Effects of an Injected Placebo on Endurance Running Performance


Medicine & Science in Sports & Exercise: August 2015 - Volume 47 - Issue 8 - p 1672–1681
doi: 10.1249/MSS.0000000000000584
Applied Sciences

Purpose: This study aims to quantify the magnitude of the placebo effect of an injected placebo (“OxyRBX”) purporting to have effects similar to those of recombinant human erythropoietin on endurance running performance in “real-world” field-based head-to-head competition settings.

Methods: Fifteen endurance-trained club-level men (mean ± SD: age, 27.5 ± 6.8 yr; body mass index, 22.9 ± 2.0 kg·m−2) with 10-km personal-best record times of 39.3 ± 4.4 min completed a randomized cross-over study of 3-km races before and after 7-d “control” and “placebo” phases. During the placebo phase, participants self-administered subcutaneous saline injections daily, believing it to be OxyRBX, with no intervention during the control phase. At the start and end of each 7-d phase, 3-km running performance was assessed. Qualitative assessments of participants’ perceptions and experiences were recorded throughout and during semistructured interviews on completion.

Results: Race time improved significantly more in response to placebo intervention (9.73 ± 1.96 s faster, P = 0.0005) than in response to control (1.82 ± 1.94 s faster, P = 0.41; Pinteraction = 0.02). In response to placebo, participants reported reductions in physical effort, increased potential motivation, and improved recovery. Beliefs and congruence between positive expectations of the effects of placebo and perceptions of physical change during training also appeared to impact on competitive performance.

Conclusions: Compared to control, the injected placebo improved 3-km race time by 1.2%. This change is of clear sporting relevance but is smaller than the performance improvement elicited by recombinant human erythropoietin administration. Qualitative data suggest that placebo may have improved performance by both reducing perception of effort and increasing potential motivation, in accord with the psychobiological model for exercise performance, and that cognitive and noncognitive processes appear to have influenced placebo response.

1Institute of Cardiovascular and Medical Sciences, College of Medical, Veterinary, and Life Sciences, University of Glasgow, Glasgow, UNITED KINGDOM; and 2Institute of Health and Wellbeing, College of Social Sciences, University of Glasgow, Glasgow, UNITED KINGDOM

Address for correspondence: Jason M.R. Gill, Ph.D. BHF Glasgow Cardiovascular Research Centre, Institute of Cardiovascular and Medical Sciences, College of Medical, Veterinary, and Life Sciences, University of Glasgow, Glasgow G12 8TA, United Kingdom; E-mail: jason.gill@glasgow.ac.uk.

Submitted for publication June 2014.

Accepted for publication November 2014.

Article Outline

Placebo effect is acknowledged as a key factor in medical research; as a consequence, it has been controlled for in clinical trials for more than 50 yr (26). Placebo effect has also been recognized in the context of sports performance, with a number of studies reporting statistically significant improvements in endurance, sprint, or strength performance with placebo interventions (reviewed by Beedie and Foad (4)).

Orally administered placebos have been typically shown to improve endurance performance by ∼2% in participants who are at least moderately well-trained (5,11,17,22,39). However, these studies have all assessed performance using either cycle ergometer or running time trial performance assessment, with tests performed on individuals performing alone and unaware of the performances of others, rather than assessment during head-to-head field-based competition scenarios—in other words, studies are often not performed under conditions which best reflect the “real-world” sporting competition for that particular event. It is known that performance is often improved in head-to-head competition settings, compared to settings without a competitive element (12,13,34,37,38). The mechanisms responsible for this “competition effect” are not fully understood but could conceivably be explained in the context of the psychobiological model (8,20,29), which applies motivational intensity theory (40) to a sporting context. Motivational intensity theory predicts that maximal exercise tolerance increases when either perception of effort is reduced or “potential motivation” (i.e., the greatest effort an individual is prepared to exert) is increased (8,20,29). Consistent with increased potential motivation, evidence indicates that competition results in increased allocation of effort to an exercise task (12), increased positive emotions (12), a similar RPE at a greater workload (37), and reduced internal attentional focus (37), together with performance-facilitating physiological changes such as enhanced sympathoadrenal system activation (12,34), increased heart rate (12,37), and higher peak oxygen uptake (34), although performance improvements with competition have also been observed in the absence of changes in peak heart rate and oxygen uptake (38). Thus, it is likely that the ergogenic benefit of competition is the result of motivational or dissociative effects enabling a greater amount of “reserve” capacity between volitional maximal effort and the true physiological capability to be utilized (13,32,33). Placebo may also reduce perception of effort via its positive effect on perceived ability and thus act, at least in part, on the same psychological mediator as competition to improve performance. Thus, the effects of placebo and competition may not be additive, and the effects of placebo on performance in a head-to-head competitive environment may be less marked than previously observed. It is therefore important to study the magnitude of placebo effect in a competitive setting to provide a clearer measure of the likely magnitude of placebo effect on sporting performance in a real-world competition setting.

Increasing use of performance-enhancing drugs in both elite-level and recreational-level sports has been reported (10,30). One such drug, which the World Anti-Doping Code bans for use in sports, is recombinant human erythropoietin (r-HuEpo), which stimulates renal erythrocyte production (15). Studies have shown that r-HuEpo administration increases hematocrit (Hct), which in turn can lead to an increase in oxygen-carrying capacity of between 7% and 13% (14,18,24,28). Such impact often also results in improvement of endurance performance in athletes (2,14). However, studies on the effects of r-HuEpo administration on sporting performance in field-based tests are limited (14), and no study has assessed its effects on performance in a placebo-controlled trial using a field-based head-to-head competition setting. Such studies are needed to assess the true effects of r-HuEpo on “real-life” sporting performance.

