An Attempt to Quantify the Placebo Effect From a Three-Week Simulated Altitude Training Camp in Elite Race Walkers

International Journal of Sports Physiology and Performance, 2010, 5, 521-534 2010 Human Kinetics, Inc. An Attempt to Quantify the Placebo Effect From a Three-Week Simulated Altitude Training Camp in Elite Race Walkers Philo U. Saunders, Christoph Ahlgrim, Brent Vallance, Daniel J. Green, Eileen Y. Robertson, Sally A.Clark, Yorck O. Schumacher, and Christopher J. Gore Purpose: To quantify physiological and performance effects of hypoxic exposure, a training camp, the placebo effect, and a combination of these factors. Methods: Elite Australian and International race walkers (n = 17) were recruited, including men and women. Three groups were assigned: 1) Live High:Train Low (LHTL, n = 6) of 14 h/d at 3000 m simulated altitude; 2) Placebo (n = 6) of 14 h/d of normoxic exposure (600 m); and 3) Nocebo (n = 5) living in normoxia. All groups undertook similar training during the intervention. Physiological and performance measures included 10-min maximal treadmill distance, peak oxygen uptake (VO 2 peak), walking economy, and hemoglobin mass (Hb mass ). Results: Blinding failed, so the Placebo group was a second control group aware of the treatment. All three groups improved treadmill performance by approx. 4%. Compared with Placebo, LHTL increased Hb mass by 8.6% (90% CI: 3.5 to 14.0%; P =.01, very likely), VO 2 peak by 2.7% ( 2.2 to 7.9%; P =.34, possibly), but had no additional improvement in treadmill distance ( 0.8%, 4.6 to 3.8%; P =.75, unlikely) or economy ( 8.2%, 24.1 to 5.7%; P =.31, unlikely). Compared with Nocebo, LHTL increased Hb mass by 5.5% (2.5 to 8.7%; P =.01, very likely), VO 2 peak by 5.8% (2.3 to 9.4%; P =.02, very likely), but had no additional improvement in treadmill distance (0.3%, 1.9 to 2.5%; P =.75, possibly) and had a decrease in walking economy ( 16.5%, 30.5 to 3.9%; P =.04, very likely). Conclusion: Overall, 3-wk LHTL simulated altitude training for 14 h/d increased Hb mass and VO 2 peak, but the improvement in treadmill performance was not greater than the training camp effect. Keywords: hypoxia, live high:train low, hemoglobin mass, peak oxygen uptake, endurance training Philo U. Saunders is with Department of Physiology, Australian Institute of Sport, Belconnen, ACT, Australia. Christoph Ahlgrim is with the Department of Sports Medicine, University of Freiburg, Freiburg, Germany. Brent Vallance is with Track and Field, Australian Institute of Sport, Belconnen, ACT, Australia. Daniel J. Green is with the Department of Physiology, Australian Institute of Sport, Belconnen, ACT, Australia. Eileen Y. Robertson is with the Department of Physiology, Australian Institute of Sport, Belconnen, ACT, Australia. Sally A. Clark is with the Department of Physiology, Australian Institute of Sport, Belconnen, ACT, Australia. Yorck O. Schumacher is with the Department of Sports Medicine, University of Freiburg, Freiburg, Germany. Christopher J. Gore is with the Department of Physiology, Australian Institute of Sport, Belconnen, ACT, Australia, and the School of Education, Flinders University, Adelaide, SA, Australia. 521

522 Saunders et al. Altitude training has been used extensively over the last 40 y with evidence that worthwhile (1 4%) improvements in performance of elite athletes can be achieved. 1 4 Due to the nature of altitude exposure, it has been difficult to determine whether the performance effect of altitude has been due to altitude exposure, a training camp effect, a placebo effect, or a combination of these factors. It has previously been observed, that performance has been consistently improved in multiple groups of athletes just by getting athletes together for a period of time where they can focus on training, the so called training camp effect. 5 To counter this effect, a 6-wk lead in period before the primary intervention component of the live high:train low (LHTL) method has been used. 2 It has also been shown that a substantial improvement in racing performance between the first and the second measurement before the intervention occurred in very experienced athletes and this improvement was at least, in part, attributed to the training camp effect. 6 Although most subsequent altitude studies have utilized a matched control group in an attempt to account for the training camp effect, the placebo effect of altitude has not been fully investigated. The new simulated altitude facility at the Australian Institute of Sport (AIS) allows individual control of all rooms so some can be at altitude and others near sea level. Current studies from our laboratory, in both elite runners and cyclists, have resulted in worthwhile improvements in performance and a 3 4% increase in Hb mass using 3 wk of 14 h d 1 LHTL. 7,8 These results have also been demonstrated by others when a sufficient dose of LHTL has been used. 1,2,9 It has also been demonstrated that LHTL improves sea-level performance by 1 3% in nationally competitive endurance athletes. 