

6 + 6 runners (crossover), 20 d at 2300 m 

1 d 



3 d 

2-mile run 

+1.3% (control not stated) 

Uncontrolled Studies 

Mizuno et al., 1990 

10 elite skiers, 14 d at 2100-2700 m 

2 d 


"not affected" 

~5-min run to exhaustion 


Dill & Adams, 1971 

6 elite runners, 17 d at 3100 m 

<1 d 



~10-min run to exhaustion 


Daniels & Oldridge, 1970 

6 elite runners, 14, 14 & 7 d at 2300 m over 45 d 

<8 d 





1- or 3-mile runs 

"14 personal bests" 

Buskirk et al., 1967 

6 runners, 63 d at 4000 m 

3-15 d 



440-yd to 5-mile runs 

 0.0% to -4.0% 

Faulkner et al., 1967 

15 swimmers, 14 d at 2300 m 




100- to 500-yd swims 

-0.8% to +1.2% 

Saltin, 1967c 

9 elite athletes, 19 d at 2250 m 



worse in 7 of the 9 subjects 

VO2max: maximum oxygen consumption. 
*Unlikely to be zero (statistically significant at the 5% level). 
aSample sizes are altitude + control (sea-level) groups respectively. Performance change is relative to control group. 
bPercent change in time to exhaustion has been divided by 14 (10-min tests) through 20 (1-min tests) to convert it to approximate change in a time trial (WGH, unpublished observations). 
cWe were unable to access the original reference. Data shown are from the review by Hahn (1991). 


Table 2: Studies of living high and training low 



Time post-altitude, performance test, and outcome (+ = better, - = worse) 

Controlled Studiesa 

Levine & Stray-Gundersen, 1997 

13 + 13 runners, 28 d at 2500/1250 m 

4 d 



3-21 d 

5-km run 


Stray-Gundersen & Levine, 1997 

13 runners, 28 d at 2500/2700/1250 m ("high-high-low") 

4 d ? 


 0.0% relative to high-low 

3 d ? 

5-km run 

+0.2% relative to high-low 

Nummela et al., 1996 

6 + 6 runners, 10 d in nitrogen house set to 2200 m 

<8 d 

peak running speed 


400-m run 

+1.0%* control not stated 

Levine et al., 1991 

6 + 3 runners, 28 d at 2500/1300 m 




5-km run 


Uncontrolled Studies 

Chapman et al., 1998 

22 elite runners, 28 d at 2500/2700/1250 m ("high-high-low") 

3 d 

3-km run 


Mattila & Rusko, 1996 

5 cyclists, 11 d in nitrogen house set to 3000 m 

5 d 

?-km cycle 


Stray-Gundersen & Levine, 1994 

6 runners, 28 d at 2500/1250 m 

0 d & 14 d 


-3.6% & -4.0% 

5-km run 

+0.9% & -1.3% 

~3-min run to exhaustion 

+1.0% &  0.0%b 

Chick et al., 1993 

5 subjects, 6 weeks of hyperoxic cycling at 1600 m 

0 d 

19-min cycle 


~6-min cycle to exhaustion 


VO2max: maximum oxygen consumption. 
*Unlikely to be zero (statistically significant at the 5% level). 
aSample sizes are altitude + control (sea-level) groups respectively. Performance change is relative to control group. 
bPercent change in time to exhaustion has been divided by 14 (10-min tests) through 20 (1-min tests) to convert it to approximate change in a time trial (WGH, unpublished observations). 


Adams, W.C., Bernauer, E.M., Dill, D.B., & Bomar, J.B. (1975). Effects of equivalent sea level and altitude training on V02max and running performance. Journal of Applied Physiology, 39, 262-266.

Baumann, I., Bonov, P., Daniels, J., & Lange, G. (1994). NSA Round Table: high altitude training. New Studies in Athletics, 9(2), 23-35.