A key feature of r-HuEpo administration is that it is given by injection. Clear evidence from clinical medicine indicates that the route of delivery is a key mediator of the size of a placebo effect, with placebos administered by injection inducing larger effects than placebos administered orally (43). For example, Benedetti et al. (6) effectively demonstrated that an intramuscular placebo injection improved pain tolerance to a greater extent than oral placebo administration. It is therefore conceivable that at least some of the performance benefits of r-HuEpo administration could be attributable to an additional conditioned response related to route of administration (36). However, Benedetti et al. (6) did not include any measure of physical performance and did not involve competitive athletes. In the context of sports performance, only one study, to our knowledge, has assessed the effect of an injected placebo (a sham spinal injection purporting to be fentanyl, an opioid analgesic) on endurance sporting performance (1). This study found no effect of placebo on 5-km cycling performance (1); however, there has been subsequent criticism of the efficacy of the placebo procedure employed, as it was argued that the leg pain experienced during the placebo cycling time trial (which would not have been experienced with fentanyl) may have unmasked the sham nature of the placebo intervention (19). Thus, to our knowledge, there have been no appropriately designed studies assessing the effect of an injected placebo, purporting to be a performance-enhancing ergogenic aid, on endurance performance.

The aim of this study was therefore to quantify the magnitude of the placebo effect of an injected placebo (which we called “OxyRBX”) purporting to have a similar effect as r-HuEpo on endurance running performance in a field-based head-to-head competition setting. In addition, to further the qualitative understanding of how placebo injections may influence endurance running performance, research team members recorded detailed notes of any comments made to them by participants about their perceptions and experiences of training and competition during the study; on completion of the study, participants were interviewed in-depth about their experiences of “taking OxyRBX.”

Back to Top | Article Outline


Back to Top | Article Outline


Nineteen endurance-trained male volunteers were initially recruited to take part in the study. Four participants dropped out of the study before the commencement of OxyRBX administration. Three of the four dropouts mentioned “fear” of the possible complications (e.g., blood clotting) associated with OxyRBX use. One individual did not give a reason for dropping out. Thus, 15 participants (mean ± SD: age, 27.5 ± 6.8 yr; height, 1.79 ± 0.05 m; body mass, 73.4 ± 7.6 kg; body mass index, 22.9 ± 2.0 kg·m−2) successfully completed the experimental protocol. The 15 participants were well-trained club-level athletes who reported engaging in 213 ± 129 min of endurance-based training and 50 ± 58 min of resistance training per week and who had personal-best record times of 39.3 ± 4.4 min for running 10 km. All participants were healthy nonsmokers, and none of them were taking any medications or supplements at the start of the study. Participants provided a written informed consent on the basis that they were undertaking a trial to investigate the effects of the legal erythropoietin-like substance OxyRBX on sporting performance, rather than a trial of placebo effect. This deception was essential for the study to be successfully undertaken and is standard practice in published studies of placebo effect (7). The dropout of three participants because of fear of the complications of OxyRBX administration illustrates the effectiveness of the procedures employed to induce participants’ belief that they were taking a powerful drug. Participants were fully debriefed about the true nature of the study, on completion of a poststudy qualitative interview (see later discussion; one participant who did not attend the interview was debriefed by E-mail), and were instructed not to discuss anything related to the trial with others until given permission by the research team (granted when all participants had completed the study). As a further precaution, participants were given specific instructions and directions on how to enter and vacate the final interview room to ensure no crossing of paths between participants. The study was approved by the research ethics committee of the College of Medical, Veterinary, and Life Sciences at the University of Glasgow and was performed according to the World Medical Association (Declaration of Helsinki) code of ethics.

Back to Top | Article Outline

Preliminary study brief

Before the commencement of performance trials, the effects of r-HuEpo administration on exercise performance were described to participants, and discussions concerning its alleged use, particularly among elite cyclists, took place. Each preliminary study brief was carried out by the same investigator (R.R.) on a one-on-one basis and followed a semistructured approach. Participants were initially provided with a detailed information sheet about the study (described as “A study to assess the effects of OxyRBX on sporting performance in well-trained individuals”) and a “drug information sheet” on OxyRBX describing its purported effects, dosage, and safety information. This was reinforced by a semistructured discussion of the purported benefits of OxyRBX on performance, with OxyRBX described as being a legal r-HuEpo-like substance that had been shown in animal studies to induce benefits similar to those of r-HuEpo and had been shown in extensive testing to be safe for use in humans. Potential side effects of OxyRBX (e.g., rash; hives; itching; difficulty breathing; tightness in the chest; swelling of the mouth, face, lips, or tongue) were also described, but it was emphasised that these were extremely rare (less than one case in a million). Participants were encouraged to ask questions and to discuss issues or topics of the study that were important to them to ensure an effective tailored priming process for each individual. This briefing process was carried out in order to reinforce beliefs on the effects of OxyRBX on exercise performance and the similarity of these effects to those of r-HuEpo.