2,4 However, we are unsure if the training camp and placebo effects account for some, or even all, of this increase in performance and this unresolved issue needs to be addressed. The placebo effect is a favorable outcome purely from the belief that one has received a beneficial treatment. Despite the widespread belief that altitude training improves performance in endurance events, 2,10 the placebo effect of caffeine and carbohydrate has been quantified at approx. 1 3% 11,12 and could also be responsible for some or all of the improvements (also 1 3%) 13 from altitude training studies. The current study was designed to determine whether the improvement in performance and physiological measures are due to simulated altitude exposure, a well-planned training camp, or the placebo effect of simulated altitude exposure. Subjects Methods Internationally competitive male and female race walkers (n = 17) were recruited for the study. Subjects were in good fitness coming into the study being in midseason build up for the Australian Championships 20 km event. Subject characteristics are presented in Table 1. All subjects were informed of the experimental procedures and possible risks involved with participation before written consent was obtained. The study and all testing procedures were approved by the Australian Institute of Sport Ethics Committee. Study Design The current study was a single blinded design where athletes were not informed about which group they were allocated. Subjects were randomly divided into three groups (LHTL, Placebo or Nocebo) matched on recent race times, training

The Placebo Effect of Simulated Altitude Training 523 Table 1 Subject characteristics at start of study. Values are mean ± standard deviation. Sum 7 skinfolds included measurements from bicep, triceps, sub scapular, supra spinale, abdominal, thigh and calf. Variable Simulated Altitude (n = 6; 3 male, 3 female) Placebo (n = 6; 3 male, 3 female) Nocebo (n = 5; 2 male, 3 female) Age (y) 24.5 ± 4.5 23.2 ± 4.6 26.6 ± 4.9 Height (cm) 171.9 ± 6.3 173.0 ± 10.3 176.3 ± 9.1 Body mass (kg) 58.4 ± 7.1 60.6 ± 11.2 63.6 ± 8.7 Sum 7 skinfolds (mm) 56.4 + 20.9 57.1 ± 25.3 65.4 ± 27.7 VO 2 peak (ml min 1 kg 1 ) 59.3 ± 6.9 61.2 ± 9.7 60.3 ± 7.4 vvo 2 peak (km h 1 ) 14.8 ± 1.4 14.3 ± 1.5 14.3 ± 1.7 10 km PB (min:s) 44:43 ± 4:29 46:14 + 3:13 44:07 ± 4:38 Training (km wk 1 ) 111 ± 37 103 ± 29 91 ± 50 Training (h:min wk 1 ) 9:21 ± 3:01 8:34 ± 2:05 7:28 ± 3:43 Note. vvo 2 peak = treadmill velocity at VO 2 peak; PB = personal best; Training is an average throughout the study period and is a combination race walking and running. history, sex and VO 2 peak, and each group spent 5 wk at the Australian Institute of Sport (AIS) where they trained together under the guidance of the AIS race walking coach. The training program during each week of the study period involved 3 4 continuous walking sessions at intensities ranging between light aerobic and threshold, up to three light aerobic runs, two interval-based sessions at or above race pace intensity using intervals of 1 2 km, and one hill session which was done at race pace intensity and consisted of a combination of continuous hill walking up to 14 km and shorter hill intervals of approx. 1 km. Average weekly training for the study period is given in Table 1. All athletes were tested under normoxic conditions in the AIS physiology laboratory in Canberra (600 m of altitude) in the week before a 21-d intervention period and within 5 d after the intervention, with weekly measures of blood parameters (Figure 1). To prevent iron-deficient anemia, participating athletes were ingesting daily the equivalent of 100 mg of elemental iron as oral medication (Ferro-Grad C, Abbott Australasia Pty Ltd, Australia) throughout the study period beginning from 2 d before altitude exposure. The altitude intervention group (LHTL, n = 6) spent 14 h d 1 at a simulated altitude of 3000 m for 21 consecutive days in the AIS Altitude House, which is a normobaric hypoxic facility. All training was done at normoxia (600 m) during an Australian summer (January/February 2008). The placebo control group (Placebo, n = 6) also spent 14 h d 1 in the AIS altitude house, but their rooms were kept at the ambient altitude of 600 m. Athletes in both groups spent no more than 2 h d 1, and no more than 1 h at a time in the kitchen of the Altitude House, which was maintained at a simulated altitude of 3000 m. The nocebo control group (Nocebo, n = 5) lived in Canberra during the intervention period and acted as a control group undertaking the same training with the other two groups. Unlike Placebo, the Nocebo group were aware that they were not at simulated altitude.