Bovard, R., Schoene, R.B., & Wappes, J.R. (1995). Don't let altitude sickness bring you down. Physician and Sportsmedicine, 23(2), 87-88.

Buick, F.J., Gledhill, N., Froese, A.B., Spriet, L, & Meyers, E.C. (1980). Effect of induced erythrocythemia on aerobic work capacity. Journal of Applied Physiology, 48, 636-642.

Burtscher, M., Nachbauer, W., Baumgartl, P, & Philadelphy, M. (1996). Benefits of training at moderate altitude versus sea level training in amateur runners. European Journal of Applied Physiology, 74, 558-563.

Buskirk, E.R., Kollias, J., Akers, R.F., et al. (1967). Maximal performance at altitude and on return from altitude in conditioned runners. Journal of Applied Physiology, 23, 259-266.

Chapman, R.F., Stray-Gundersen, J., & Levine, B.D. (1998). Individual variation in response to altitude training. Journal of Applied Physiology, in press.

Chick, T.W., Stark, D.M., & Murata, G.H. (1993). Hyperoxic training increases work capacity after maximal training at moderate altitude. Chest, 104, 1759-1762.

Coote, J. H. (1995). Medicine and mechanisms in altitude sickness: recommendations. Sports Medicine, 20, 148-159.

Daniels, J., & Oldridge, N. (1970). The effects of alternate exposure to altitude and sea level on world-class middle-distance runners. Medicine and Science in Sports, 2, 107-112.

Dill, D.B., & Adams, W.C. (1971). Maximal oxygen uptake at sea level and at 3,090-m altitude in high school champion runners. Journal of Applied Physiology, 30, 854-859.

Ekblom, B., & Berglund, B. (1991). Effect of erythropoietin administration on maximal aerobic power. Scandinavian Journal of Medicine and Science in Sports, 1, 88-93.

Faulkner, J.A., Daniels, J.T., & Balke, B. (1967). The effects of training at moderate altitude on physical performance capacity. Journal of Applied Physiology, 23, 85-89.

Goforth, H.W., Campbell, N.L., Hodgdon, J.A., & Sucec, A.A. (1982). Hematological parameters of trained distance runners following induced erythrocythemia. Medicine and Science in Sports and Exercise, 14, 174 (Abstract).

Gore, C.J., Hahn, A.G., Burge, C.M., & Telford, R.D. (1997). VO2max and haemoglobin mass of trained athletes during high intensity training. International Journal of Sports Medicine, 18, 477-482.

Hahn, A.G. (1991). The effect of altitude training on athletic performance at sea level - a review. Excel, 7(2), 9-23.

Jensen, K., Nielsen, T.S., Fiskastrand, A., et al. (1993). High-altitude training does not increase maximal oxygen uptake or work capacity at sea level in rowers. Scandinavian Journal of Medicine and Science in Sports, 3, 256-262.

Karvonen, J., Peltola, E., & Saarela, J. (1986). The effect of sprint training performed in a hypoxic environment on specific performance capacity. Journal of Sports Medicine and Physical Fitness, 26, 219-224.

Knaupp, W., Khilnani, S., Sherwood, J., et al. (1992). Erythropoietin response to acute normobaric hypoxia in humans. Journal of Applied Physiology, 73, 837-40.

Krasney, J. A. (1994). A neurogenic basis for acute altitude illness. Medicine and Science in Sports and Exercise, 26, 195-208.

Levine, B.D., & Stray-Gundersen, J. (1992). Altitude training does not improve running performance more than equivalent training near sea level in trained runners. Medicine and Science in Sports and Exercise, 24, S95 (Abstract 569).

Levine, B.D., & Stray-Gundersen, J. (1997). "Living high-training low": effect of moderate-altitude acclimatization with low-altitude training on performance. Journal of Applied Physiology, 83, 102-112.

Levine, B.D., Stray-Gundersen, J., Duhaime, G., et al. (1991). "Living high - training low": the effect of altitude acclimatization/normoxic training in trained runners. Medicine and Science in Sports and Exercise, 23, S25 (Abstract 145).