Back to Top | Article Outline

Experimental design

The experimental design is outlined in Figure 1. Participants initially underwent an individual 3-km time trial familiarization run on an indoor 200-m running track located at the Kelvin Hall Arena (Glasgow, United Kingdom). This was carried out to familiarize participants with the running track used for the four competition runs and with racing over the 3-km distance. With participants unaware, individual 3-km completion times for this time trial were used to handicap for the 3-km competition runs in the main study. Participants then followed a randomized cross-over study design, where each participant underwent tests before and after a 7-d “control” phase, during which no intervention was given, and before and after a 7-d “placebo” phase, in which participants came to the laboratory every day to receive daily subcutaneous saline injections (0.5 mL of sterile 0.9% NaCl). Participants were informed that these injections contained OxyRBX, which should elicit effects similar to those of r-HuEpo. Eight participants underwent the control phase before the placebo phase, with the other seven participants undergoing the placebo phase before the control phase. There was a 2-wk interval between the two study phases; participants were told that this 2-wk interval ensured a suitable OxyRBX “washout period” for participants who started with the OxyRBX administration phase. Participants were regularly reminded not to discuss which group they have been assigned to and, as far as possible, were monitored to ensure that no such discussions took place.

Back to Top | Article Outline

3-km running races

Each participant in each cohort undertook four performance tests as competitive 3-km running races (in groups of seven or eight participants) on an indoor 200-m running track. During the races, which took place at the start and end of both placebo and control weeks, ambient temperature and humidity conditions were 17.9°C ± 0.8°C and 39.6% ± 3.4%, respectively. Participants were asked to prepare for each race as they would normally prepare for competition and were instructed to refrain from consuming alcohol 48 h preceding each race. Participants were given no specific instructions regarding hydration and caffeine intake other than being told to prepare as they would normally prepare for a competitive event. Participants were asked to undertake the same preparation for all the races. To ensure that the 3-km races were competitive, we handicapped participant starting times based on the times achieved during the individual 3-km time trial familiarization runs (i.e., a participant who completed that familiarization run in 11 min would start 30 s ahead of a participant who completed the familiarization run in 10 min 30 s; this handicapping remained constant throughout the study), and small monetary prizes (£35 [∼US$50] for first place, decreasing by £5 per position down to seventh place) were provided according to finishing position. Participants were instructed to aim to achieve as high a finishing position as possible and to complete the race in the shortest possible time. Heart rates were recorded second by second throughout using a telemetry system (Fitpulse; TT Sport SRL, San Marino, Italy), and lap times were recorded manually by the investigators. RPE were collected by a research team member from each participant immediately after the participant finished each race (including familiarization run). Participants were asked to report their RPE at the end of the race using the Borg scale from 6 to 20 (9) (6 = “no effort at all”; 20 = “maximal exertion”). Before each race, they were briefed that this rating should reflect their overall perception of effort, taking into account all sensations. During each race, participants were given positive verbal encouragement by research team members who were blinded to the participants’ condition allocation and received verbal information on the number of laps remaining. Participants were not given any further information (e.g., lap times, heart rates, etc.) or results until the completion of the study.

Back to Top | Article Outline

Hematological measures

Venous blood samples (∼4 mL) from an antecubital vein were collected into ethylenediaminetetraacetic acid tubes, after 10 min of supine rest, 60–120 min before each 3-km race. Red blood cell count, hemoglobin concentration, Hct, mean cell volume, and mean corpuscular volume were quantified using a hematology analyzer (XT-2000i; Sysmex, Norderstedt, Germany).

Back to Top | Article Outline

Training and diet

Participants were instructed to maintain their normal training and diet regimen throughout the study and were asked to record all training sessions in a diary provided by the experimenters. Participants were asked to replicate training for the week and food intake for the 2 d leading up to each 3-km race.

Back to Top | Article Outline

Qualitative assessments

During the placebo week, when they attended the university to receive their daily placebo injection, participants were asked to describe any changes that they had noticed while taking OxyRBX. Immediately after each 3-km race (during both placebo and control weeks), participants were asked to assess their performance during the race. Notes of daily and postrace accounts were taken by research team members and written electronically. All 15 participants provided accounts during the placebo week; 9 of 15 participants provided accounts during the control week. In addition, 14 of 15 participants took part in an in-depth semistructured interview on completion of the study, before being informed of the true nature of the experiment. The interviews were audiorecorded with participant consent and transcribed verbatim. Questions were asked in two stages. First, participants were asked about their experiences of taking OxyRBX during the trial, including whether they had any anxieties about taking the substance; whether they thought it would improve their performance; whether they felt different while “taking” the substance; how they felt during training and during the races; whether they felt that their recovery was different; and whether they experienced any positive or negative side effects. After their race times had been revealed to them, they were asked further questions, including the extent to which they felt that any improvement was caused by the substance allowing them to work harder. After completion of the interview, participants were fully debriefed about the nature of the study.

Back to Top | Article Outline

Quantitative data analysis

Statistical analyses were conducted using Statistica 6.0 (StatSoft, Tulsa, OK). Data are presented as mean ± SEM (unless otherwise indicated). All data were tested for normality and homogeneity of variance before statistical analysis and were found to conform to assumptions for parametric statistical testing. Two-way repeated-measures ANOVA [intervention condition (control vs placebo) × test (pre vs post)] were used to identify any between-trial differences in 3-km race time, mean heart rate, RPE, and hematological variables. The intervention condition–test interaction term was used to assess whether preintervention-to-postintervention changes under the placebo condition differed from those under the control condition. Three-way repeated-measures ANOVA [intervention condition × test × race segment (first, second, and third 1000-m segments)] was used to identify any between-trial differences in running speed and heart rate throughout the races. For all ANOVA, post hoc Fisher test was used to identify where any differences lay. Statistical significance was accepted at P < 0.05.