524 Saunders et al. Figure 1 Study time-line indicating testing before (negative numbers), during (no sign on numbers) and after (positive numbers) the 3-wk intervention period. The LHTL and Placebo groups were told that they could be assigned to either group but only informed about the actual group at the conclusion of all testing and races. The 2 h d 1 that Placebo spent at 3000 m is less than the 3 h d 1 at 4500 5500 m, which was insufficient to produce any increase in reticulocytes, soluble transferrin receptor (stfr), Hb mass, and red cell volume as demonstrated by Gore et al. 14 Moreover, even 60 min at 5450 m is not enough to increase serum erythropoietin, which requires at least 2 h of continuous exposure to be elevated. 15 Therefore, there is strong scientific evidence to support the notion that the total of 2 h d 1 in hypoxia for Placebo will not induce an increase in the total number of red blood cells. On completion of the intervention and posttesting, all LHTL and Placebo athletes completed a questionnaire to ascertain which group they thought they were in, that is, simulated altitude or placebo, and the certainty of their selection on a scale of 1 to 5, with 5 being complete certainty. Methodology A treadmill test was used to assess peak oxygen uptake (VO 2 peak), walking economy, velocity at VO 2 peak (vvo 2 peak) and maximal distance covered in a 10-min performance test, on a custom-built motorized treadmill (Australian Institute of Sport). This test involved continuous walking for 4 min at 4 5 incrementally faster speeds ranging from 9 15 km h 1 at 0% gradient with a 1 min break between each speed. Five minutes after the last submaximal walking speed, the subject performed a maximal 10 min walk at a gradient of 4% and the speed self-selected with the

The Placebo Effect of Simulated Altitude Training 525 walker able to modify speed every 30 s. The treadmill speed was not blinded to the subjects or the testers. The subjects were instructed to walk the furthest distance in the 10 min after being paced for the initial 2 min at a speed calculated from their performance results. The VO 2 peak was determined as the highest value that was held for 60 s during this 10-min performance trial. Expired ventilation samples collected by a custom built open-circuit indirect calorimetry system with associated in-house software for determination of oxygen uptake described in full previously. 16 Before the study we conducted pilot testing on five subjects in the study with both the 10-min protocol and the incremental protocol normally used in our laboratory. 17 The difference in VO 2 peak between the protocols was 0.1 ± 1.5 ml min 1 kg 1, and using a Bland-Altman analysis 18 the mean bias between the 10 min protocol and the incremental protocol expressed as a percent was 0.1% (95% confidence interval: 3.1 to 2.9%) which provided solid data that the VO 2 peak achieved with either protocol are equal. In addition, the typical error (TE) associated with equipment, testing and biological variation, established in our laboratory was 2.4% (90% confidence interval: 1.7 to 4.6%) for submaximal VO 2 and 2.1% (1.4 to 4.1%) for VO 2 peak from duplicate measures conducted upon 10 subjects before commencement of this study. Before the intervention period, each participant had 2 11 ml samples of venous blood drawn after 15 min of supine rest for duplicate baseline measurement of full blood count (FBC), ferritin, percentage of reticulocytes and soluble transferrin receptor (stfr). Samples were then withdrawn weekly and at the end of the intervention. FBC and reticulocyte data was analyzed within 4 h after withdrawal using an Advia 120 Hematology Analyzer (Bayer Diagnostics, Tarrytown, NY). Serum tubes were spun for 5 min immediately after withdrawal and frozen at 80 C. After completion of the study, all samples were thawed and analyzed as a batch for stfr and ferritin on an automated Hitachi 911 Automatic Analyzer (Boehringer Mannheim, Germany). Ferritin index, a marker displaying functional and storage iron compartment, was calculated using stfr and Ferritin values 19 to display relative iron deficiency due to erythropoiesis, which would be illustrated by an increased Ferritin index. The TE (and 90% confidence interval) of hematological parameters as taken from the double baseline measures from the 17 subjects in the current study are as follows: Hematocrit = 3.7% (2.8 to 5.8%), hemoglobin concentration ([Hb]) = 1.8% (1.4 to 2.8%), blood volume = 3.2% (2.4 to 5.1), plasma volume = 4.5% (3.4 to 7.1%), red cell volume = 3.4% (2.6 to 5.3%), ferritin = 18.4% (14.7 to 32.3%), stfr = 12.4% (9.7 to 20.8%), ferritin index = 11.2% (8.7 to 18.6%), and reticulocytes = 10.5% (8.1 to 17.3%). Total hemoglobin mass (Hb mass ) was measured with the optimized 2 min carbon monoxide (CO) rebreathing test adapted from Schmidt and Prommer. 20 This parameter was measured twice before the intervention period to establish a double baseline value, weekly during the study period, and approximately one week after the completion of the intervention (Figure 1). The TE of Hb mass was 2.2% (1.6 to 3.4%) as determined from the double baseline measures from the 17 subjects in the current study. Blood volumes (total blood volume, red cell volume and plasma volume) were calculated using [Hb] and mean cell hemoglobin concentration (MCHC) from FBC analysis and Hb mass from the CO rebreathing test as described by equations as follows:

