Motorsport is one of the largest spectator sports in the world with similar viewing audiences to football (soccer) (1), yet unlike football, there is limited scientific peer reviewed literature on the physiological responses of driver athletes (2–4). Like athletes in other sports, drivers compete at 65%–85% of their maximum heart rate (HR) because of the metabolic demand of skeletal muscle required to perform sport specific skills (pilot the vehicle) (5), emotional arousal (6), and elevated core body temperature (2,7).
Race car drivers wear fire protective suits, shoes, socks, gloves, balaclava, and a helmet during competition, with a clo value (indicator of the thermal properties of clothing) of 1.56 (7), which is similar to a winter ski suit. As such, the insulated race suit limits race car drivers’ ability to effectively cool themselves via sweating, which is alarming as it is not uncommon for the cockpit of a race car to have a temperature range of 32°C–58°C depending on the ambient temperature at the race track, placement of the race car’s drive train, and airflow through the cockpit (7).
Endurance sports car racing hosts some of the longest races in motorsports (up to 24 h) with multiple drivers sharing a single car. As race car drivers compete for multiple long duration stints over a 24-h period, it is important to monitor drivers during competition to ensure overall health. In 2017, the Federation Internationale de l’Automobile (FIA) enacted regulations limiting the temperature of the cockpit to reduce the risk of heat injury. The regulation (901-1) is detailed in Table 1.
TABLE 1 -
FIA regulation 901-1.
|A: When the car is in motion, the temperature around the driver must be maintained at:
| 1. 32°C maximum when the external temperature is below 32°C
| 2. A figure less than or equal to the external temperature if the latter is above 32°C
|B: The temperature must go back down to the figure defined above (case 1 or 2) in 8 min maximum after a car stop.
|C: There is a limitation of the continuous driving time to 80 min if the ambient temperatures reach 32°C. The minimum rest time in the period of application of this rule must be 30 min between consecutive stints. The official maximum ambient temperature predicted, and the periods of application will be published 2 h before the start of the race.
The incident rates of “hot cockpit temperature” violations are private and only communicated over secure channels between the FIA and the racing teams. Yet, according to popular and social media, in 2017 (following the issue of the regulation), there were at least four instances of the FIA notifying teams that the cockpit temperature exceeded the designated threshold. The conservative incidence rates reported by the media indicate that the race car cockpits approach unsafe limits, but existing literature on fire fighters, astronauts, military personal, and hazardous material workers demonstrates that insulated clothing creates a microenvironment between the individual and the ambient environment. The physical properties of such clothing do not allow for an effective determination of core body temperature by evaluating the ambient environment (8–10). As the required motorsport fire suit creates a microenvironment around the race car driver (1,4,7,11), it is hypothesized that cockpit temperature is not an accurate indicator of thermal strain on the race car driver.
Thus, the purpose of this investigation was to determine the correlation between cockpit temperature and indicators of thermal strain of race car drivers during an endurance sports car competition when ambient temperature exceeded 32°C. The outcome of the present investigation will inform evidence-based practices to improve driver health and safety in all forms of motorsport competition.
This study was approved by the Institutional Review Board at Michigan State University. Before the start of testing, the protocols were explained, and the participants provided an up-to-date health history and signed an institutionally approved informed consent. Participants also provided consent for their race team to be acknowledged in a scientific publication, which could potentially lead to their identification. All procedures performed were in accordance with the ethical standards established by the 1964 Declaration of Helsinki and its later amendments.
Study participants were four male sports car drivers with at least 15 yr of professional racing experience. All participants were members of the same racing team and drove identical Corvette C7.R GTLM class racing cars (two drivers drove car 1 and two drivers drove car 2) in the International Motor Sports Association (IMSA) WeatherTech SportsCar Championship, governed by the FIA Technical Regulations.
Race venue data collection
All data were collected during the 2018 “Sahlen’s 6 Hours at the Glen” race, located in Watkins Glen, New York, on July 1. One lap of the race course is 5.47 km in length and includes 11 turns. The race began at 0945 h and ended at 1545 h. The race was not a sanctioned FIA event, but participating teams were encouraged to follow temperature regulations.
Cockpit temperature measures
Cockpit temperatures were constantly monitored through a standard PT1000 thermal sensor placed in the middle of the car at the same height as the driver’s head. This sensor is required to be shielded from any direct airflow to maintain an accurate measure. The sensor recorded at a rate of 100 Hz.
All participants underwent fitness tests at the Spartan Motorsports Performance Laboratory at Michigan State University before track measurements. Body composition was determined using the BodPod (Body Composition System; Life Measurement Instruments, Concord, CA). After the pretesting, calibration was completed, and the subject put on the appropriate clothing (a swim cap, worn to minimize air trapped within the hair, and spandex shorts). The subject entered the BodPod for two trials of approximately 45 s. The Bodpod software and the Siri equation were used to calculate body fat percentage (12).
A maximal aerobic capacity test (V˙O2peak) was performed according to testing criteria established by the American College of Sports Medicine (13). A validated discontinuous treadmill protocol was used (14), which consisted of 3-min exercise stages of increasing intensity followed by 1.5-min recovery stages. Oxygen consumption was measured continuously using a metabolic cart (Parvo Medics, Salt Lake City, UT) connected to a Hans Rudolf face mask (worn by the participant). HR was recorded every 15 s using a Polar T31 Transmitter (Polar Electro, Kempele, Finland).