Back to Top | Article Outline

Qualitative data analysis

Notes of participants’ accounts and interview transcripts were read repeatedly (by R. R. and C. M. G.) to identify questions of interest (e.g., physical effects of taking OxyRBX) and any emerging issues (e.g., impact of regular competition on performance). Thematic analysis was conducted using an adapted framework approach (27), where data are coded, indexed, charted systematically, and organized using matrices in which each participant is represented by a row and each theme is represented by a column. NVivo10 software was used to assist data coding and organization. Constant comparison (checking the emerging analysis against every instance of similarly indexed data) ensured that all perspectives were represented (31).

Summary analyses of four key themes are relevant here: “Expectations,” which captured participants’ views of taking OxyRBX before the placebo week; “Physical effects,” which included all references to physical changes experienced by participants during the study (both placebo and control weeks); “Psychological effects,” which included all references to nonphysical changes experienced by participants during the study; and “Competition,” which included participants’ views of the competitive element of the study. To compare and contrast the accounts of participants whose performance improved during the placebo week and participants who showed no improvement, we grouped participants according to change in race times during the placebo week. Participants were assigned to one of three “performance” groups: “Marked Improvement” group (n = 6), which included those whose races times decreased by ≥10 s; “Slight Improvement” group (n = 5), which included those whose race times decreased by <10 s; and “No Improvement” group (n = 4), which included those whose race times increased during the placebo week compared to the control week. Participant numbers were assigned on recruitment to the study and were not reassigned when a participant dropped out; thus, participant numbers for interviews in the results are recorded as P01 to P19. Accordingly, extracts are presented with labels to indicate the participant’s ID (P01–P19), data type (“Log” = researcher notes; “Int” = poststudy interview), and performance during the placebo week (“Improver” = marked improvement; “Slight” = slight improvement; “No” = no improvement).

Back to Top | Article Outline


Back to Top | Article Outline

Quantitative results

Completion times for the 3-km races preintervention and postintervention under the control and placebo conditions are shown in Table 1. There was a significantly greater improvement in race time from preintervention to postintervention in response to the placebo intervention than in response to the control intervention (intervention condition–test interaction; F1,14 = 6.82, P = 0.02). Post hoc tests revealed that race time was significantly faster by 9.73 s [95% confidence interval (CI), 5.14–14.33 s faster] in response to the placebo intervention (P = 0.0005) but did not improve significantly in response to the control condition (1.82 s faster; 95% CI, 2.77 s slower–6.41 s faster; P = 0.41). Figure 2A shows the mean ± SEM change in 3-km race time in response to the control and placebo interventions (i.e., postintervention − preintervention). Figure 2B shows changes in individual participants. Eleven participants improved performance more in response to the placebo intervention than in response to the control intervention; change in performance was similar in response to both interventions for one participant; and three participants had greater performance increases in response to the control intervention.

Figures 3A and B show the participants’ running speeds in the first, second, and third 1000-m segments of the 3-km races before and after the control and placebo interventions, respectively. In three-way repeated-measures ANOVA (intervention condition × test × race segment), a significant intervention condition–test interaction for running speed was observed (F1,14 = 7.117, P = 0.018). Intervention condition × race segment (F2,28 = 0.085, P = 0.92), test × race segment (F2,28 = 0.645, P = 0.53), and intervention condition × test × race segment (F2,28 = 0.378, P = 0.69) interaction terms were not significant. Post hoc tests revealed that participants started the 3-km race more aggressively after the placebo intervention, running the first 1000-m segment at a 2% faster pace than preintervention (0.094 m·s−1 faster; 95% CI, 0.032–0.156 m·s−1 faster; P = 0.004) and maintaining a running speed 1.4% faster than preintervention in the second 1000-m segment (0.061 m·s−1 faster; 95% CI, 0.001 m·s−1 slower–0.123 m·s−1 faster; P = 0.053) and in the third 1000-m segment (0.062 m·s−1 faster; 95% CI, 0.000–0.124 m·s−1 faster; P = 0.050; Fig. 3B). In contrast, there was no significant difference in running speed in any of the three 1000-m segments of the postintervention 3-km race compared to preintervention under the control condition (Fig. 3A).

Figures 4A and B show participants’ heart rates during the first, second, and third 1000-m segments of the 3-km races before and after the control and placebo interventions, respectively, for participants in whom complete sets of heart rate data across all four races were obtained (n = 10). In three-way repeated-measures ANOVA (intervention condition × test × race segment), there was a significant main effect of race segment on heart rate (F2,18 = 51.058, P < 0.0005), but there were no significant two-way or three-way interactions indicating that heart rate responses during the 3-km races were not significantly influenced by the placebo intervention. There were also no significant differences in RPE between preintervention and postintervention races under either the control condition or the placebo condition (Table 1). As expected, there were no significant differences in any hematological variables between control and placebo conditions, either preintervention or postintervention (Table 1). Participants’ training remained constant throughout the period of the study and did not differ significantly between the weeks preceding each of the four 3-km races (Table 2).