526 Saunders et al. Total blood volume (ml) = Hb mass (g) 100 / [Hb] (g/dl) Red cell volume (ml) = Hb mass (g) 100 / MCHC (g/dl) Plasma volume (ml) = blood volume (ml) red cell volume (ml) 21 Statistical Analysis Physiological and performance measures were log-transformed for the analyses to reduce bias arising from any nonuniformity of error and back-transformed to obtain, as percents, the changes in means and variation. Simple group statistics are shown as mean ± between-subject standard deviation. Mean effects and their 90% confidence limits were estimated with a spreadsheet for randomized controlled trials (http://www.sportsci.org/resource/stats/xparallelgroupstrial.xls) using unpaired t tests. The magnitude of differences between change scores (pre- versus post-) were expressed as the effect of the altitude intervention compared with control conditions (Placebo and Nocebo) with the mean percentage change and the lower and upper 90% confidence intervals. The percentage likelihood are expressed qualitatively (<0.5%, almost certainly not; 0.5 5%, very unlikely; 5 25%, unlikely or probably not; 25 75%, possibly; 75 95%, likely or probable; 95 99.5%, very likely; >99.5% most likely or almost certainly) based on previous recommendations. 22 These statistical methods are described in full by Hopkins et al. 23 To evaluate the time course of hematological parameters in each group we used a paired t test to evaluate the mean change 1 wk after the intervention (within one group, not comparing between groups) with Pre, 1, 2 and 3 wk of intervention. This mean change was calculated by averaging every subject s individual change (eg, post- to pre-) and then the distribution of the mean change were analyzed. As we could not be certain about the direction of the change, a two-tailed t test was used with significance set at P <.1. Results All subjects in both LHTL and Placebo correctly identified their group allocation with a certainty of 4.3 ± 0.9 on a scale of 1 to 5. Because our blinding failed, hereafter the Placebo group will be denoted as Plaecbo (F), to indicate that the it was a failed placebo group. Treadmill performance during the 10-min walk for maximal distance was 3 4% better in all groups within 5 d after the 21 d intervention period (Figure 2). However, there was no difference between LHTL and Placebo (F) ( 0.7, 4.8 to 3.6%; mean and 90% confidence interval) with only a 23% (P =.75, very unlikely) likelihood of LHTL improving performance compared with Placebo (F). Similarly, there was no difference between LHTL and Nocebo (0.3, 1.9 to 2.5) with a 28% (P =.84, unlikely) likelihood of LHTL improving performance compared with Nocebo. When Placebo (F) was compared with Nocebo, the difference in performance was 1.0% ( 3.2 to 5.3%; likelihood 49%, P =.67, possibly). Three weeks of LHTL simulated altitude improved Hb mass by 8.6% (3.5 to 14.0%) with a likelihood of 99% (P =.01, very likely) when compared with Placebo (F), and by 5.5% (2.5 to 8.7%) with a likelihood of 99% (P =.01, very likely) when compared with Nocebo. When Placebo (F) was compared with Nocebo, Hb mass

The Placebo Effect of Simulated Altitude Training 527 Figure 2 Individual changes (gray lines) in performance (distance covered) and peak oxygen uptake (VO 2 peak) during 10 min race walking test on treadmill at a set grade of 4%. Means ± SD are represented by black lines. LHTL (n = 6), Placebo (n = 5), and Nocebo/ Control (n = 5). for Placebo (F) was 4.0% ( 7.4 to 0.5%; likelihood 93%, P =.06, likely) lower. Figure 3 presents individual and mean responses of Hb mass during the intervention. The time course of further hematologic parameters throughout the study period is displayed in Table 2. One female subject in LHTL initially was iron depleted (baseline ferritin: 9.1 ng ml 1, stfr: 4.0 mg L 1, ferritin-index: 4.3) but normal ferritin levels were observed after 1 wk of iron supplementation (ferritin 20.6 ng ml 1 ) Throughout the intervention period, no further ferritin levels below 20 ng ml 1 were observed in any subject.