MacDougall, J.D., Green, H.J., Sutton, J.R., et al. (1991). Operation Everest II: structural adaptations in skeletal muscle in response to extreme simulated altitude. Acta Physiologica Scandinavica, 141, 421-427.

Martin, D.T., Ashenden, M., Parisotto, R., et al. (1997). Blood testing for professional cyclists: what's a fair hematocrit limit? Sportscience News (Mar-Apr).

Martino, M., Myers, K., & Bishop, P. (1995). Effects of 21 days training at altitude on sea-level anaerobic performance in competitive swimmers. Medicine and Science in Sports and Exercise, 27, S7 (Abstract 37).

Matttila, V., & Rusko, H. (1996). Effect of living high and training low on sea level performance in cyclists. Medicine and Science in Sports and Exercise, 28, S156 (Abstract 928).

Mizuno, M., Juel, C., Bro-Rasmussen, T., et al. (1990). Limb skeletal muscle adaptations in athletes after training at altitude. Journal of Applied Physiology, 68, 496-502.

Nummela, A., Jouste, P., & Rusko, H. (1996). Effect of living high and training low on sea level anaerobic performance in runners. Medicine and Science in Sports and Exercise, 28, S124 (Abstract 740).

Rahkila, P., & Rusko, H. (1982). Effect of high altitude training on muscle enzyme activities and physical performance characteristics of cross-country skiers. In P.V. Komi (ed.) Exercise and Sport Biology&#150;International Series on Sports Sciences, Volume 12. Human Kinetics: Champaign Il., pp. 143-151.

Rusko, H.R. (1996). New aspects of altitude training. American Journal of Sports Medicine, 24(6), S48-S52.

Rusko, H.K., Kirvesniemi, H., Paavonlainen, L., et al. (1996).Effect of altitude training on sea level aerobic and anaerobic power in elite athletes. Medicine and Science in Sports and Exercise, 28, S124, (Abstract 739).

Saltin, B. (1967). Aerobic and anaerobic work capacity at an altitude of 2,250 meters. In R.F. Goddard (ed.), The effects of altitude on physical performance, Athletic Institute: Chicago, pp. 97-102.

Sawka, M.N., Joyner, M.J., Miles, D.S., et al. (1996). The use of blood doping as an ergogenic aid. Medcicine and Science in Sports and Exercise, 28(6), i-viii.

Seiler, S. (1997). Tighter control on EPO use by skiers. Sportscience News (Jan-Feb).

Stray-Gundersen, J., Alexander, C., Hochstein, A., et al. (1992). Failure of red cell volume to increase to altitude exposure in iron deficient runners. Medicine and Science in Sports and Exercise, 24, S90 (Abstract 540).

Stray-Gundersen, J., & Levine, B.D. (1994). Altitude acclimatization/normoxic training (High/Low) improves sea level endurance immediately on descent from altitude. Medicine and Science in Sports and Exercise, 26, S64 (Abstract 360).

Stray-Gundersen, J., & Levine, B.D. (1997). "Living high-training high and low" is equivalent to "living high-training low" for sea-level performance. Medicine and Science in Sports and Exercise, 29, S136 (Abstract 783).

Szygula, Z. (1990). Erythrocytic system under the influence of physical exercise and training. Sports Medicine, 10, 181-97.

Telford, R.D., Graham, K.S., Sutton, J.R., et al. (1996). Medium altitude training and sea-level performance. Medicine and Science in Sports and Exercise, 28, S124, (Abstract 741).

Wolski, L.A., McKenzie, D.C., & Wenger, H.A. (1996). Altitude training for improvements in sea level performance: is there scientific evidence of benefit? Sports Medicine, 22, 251-263. · · Homepage · Copyright ©1998
Edited by Mary Ann Wallace · Webmastered by Will Hopkins
Last updated 20 July 98