Back to Top | Article Outline

Qualitative results

In poststudy interviews, a few participants—despite consenting to take part in the study—admitted that they were “worried about taking the injections” (P07_Int_Improver); “a wee (a little) bit guilty I suppose, putting something in my body that’s not always going to be there” (P16_Int_Improver); or “a wee bit scared, just because I didn’t know anything about OxyRBX” (P09_Int_No Improver). Almost all of the participants reported that they thought they were taking a real performance-enhancing substance during the study, providing further evidence that the briefing procedures used were highly effective in inducing participants’ beliefs that the substance would enhance their performance. Almost all of those who recorded a marked improvement or no improvement in their race times had expected to see positive changes after taking OxyRBX. Many said that they had been interested to see whether and how OxyRBX would work for them. Some participants whose performance had improved markedly described a real sense of anticipation beforehand:

P13_Int_Improver: “I wanted to kind of like count down until I was going to take it, and see if there was any differences in my performance and in like everyday general life. So yeah, I was really looking forward to seeing differences... the advantages of it.”

Among those whose performance improved less markedly, expectations tended to be more measured. Some admitted that they had not anticipated any changes from taking OxyRBX:

P15_Int_Slight: “I didn’t think taking the drug would have any effects at all because it was a tenth of the normal dose tested before. Ethically, I thought you couldn’t give any therapeutic doses, so didn’t think... I wouldn’t expect anything to happen, not through any evidence, just thought the dose was too small in comparison to therapeutic doses.”

Although a few participants—in describing their experiences of taking OxyRBX—reported minor negative effects (changing sleep patterns, slight pain at the injection site, and nausea), almost all mentioned at least some positive effects. During the placebo week, participants whose race times improved (both markedly and slightly) reported back to research team members that they were feeling “more comfortable” (P13_Improver) both during training and in competition:

P15_Log_Slight: Run today outdoor about 3 mile easy run in approximately 30 min, last 10 min more difficult, however much better than normal. This was a repeated route. A lot easier, at beginning took him longer to get tired [...] Breathing was easier during running, noticeably different.

P16_Log_Improver: Thinks best run was today. Doesn’t know why, says the only thing that’s different is the drug, so drug potentially had effect. Breathing was better, utilizing oxygen really well. Muscles more efficient, good push off every step.

This reduction in perceived physical effort both during training and in competitive races was also evident in poststudy interviews, where some participants described how taking OxyRBX had increased their enjoyment of training sessions:

P01_Int_Slight: “... when I started taking the drug, particularly noticed in the gym that I was doing a lot more in my sets than normal, and also running I did feel less tired even when I was on the treadmill... I don’t like running inside, but I felt I could run longer than normal. Even in training sessions, I felt I was running better, felt less out of breath, enjoying the sessions more. One session, it was windy, but I was still running well, coach said I was running faster.”

In contrast, participants whose race times did not improve often appeared less convinced that taking OxyRBX had had a positive physical effect, particularly during training (P19_Log_No: “Alright, nothing to report really. Feels no difference really”). One (who had expected the placebo to work) even suggested an increase in perceived effort (or a nocebo effect) during the placebo week:

P10_Int_No: “I thought the week when I was taking the supplement, I felt like my legs were really heavy, felt like it was working against me, particularly the drug week. The following weeks I was getting better, maybe like a delayed effect... like my legs felt better.”

Another noticeable physical effect that many participants experienced during the placebo week (including those whose race times showed no improvement) was enhanced recovery both after training sessions and after the races:

P15_Int_Slight: “... the last race we did, recovery was pretty poor... that was post control, compared to taking the drug; you seemed to regenerate really quickly. After 1 or 2 days on drug I started thinking the drug was having an effect.”

Many participants also described enhanced potential motivation, particularly during the competitive races. They commonly reported “pushing” themselves harder during the placebo week races, and some felt that taking OxyRBX presented an opportunity for them to experience their full potential as an endurance athlete:

P07_Int_Improver: “I don’t know, I always had it in my head, when I run, psychology is what stops me, but I think having this in me just made me push harder, I was knackered (tired) throughout that race, but I just kept pushing. It didn’t make the race easier to get that time, I don’t know, worked hard...”

Only a couple of participants reported that their potential motivation had increased during training. However, one reported that he had almost found this counterproductive:

P02_Int_Slight: “I think the training effect of the substance, everything was much quicker, I have to admit when I was on the substance I was on the verge of injury, I kept pushing myself too hard, just because I could... and because it was fun.”

In addition to increased potential motivation, some participants reported feeling greater confidence in their ability during the placebo week race. However, this perceived advantage did not always translate into improved time:

P09_Int_No Improver: “During the races, I always gave it all, so you’re finishing the same way. Just when I was doing the runs... when on the drug I thought, I thought there are at least two or three guys that I should have an advantage over.”

A number of participants also described how the competitive element of the study had had an effect in itself. As is revealed in this participant’s account, the regular races appeared to act as an additional spur for some during training:

P04_Log_Slight: “Feeling really good. Drug is having an effect. Feels more up for it, especially due to competition tomorrow.”

Indeed, one participant who had taken part in a previous performance-enhancing drug trial and remained sceptical throughout the study (“...having done the EPO [erythropoietin] trial and having a medic on board... this time not having one, seemed a bit unreal”) showed a marked improvement in his race time despite reporting no physical or psychological effects during the placebo week. He maintained that competition and regular assessment were enough to improve performance:

P5_Int_Improver: “The biggest change in training was the knowledge that you’re competing every week etc... You try and do race prep... Interval sessions improved noticeably one day. When you get onto one of these trials [research studies], you tend to find you can push harder, there’s something to focus on, prepare for.”