Figure 3 Individual changes (gray lines) in hemoglobin mass (Hb mass ) before, weekly during the 3-wk intervention and after. Means ± SD are represented by black lines. LHTL (n = 6), Placebo (n = 6), and Nocebo/Control (n = 5). 528

Table 2 Hematologic parameters before and each week during the intervention period. Values are mean ± standard deviation. Variable Before Week 1 Week 2 Week 3 ALT (n = 6) Hb mass (g) 733 ± 179 746 ± 200 766 ± 187* 778 ± 183* Red cell volume (ml) 1955 ± 483 2057 ± 516* 2129 ± 505* 2192 ± 484* Plasma volume (ml) 3536 ± 559 3421 ± 651* 3456 ± 708 3399 ± 644* Blood volume (ml) 5492 ± 1033 5478 ± 1152 5585 ± 1179 5591 ± 1101 [Hb] (g dl 1 ) 13.2 ± 1.1 13.5 ± 1.0 13.6 ± 0.9 13.8 ± 1.0* Hct (%) 0.35 ± 0.02 0.37 ± 0.02* 0.38 ± 0.03* 0.39 ± 0.02* Reticulocytes (%) 1.7 ± 0.4 2.1 ± 0.6* 2.1 ± 0.4* 1.8 ± 0.3 Ferritin (ng ml 1 ) 39.0 ± 22.3 39.5 ± 17.6 39.2 ± 15.8 33.9 ± 13.7 stfr (mg L 1 ) 2.08 ± 1.04 1.87 ± 0.67 2.03 ± 0.64 1.98 ± 0.61 stfr (mg L 1 ) 1.69 ± 0.48 1.64 ± 0.42 1.88 ± 0.58 1.86 ± 0.60 Ferritin index 1.61 ± 1.37 1.25 ± 0.57 1.30 ± 0.39 1.33 ±0.42 Ferritin index 1.07 ± 0.37 1.04 ± 0.30 1.20 ± 0.35 1.24 ± 0.42 Placebo (n = 6) Hb mass (g) 840 ± 258 817 ± 259* 835 ± 242 815 ± 228 Red cell volume (ml) 2302 ± 732 2242 ± 705* 2304 ± 671 2261 ± 657 Plasma volume (ml) 3421 ± 894 3542 ± 829 3690 ± 818* 3566 ± 896* Blood volume (ml) 5723 ± 1613 5783 ± 1522 5994 ± 1484* 5827 ± 1550 [Hb] (g dl 1 ) 14.6 ± 0.8 14.0 ± 1.0* 13.8 ± 0.7* 14.0 ± 0.4* Hct (%) 0.40 ± 0.03 0.39 ± 0.03* 0.38 ± 0.02* 0.39 ± 0.02* Reticulocytes (%) 1.4 ± 0.1 1.7 ± 0.4* 1.9 ± 0.4* 1.6 ± 0.3 Ferritin (ng ml 1 ) 40.6 ± 16.5 45.0 ± 21.6 36.0 ± 20.0 39.0 ± 26.0 stfr (mg L 1 ) 2.20 ± 0.61 1.87 ± 0.52* 1.88 ± 0.43 1.92 ± 0.31 Ferritin index 1.42 ± 0.40 1.20 ± 0.42 1.31 ± 0.45 1.34 ± 0.47 Nocebo (n = 5) Hb mass (g) 803 ± 228 814 ± 235 817 ± 244 809 ± 230 Red cell volume (ml) 2184 ± 615 2229 ± 623 2248 ± 634* 2245 ± 624 Plasma volume (ml) 3689 ± 912 3809 ± 977 3766 ± 887 3635 ± 895 Blood volume (ml) 5873 ± 1523 6039 ± 1584 6014 ± 1513 5880 ± 1503 [Hb] (g dl 1 ) 13.6 ± 0.4 13.4 ± 0.9 13.5 ± 0.7 13.7 ± 0.8 Hct (%) 0.37 ± 0.01 0.37 ± 0.02 0.37 ± 0.02 0.38 ± 0.02 Reticulocytes (%) 1.7 ± 0.7 1.7 ± 0.4 1.6 ± 0.4 1.6 ± 0.3 Ferritin (ng ml 1 ) 26.8 ± 14.0 32.7 ± 15.0* 41.1 ± 17.2* 31.2 ± 11.3 stfr (mg L 1 ) 2.00 ± 0.49 1.96 ± 0.60 2.16 ± 0.24 1.74 ± 0.32* Ferritin index 1.48 ± 0.48 1.31 ± 0.32 1.38 ± 0.23 1.19 ± 0.23* After exclusion of a subject who initially was iron deficient. *Indicates significant (P <.1) group difference from baseline measure as yielded by a two-tailed matched pair analysis (t test). 529

530 Saunders et al. The VO 2 peak was increased by 5.4 ± 4.1% in LHTL (Figure 2). Compared with Placebo (F), VO 2 peak of LHTL was 2.7% ( 2.2 to 7.9%) higher with a likelihood of 72% (P =.34, possibly). Compared with Nocebo, VO 2 peak of LHTL was 5.8% (2.3 to 9.4%) higher with a likelihood of 98% (P =.02, very likely). When Placebo (F) was compared with Nocebo, VO 2 peak of Placebo (F) was 3.0% ( 1.4 to 7.6%; likelihood 80%, P =.23, likely) higher. Economy, measured by the slope of speed versus VO 2 for the 4 5 walking speeds was not improved by LHTL compared with Placebo (F) ( 8.2, 24.1 to 5.7%; likelihood 13%, P =.31, unlikely) and was worse compared with Nocebo ( 16.5, 3.9 to 30.5%; likelihood 1%, P =.04, very likely). When Placebo (F) was compared with Nocebo, economy of Placebo (F) was 7.6% ( 7.0 to 24.5%; likelihood 78%, P =.37, likely) worse. The vvo 2 peak was better in all groups after the 21 d intervention period: vvo 2 peak increased in LHTL by 3.4 ± 3.4%, Placebo (F) by 5.4 ± 3.5% and Nocebo by 3.7 ± 1.8% However, there was no difference between LHTL and Placebo (F) ( 2.0, 6.2 to 2.3%) with only a 12% (P =.41, unlikely) likelihood of LHTL improving vvo 2 peak when compared with Placebo (F). Similarly, there was no difference between LHTL and Nocebo ( 0.3, 4.0 to 3.5%) with a 26% (P =.87, possibly not) likelihood of LHTL improving performance when compared with Nocebo. The difference between Placebo (F) and Nocebo was 1.7% ( 1.7 to 5.2%; likelihood 65%, P =.36, possibly). Discussion Unfortunately, we were unable to quantify the magnitude of the placebo effect, because our blinding failed. Nevertheless, the current study demonstrated that, in internationally-competitive race walkers, 21 d of LHTL simulated moderate altitude