Back to Top | Article Outline


The aim of this study was to determine whether an injectable placebo, claiming to be a legal substance with effects similar to those of r-HuEpo, would improve endurance running performance. The principal finding was that participants completed the 3-km distance 1.2% faster under the postplacebo condition compared to the postcontrol condition (and 1.5% faster compared to the preplacebo condition)—a difference that is statistically significant, physiologically relevant, and clearly important in a competitive sporting setting. To put these results into context, in the 2012 Olympics, the difference between the gold medal and the fourth place was less than 1% in all track events from 1500 to 10,000 m for both men and women.

The ergogenic effects of placebo in the present study can potentially be explained within the framework of the psychobiological model, which postulates that maximal exercise tolerance increases when either perception of effort is reduced or potential motivation (i.e., the greatest effort an individual is prepared to exert) is increased (8,20,29). This model attempts to provide an overarching paradigm that encompasses both physiological and psychological factors affecting performance. For example, according to the psychobiological model, physiological factors such as exercise training (16) or carbohydrate ingestion (3) would ultimately act to improve performance by reducing perception of effort for a given absolute exercise intensity. Similarly, psychological interventions such as motivational “self-talk” lead to a lower RPE for equivalent work rate and enhanced endurance performance (8). Consistent with these previous observations, improved exercise performance for the same degree of perceived effort was observed after placebo administration in the present study. This suggests that placebo facilitated an enhancement in performance by providing a degree of decoupling in the normal relationship between RPE and exercise intensity, an observation supported by qualitative data in which a number of participants described reduced physical effort (and enhanced recovery) when taking the placebo both during training and in competitive races.

It was not the intention of this study to establish the potential underlying neurological mechanisms for this reduction in effort perception; however, previous work on the effects of placebo analgesia on pain perception, using functional magnetic resonance imaging, may provide some insights on this. Wager et al. (35) used functional magnetic resonance imaging to reveal decreased activity in pain-sensitive areas of the brain in response to a placebo analgesic cream, demonstrating a measurable effect of placebo on brain function. Thus, given the evidence that pharmaceutical interventions acting to elevate pain threshold also reduce RPE and improve exercise performance (21,33), it is conceivable that the effect of placebo on effort perception may act, at least in part, via the same neurological mechanisms (i.e., reduction in pain sensitivity). However, further study is needed to establish whether this is the case.

In addition to reducing perceived effort, evidence from qualitative data suggests that placebo administration increased potential motivation. Participants commonly reported pushing themselves harder during the races at the end of the placebo intervention week. This may be related to the effect of placebo on perceived ability (a number of participants also reported increased confidence in their ability after taking placebo), which has been shown to increase willingness to exert effort in challenging tasks (41,42). Thus, evidence from the present study suggests that placebo administration potentially acts to improve performance by both reducing perception of effort and increasing potential motivation.

The 1.2% improvement in performance with the injected placebo is of similar magnitude to the performance improvements seen in response to orally administered placebos (5,11,17,22,39). One interpretation of this finding is that route of placebo administration does not substantially influence the magnitude of placebo effect on endurance sporting performance. However, the lack of head-to-head competition in laboratory-based performance assessments after oral placebo administration in earlier studies may have led to the observed placebo effect being greater than would have been observed under true competitive conditions. The present findings suggest that placebo administration is likely to improve performance by reducing perception of effort and by increasing potential motivation—the same psychological mediators thought to be responsible for the ergogenic effect of competition (12,13,32,33,37). Thus, as capacity to reduce effort perception and to increase potential motivation is likely to be finite, the beneficial effects of competition and placebo are unlikely to be independent or fully additive. Thus, the present findings uniquely provide an estimate of the likely magnitude of the effect of an injected placebo on endurance sporting performance in a real-world competition setting. Further study is needed to determine whether there are differences in the effects of orally administered placebos and injected placebos on performance under such real-world conditions.

Analysis of qualitative data provided insights that helped explain how placebo influenced participants’ approach to and perceptions of training and racing, revealing that endurance athletes taking a placebo drug experienced reductions in perceptions of physical effort during training and competition, an increase in potential motivation during competition, and perceptions of increased recovery both during training and after races. According to the conditioned response model (36), the method of administration (injection; the conditioned stimulus) and the credibility of the university setting may have contributed to performance improvement; however, interestingly, the participants’ accounts suggest that an interaction between their expectations from taking a placebo and their actual experiences of physical changes during training influenced performance outcomes. Those who anticipated the greatest positive change from taking the placebo and also perceived decreased physical effort during training tended to demonstrate the greatest improvement in performance; those with more moderate expectations who perceived decreased physical effort during training showed some improvement in performance; whereas those whose expectations of positive change were not supported by their experiences (i.e., they did not perceive decreased physical effort during training) did not show any improvement in their race times. These findings are in line with experimental evidence indicating that placebo effect is mediated by cognitive processes when conscious physiological processes (e.g., motor performance or pain) are targeted (7,23). However, it has previously been suggested that a nocebo response (i.e., wherein an inert substance produces a “harmful” effect) is the result of negative expectations (7). Our study suggests that the relationship is not so straightforward and that, where experience thwarts expectations, a negative outcome may be observed. Previous research on placebo analgesia has demonstrated that verbal instructions on the effect of a placebo can also influence placebo response, with participants who received the strongest suggestion that a placebo will be effective showing the greatest pain reduction (25). In the current study, the initial detailed documentation and verbal reinforcement of participants’ belief that the placebo could produce effects similar to those of the well-known (illegal) performance enhancer r-HuEpo served to heighten emotions and expectations of the outcomes of taking the placebo. This in turn will likely have been an important mediator of the physical, motivational, and performance outcomes (26).

Only one study, to our knowledge, has determined the effect of r-HuEpo administration on running performance in a field-based test (14). In that report by Durussel et al. (14), r-HuEpo administration improved performance in a 3-km running time trial by ∼6%, compared to baseline, in a group of men of similar ability to participants in the present study (3-km race time of 668 ± 75 s at baseline). However, the study by Durussel et al. (14) did not include a placebo intervention arm; therefore, performance improvement from baseline would have reflected the true ergogenic effect of the drug plus any additional cognitive (i.e., placebo) effects associated with expectations and beliefs related to the impact of r-HuEpo administration. The results of the present study indicate that this cognitive component is real. In addition, the performance trial employed in the study by Durussel et al. (14) did not include head-to-head competition, thus representing an experimental model in which potential placebo effects are likely to be maximized. Thus, taking together the results of the present study and those of the study by Durussel et al. (14), it would seem that the likely true effect of r-HuEpo administration on endurance performance is somewhat less than 5%; further study using a placebo-controlled trial and performance testing in a competitive environment is needed to quantify the true ergogenic effect of r-HuEpo on endurance performance in real-life sporting context.

The present study has a number of strengths. We used a randomized cross-over design to determine the true effect of placebo over and above any familiarization or order of testing effects. We also used a field-based performance test that included head-to-head competition and prizes to simulate real competition. Thus, the results should provide an estimate of the true magnitude of changes in performance with an injected placebo in a competitive setting. In addition, we adopted a mixed-methods approach that supplemented quantitative data with qualitative insights into the participants’ experiences after placebo administration. However, an important limitation of the study is that, although the participants were well-trained, they were not elite runners. Thus, further study is needed to determine whether placebo effect size would be comparable in elite athletes, who may have greater experience in providing maximal physical effects and thus may have higher baseline potential motivation with less capacity for this to be augmented.

In conclusion, this study provides novel insights into the effects of an injected placebo, purporting to be a performance-enhancing agent similar to r-HuEpo, on endurance performance in a head-to-head competitive setting. The magnitude of benefit (at 1.2%) is of clear sporting relevance but is substantially smaller than the performance improvement elicited by r-HuEpo administration. The data are consistent with placebo acting to improve competitive performance by both reducing perception of effort and increasing potential motivation, in accord with the psychobiological model for exercise performance (8,20,29), but also suggest that other factors (including cognitive beliefs and expectations) may mediate placebo response. Further study is needed to determine whether the magnitude of placebo effect is similar in athletes at the elite level.

We thank the participants for their time and effort. Their cooperation is greatly appreciated.

This work received no external funding.

The authors declare no conflicts of interest.