training for 14 h d 1 increased Hb mass and VO 2peak in addition to any improvements gained from the training camp in two control groups (Nocebo and Placebo (F) ). However, despite the increases in both Hb mass and VO 2peak, the simulated altitude group did not improve 10 min performance more than either of the control groups, which indicates that the performance improvement immediately after LHTL exposure was not greater than that gained from a well-designed training camp. Accelerated red blood cell production is effort independent and thus not affected but the failure of our placebo blinding. In accordance with recent work from our laboratory, increasing the duration of simulated altitude exposure was effective in eliciting increases in Hb mass. The 5 9% improvement in Hb mass in LHTL compared with Placebo (F) and Nocebo is in the same range as the recent studies from our laboratory using an extended duration LHTL protocol in highly trained runners, 24 the same LHTL protocol as the current study in highly trained cyclists and runners, 7,8 and of that obtained by other research groups using a similar altitude stimulus and duration. 1,2,9 In the current study, increased Hb mass occurred together with an increased total red cell volume, a reduction in plasma volume and no change in total blood volume in the LTHL athletes. In contrast, an expansion of plasma volume was observed in Nocebo and Placebo (F) during the first weeks of the study, likely due to the hot training conditions. 25 Therefore, assuming that SaO 2 and PaO 2 remain constant, relative oxygen transport capacity [Hb] was increased during the 3 wk of training and may have contributed to higher maximal aerobic

The Placebo Effect of Simulated Altitude Training 531 power in LHTL, assuming that peak cardiac output and arteriovenous O 2 difference were unchanged. In this context it may be of some importance that all subjects received iron supplementation throughout the intervention period which successfully prevented a state of low body iron stores. This occurred despite one LHTL subject being iron deficient during the first days of the study, which was improved after one week of iron supplementation. However, in a placebo-controlled trial by Friedmann et al, 21 12 wk of iron supplementation in iron-deficient athletes, defined by a blood ferritin of <20 ng gml 1, did not increase red cell volume significantly. Therefore, the rise in Hb mass observed in our LHTL subjects can likely be attributed to altitude exposure rather than the iron-supplementation. stfr is tightly correlated with erythropoiesis and the lack of an increase in stfr with LHTL is somewhat unexpected. The typical error of stfr was six-fold higher than that of Hb mass in the current study and this fact may be the cause of this discrepancy. However, when the one LHTL subject who was iron deficient during the first few days of the study is removed from the anlysis, the trend is for an increase in stfr at week 2 and 3 of the intervention period in the LHTL group. There is substantial evidence that improvements in endurance performance at sea level can be obtained through altitude exposure via nonhematological mechanisms, 13 although this is strongly contested by other researchers. 26 The current study did not indicate any improvement in race walking economy, measured by the slope of VO 2 versus speed during 4 5 submaximal walking bouts. In fact, race walking economy in Nocebo was better than LHTL after the 3 wk intervention period. Collectively, these results suggest that improvements in economy are not a consistent mechanism associated with altitude training. Although the physiological responses described above are affected by altitude exposure and may be associated with changes in performance, the main outcome of any altitude training regimen is to improve competitive performance. The current study was successful in improving 10 min treadmill performance by approx. 4%, which is a worthwhile change since the relative improvement in performance required by a top individual athlete to increase their chance of winning medals at international competition is 0.5 1%. 