The results of the present investigation do not in any way constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Amann M, Proctor LT, Sebranek JJ, Pegelow DF, Dempsey JA. Opioid-mediated muscle afferents inhibit central motor drive and limit peripheral muscle fatigue development in humans. J Physiol. 2009; 587: 271–83.
2. Ashenden MJ, Hahn AG, Martin DT, Logan P, Parisotto R, Gore CJ. A comparison of the physiological response to simulated altitude exposure and r-HuEpo administration. J Sports Sci. 2001; 19: 831–37.
3. Backhouse SH, Bishop NC, Biddle SJ, Williams C. Effect of carbohydrate and prolonged exercise on affect and perceived exertion. Med Sci Sports Exerc. 2005; 37 (10): 1768–73.
4. Beedie CJ, Foad AJ. The placebo effect in sports performance: a brief review. Sports Med. 2009; 39: 313–29.
5. Beedie CJ, Stuart EM, Coleman DA, Foad AJ. Placebo effects of caffeine on cycling performance. Med Sci Sports Exerc. 2006; 38 (12): 2159–64.
6. Benedetti F, Pollo A, Colloca L. Opioid-mediated placebo responses boost pain endurance and physical performance: is it doping in sport competitions? J Neurosci. 2007; 27: 11934–39.
7. Benedetti F, Pollo A, Lopiano L, Lanotte M, Vighetti S, Rainero I. Conscious expectation and unconscious conditioning in analgesic, motor, and hormonal placebo/nocebo responses. J Neurosci. 2003; 23: 4315–23.
8. Blanchfield AW, Hardy J, de Morree HM, Staiano W, Marcora SM. Talking yourself out of exhaustion: the effects of self-talk on endurance performance. Med Sci Sports Exerc. 2014; 46 (5): 998–1007.
9. Borg GA. Perceived exertion: a note on history and methods. Med Sci Sports. 1973; 5: 90–3.
10. Carpenter PC. Performance-enhancing drugs in sport. Endocrinol Metab Clin North Am. 2007; 36: 481–95.
11. 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 (9): 1642–47.
12. Cooke A, Kavussanu M, McIntyre D, Ring C. Effects of competition on endurance performance and the underlying psychological and physiological mechanisms. Biol Psychol. 2011; 86: 370–378.
13. Corbett J, Barwood MJ, Ouzounoglou A, Thelwell R, Dicks M. Influence of competition on performance and pacing during cycling exercise. Med Sci Sports Exerc. 2012; 44 (3): 509–15.
14. Durussel J, Daskalaki E, Anderson M, et al. Haemoglobin mass and running time trial performance after recombinant human erythropoietin administration in trained men. PLoS One. 2013; 8: e56151.
15. Egrie JC, Strickland TW, Lane J, et al. Characterization and biological effects of recombinant human erythropoietin. Immunobiology. 1986; 172: 213–24.
16. Ekblom B, Goldbarg AN. The influence of physical training and other factors on the subjective rating of perceived exertion. Acta Physiol Scand. 1971; 83: 399–406.
17. Foad AJ, Beedie CJ, Coleman DA. Pharmacological and psychological effects of caffeine ingestion in 40-km cycling performance. Med Sci Sports Exerc. 2008; 40 (1): 158–65.
18. Lundby C, Robach P, Boushel R, et al. Does recombinant human Epo increase exercise capacity by means other than augmenting oxygen transport? J Appl Physiol (1985 ). 2008; 105: 581–87.
19. Marcora SM. Rebuttal from Marcora. J Appl Physiol. 2010; 108: 457.
20. Marcora SM, Staiano W. The limit to exercise tolerance in humans: mind over muscle? Eur J Appl. Physiol. 2010; 109: 763–70.
21. Mauger AR, Jones AM, Williams CA. Influence of acetaminophen on performance during time trial cycling. J Appl Physiol (1985). 2010; 108: 98–104.
22. McClung M, Collins D. “Because I know it will!”: placebo effects of an ergogenic aid on athletic performance. J Sport Exerc Psychol. 2007; 29: 382–94.
23. Montgomery GH, Kirsch I. Classical conditioning and the placebo effect. Pain. 1997; 72: 107–13.
24. Parisotto R, Gore CJ, Emslie KR, et al. A novel method utilising markers of altered erythropoiesis for the detection of recombinant human erythropoietin abuse in athletes. Haematologica. 2000; 85: 564–72.
25. Pollo A, Amanzio M, Arslanian A, Casadio C, Maggi G, Benedetti F. Response expectancies in placebo analgesia and their clinical relevance. Pain. 2001; 93: 77–84.
26. Price DD, Finniss DG, Benedetti F. A comprehensive review of the placebo effect: recent advances and current thought. Annu Rev Psychol. 2008; 59: 565–90.
27. Richie J, Lewis J. Qualitative Research Practice. London: Sage; 2003.
28. Robach P, Calbet JA, Thomsen JJ, et al. The ergogenic effect of recombinant human erythropoietin on V˙O2max depends on the severity of arterial hypoxemia. PLoS One. 2008; 3: e2996.
29. Smirmaul BPC, Dantas JL, Nakamura FY, Pereira G. The psychobiological model: a new explanation to intensity regulation and (in)tolerance in endurance exercise. Rev Bras Educ Fis Esporte. 2014; 27: 333–40.
30. Strano RS, Botre F. Prevalence of illicit drug use among the Italian athlete population with special attention on drugs of abuse: a 10-year review. J Sports Sci. 2011; 29: 471–76.
31. Strauss A, Corbin J. Basics of Qualitative Research: Grounded Theory Procedures. London: Sage; 1990.
32. Swart J, Lamberts RP, Lambert MI, et al. Exercising with reserve: exercise regulation by perceived exertion in relation to duration of exercise and knowledge of endpoint. Br J Sports Med. 2009; 43: 775–81.
33. Swart J, Lamberts RP, Lambert MI, et al. Exercising with reserve: evidence that the central nervous system regulates prolonged exercise performance. Br J Sports Med. 2009; 43: 782–88.
34. Viru M, Hackney AC, Karelson K, Janson T, Kuus M, Viru A. Competition effects on physiological responses to exercise: performance, cardiorespiratory and hormonal factors. Acta Physiol Hung. 2010; 97: 22–30.
35. Wager TD, Rilling JK, Smith EE, et al. Placebo-induced changes in FMRI in the anticipation and experience of pain. Science. 2004; 303: 1162–67.
36. Wickramasekera I. A conditioned response model of the placebo effect predictions from the model. Biofeedback Self Regul. 1980; 5: 5–18.
37. Williams EL, Jones HS, Andy SS, Marchant DC, Midgley AW, Mc Naughton LR. Competitor presence reduces internal attentional focus and improves 16.1 km cycling time trial performance. J Sci Med Sport. 2015; 18 (4): 486–91.
38. Wilmore JH. Influence of motivation on physical work capacity and performance. J Appl Physiol. 1968; 24: 459–63.
39. Wright G, Porcari JP, Foster CC, et al. Placebo effects on exercise performance. Gundersen Lutheran Med J. 2009; 6: 3–7.
40. Wright RA. Refining the prediction of effort: Brehm’s distinction between potential motivation and motivation intensity. Soc Pers Psychol Compass. 2008; 2: 682–701.
41. Wright RA, Dill JC. Blood pressure responses and incentive appraisals as a function of perceived ability and objective task demand. Psychophysiology. 1993; 30: 152–60.
42. Wright RA, Dismukes A. Cardiovascular effects of experimentally induced efficacy (ability) appraisals at low and high levels of avoidant task demand. Psychophysiology. 1995; 32: 172–76.
43. Zhang W, Robertson J, Jones AC, Dieppe PA, Doherty M. The placebo effect and its determinants in osteoarthritis: meta-analysis of randomised controlled trials. Ann Rheum Dis. 2008; 67: 1716–23.


© 2015 American College of Sports Medicine