27 However, all three groups in the current study improved performance in the 10 min treadmill test equally and although only LHTL had substantial hematological changes (due to altitude exposure) this did not result in better performance than the Placebo (F) and Nocebo immediately after simulated altitude exposure. Despite the failure to blind the Placebo group, our results have two important implications: 1) that a well-designed training camp that brings top athletes together to train is conducive to gaining substantial improvements in performance; and 2) in order to maximize the physiological improvements gained from altitude training a period longer than 2 d between completion of altitude and competition may be required. If the extra stress of altitude together with hard training accumulates substantial fatigue, a period of a few days to several weeks may be required to dissipate the fatigue and allow the increase in fitness to be manifested. 28 Of note, 4 wk postintervention four athletes in LHTL walked an average of 6% faster than their previous best 20 km before commencement of the study in the Australian 20 km Olympic Selection trials and three of these athletes walked qualifying times and were selected for the Australian team. It is a major limitation that we were unable to attract higher number of subjects. However, high caliber athletes are difficult to recruit and retain for the duration of

532 Saunders et al. research studies. By using relevant statistics 29 we have moderate estimates of the effect of LHTL on both physiological and performance variables. Another serious limitation of the current study was our failure to blind the Placebo and LHTL groups. Despite our best efforts to blind the study, all athletes correctly guessed their group allocation with a high certainty. Our attempt to quantify the placebo effect failed, and consequently we ended up with two control groups and one LHTL group. Nevertheless, we confirmed that for our subject group of elite level race walkers, LHTL was no more beneficial for performance than the training camp, despite the difference in physiological adaptations of Hb mass and VO 2 peak, which should be independent of the placebo effect. This result reinforces that VO 2 peak is not the sole determinant of performance in highly trained athletes. 30,31 A final limitation is that training was not accurately monitored in the lead up period before the study. Although groups were matched for training history, there is the possibility that athletes did more or less training than normal during the study period that may have influenced the physiological and performance results. However, any effect of large variations in training should have been random between groups. In conclusion, 3 wk of simulated altitude exposure of 3000 m for 14 h d 1 substantially improved hematological parameters (Hb mass ) and VO 2 peak in elite race walkers when compared with two control groups (Placebo (F) and Nocebo groups). Although, our attempt at quantifying the placebo effect failed, treadmill performance within 5 d after the intervention was substantially better in all three groups, which was surprising given the higher physiological capacity of the simulated altitude group. Our attempts illustrate the extreme challenge of blinding elite athletes to moderate altitude, either terrestrial or simulated given that many of them may have had exposure previously. Studies with younger or less-experience athletes may be useful to quantify the placebo effect, but the relevance to truly elite athletes will be dubious. 29 Practical Applications The current study demonstrates that a well designed training camp with elite athletes can provide significant improvements in performance. The use of LHTL altitude exposure for 3 wk at 14 h/d provides improvement in physiological capacities such as greater Hb mass and VO 2 peak which should lead to improved performance with correct training during and after the altitude exposure. Optimal performance after LHTL altitude exposure may be some time after the completion of the exposure and may be related to an increase in training capabilities due to the increase in aerobic capacity from the altitude exposure. References 1. Brugniaux JV, Schmitt L, Robach P, et al. Eighteen days of living high, training low stimulate erythropoiesis and enhance aerobic performance in elite middle-distance runners. J Appl Physiol. 2006;100:203 211. 2. Levine BD, Stray-Gundersen J. Living high-training low : effect of moderatealtitude acclimatization with low-altitude training on performance. J Appl Physiol. 1997;83:102 112.

The Placebo Effect of Simulated Altitude Training 533 3. Saunders PU, Telford RD, Pyne DB, Gore CJ, Hahn AG. Improved race performance in elite middle-distance runners after cumulative altitude exposure. Int J Sports Physiol Perform. 2009;4:134 138. 4. Stray-Gundersen J, Chapman RF, Levine BD. Living high-training low altitude training improves sea level performance in male and female elite runners. J Appl Physiol. 2001;91:1113 1120. 5. Levine BD, Stray-Gundersen J. A practical approach to altitude training: where to live and train for optimal performance enhancement. Int J Sports Med. 1992;13(Suppl 1):S209 S212. 6. Julian CG, Gore CJ, Wilber RL, et al. Intermittent normobaric hypoxia does not alter performance or erythropoietic markers in highly trained distance runners. J Appl Physiol. 2004;96:1800 1807. 7. Clark SA, Quod MJ, Clark MA, Martin DT, Saunders PU, Gore CJ. Time course of haemoglobin mass during 21 days live high:train low simulated altitude. Eur J Appl Physiol. 2009;106:399 406. 8. Robertson EY, Saunders PU, Pyne DB, Aughey RJ, Anson JM, Gore CJ. Reproducibility of performance changes to simulated live high/train low altitude. Med Sci Sports Exerc. 2010;42:394 401. 9. Wehrlin JP, Zuest P, Hallen J, Marti B. Live High - Train Low For 24 Days Increases Hemoglobin Mass And Red Cell Volume In Elite Endurance Athletes. J Appl Physiol. 2006;100:1938 1945. 10. Dick FW. Training at altitude in practice. Int J Sports Med. 1992;13:S203 S206. 11. Beedie CJ, Stuart EM, Coleman DA, Foad AJ. Placebo effects of caffeine on cycling performance. Med Sci Sports Exerc. 2006;38:2159 2164. 12. Clark VR, Hopkins WG, Hawley JA, Burke LM. Placebo effect of carbohydrate feedings during a 40-km cycling time trial. Med Sci Sports Exerc. 2000;32:1642 1647. 13. Gore CJ, Clark SA, Saunders PU. Nonhematological mechanisms of improved sea-level performance after hypoxic exposure. Med Sci Sports Exerc. 2007;39:1600 1609. 14. Gore CJ, Rodriguez FA, Truijens MJ, Townsend NE, Stray-Gundersen J, Levine BD. Increased serum erythropoietin but not red cell production after 4 wk of intermittent hypobaric hypoxia (4000-5500m). J Appl Physiol. 2006;101:1386 1393. 15. Knaupp W, Khilnani S, Sherwood J, Scharf S, Steinberg H. Erythropoietin response to acute normobaric hypoxia in humans. J Appl Physiol. 1992;73:837 840. 16. Saunders PU, Telford RD, Pyne DB, et al. Improved running economy in elite runners after 20 days of simulated moderate-altitude exposure. J Appl Physiol. 2004;96:931 937. 17. Saunders PU, Pyne DB, Telford RD, Hawley JA. Reliability and Variability of Running Economy in Elite Distance Runners. Med Sci Sports Exerc. 2004;36:1972 1976. 18. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307 310. 19. Suominen P, Punnonen K, Rajamaki A, Irjala K. Serum transferrin receptor and transferrin receptor-ferritin index identify healthy subjects with subclinical iron deficits. Blood. 1998;92:2934 2939. 20. Schmidt W, Prommer N. The optimized CO-rebreathing method - a new tool to determine total haemoglobin mass routinely. Eur J Appl Physiol. 2005;95:486 495. 21. Friedmann B, Jost J, Rating T, et al. Effects of iron supplementation on total body hemoglobin during endurance training at moderate altitude. Int J Sports Med. 1999;20:78 85. 22. Liow DK, Hopkins WG. Velocity specificity of weight training for kayak sprint performance. Med Sci Sports Exerc. 2003;35:1232 1237. 23. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41:3 13.

534 Saunders et al. 24. Saunders PU, Telford RD, Pyne DB, Hahn AG, Gore CJ. Improved running economy and increased hemoglobin mass in elite runners after extended moderate altitude exposure. J Sci Med Sport. 2009;12:67 72. 25. Sawka MN, Convertino VA, Eichner ER, Schnieder SM, Young AJ. Blood volume: importance and adaptations to exercise training, environmental stresses, and trauma/ sickness. Med Sci Sports Exerc. 2000;32:332 348. 26. Lundby C, Calbet JA, Sander M, van Hall G, Mazzeo RS, Stray-Gundersen J, M. SJ, Chapman RF, Saltin B, Levine BD. Exercise economy does not change after acclimatization to moderate to very high altitude. Scand J Med Sci Sports. 2006;E-published April 3 27. Hopkins WG, Hewson DJ. Variability of competitive performance of distance runners. Med Sci Sports Exerc. 2001;33:1588 1592. 28. Morton RH, Fitz-Clarke JR, Banister EW. Modeling human performance in running. J Appl Physiol. 1990;69:1171 1177. 29. Bonetti DL, Hopkins WG. Sea-Level exercise performance following adaptation to hypoxia: A meta-analysis. Sports Med. 2009;39:107 127. 30. Bulbulian R, Wilcox AR, Darabos BL. Anaerobic contribution to distance running performance of trained cross-country athletes. Med Sci Sports Exerc. 1986;18:107 113. 31. di Prampero PE. The energy cost of human locomotion on land and in water. Int J Sports Med. 1986;7:55 72.