Muscular fatigue resistance and power output were assessed through a standard Wingate power test using a cycle ergometer with a mechanically loaded flywheel (Velotron, Seattle, WA). The participant was fitted to the cycle ergometer, so that the pedals and seat were in proper positions (12). The participant then warmed up by pedaling at a light resistance for 1 min. The ergometer flywheel was loaded with a resistance of 7.5% of the subject’s body weight. The participant was instructed to pedal at a rate above 80 rpm for the 30-s test period. Maximum and minimum power were measured and rate of fatigue [(maximum power − minimum power) × 30 s−1] and glycolytic capacity [mean power (W) × body mass−1 (kg)] were calculated.
Physiological responses to racing were recorded during the “2018 Sahlen’s 6 Hours at the Glen,” using a lightweight ambulatory monitoring system (Equivital EQ02; Hidalgo Ltd., Cambridge, UK). The Equivital LifeMonitor system recorded HR (bpm) and skin temperature (Tskin;°C). Core temperature (Tcore;°C) measurements were obtained from an ingestible pill sensor (VitalSense, Mini Mitter, Philips Respironics, The Netherlands) at 15-s intervals. Ingestible pill sensors were swallowed at least 3 h before the start of the first driving session in accordance with established guidelines to ensure good reliability and validity of measurement (15).
Participants wore the Equivital Life Monitor under their racing attire. All variables were recorded without interference from any race car electrical or communication systems. The percentage of the driver’s maximum HR (HRmax) was calculated using the driver’s peak HR obtained during the maximal aerobic capacity test (described above).
Physiological strain index (PSI) was determined based on Tcore (adjusted for core pill as opposed to rectal temperature) and HR using the following calculation (16):
In equation 1, Tcoret and HRt are simultaneous measurements taken during the data collection period, and Tcore0 and HR0 are the initial measurements taken once the racing attire was put on but before the race started (16). Drivers consumed 10 mL·kg body weight−1 of water 60 min before the start of the race to achieve euhydration. After the start of the race, drivers consumed food and water ad libitum.
Because all drivers drove identical vehicles, under the same conditions and on the same track, the collective effects of cockpit temperature on thermal strain (%HRmax, Tskin, Tcore, and PSI) were analyzed via linear regression analysis. The raw cockpit temperature data were plotted against the corresponding thermal strain variable. A best fit line and coefficient of determination (r2) was compared with a line with a slope of zero to determine whether there was a significant relationship between variables (alpha level of 0.05 set a priori).
The cross-sectional analyses permit the evaluation of the reliability of cockpit temperature to indicate thermal strain on racing drivers. A component of the FIA regulations refers to the amount of time racing drivers are exposed to hot conditions during a racing stint. All driving stints in this race were 60 min in length, and cockpit temperature did not differ between driving stints (P = 0.654), as such cockpit temperature data and associated thermal strain variables (%HRmax, Tskin, Tcore, and PSI) were pooled to generate a representative 60-min driving stint. A best fit line and coefficient of determination (r2) was compared with a line with a slope of zero to determine the presence or absence of a significant relationship (alpha level of 0.05 set a priori) among variables, over time. The resulting r2 values were classified based on the standards established by Moore et al. (17). The results from the longitudinal analysis allowed for the elucidation of the influence of repeated driving stints on thermal strain in race car drivers.
Recovery from thermal strain
Percent HRmax, Tskin, Tcore, and PSI were analyzed over the course of the racing event (during which drivers were both in the car and out of the car). A repeated-measures ANOVA with the main effects of (i) driving stint number (1–3) and (ii) whether the driver was in the race car (in or out) was run with an alpha level of 0.05 set a priori. If the main effects were found to be significant, a Tukey’s HSD post hoc test was run. The results from this ANOVA analysis determined whether being outside the race car reduces thermal strain on racing drivers. All analyses were performed in GraphPad Prism version 8 (GraphPad Software, San Diego, CA).
Participant fitness level and race venue information
The participant’s demographic and fitness information is displayed in Table 2. The ambient temperature at the start of race (0945 h) was 30.5°C and 36.7°C at the finish (1545 h), with an average temperature and SD of 33.8°C ± 1.3°C. Throughout the race, cockpit temperature ranged from 29.0°C to 42.4°C with an average temperature and standard deviation of 36.6°C ± 2.4°C.
TABLE 2 -
Participant fitness profile.
|Body fat (%)
|Relative V˙O2peak (mL·kg−1·min−1)
|Peak HR (bpm)
|Peak power (W)
|Mean power (W)
|Rate of fatigue (W·s−1)
|Glycolytic capacity (W·kg−1)
The best fit line for cockpit temperature and percent of HRmax (slope = 0.069 ± 0.049, r2 < 0.001, P = 0.158; Fig. 1A) was not significantly different from the zero line of best fit and the data demonstrated weak “goodness of fit.” The relationship between cockpit temperature and Tskin (slope = −0.035 ± 0.011, r2 < 0.001, P = 0.753; Fig. 1B) was also not significantly different from the zero line of best fit, and the data demonstrated weak “goodness of fit.” The cockpit temperature and Tcore (slope = −0.017 ± 0.0037, r2 = 0.007, P < 0.01; Fig. 1C) relationship was significantly different from zero, but the “goodness of fit” of the data was negative and very weak. Lastly, the relationship between cockpit temperature and PSI (slope = −0.013 ± 0.0083, r2 = 0.001, P = 0.109; Fig. 1D) was not significantly different from the zero line of best fit, and the “goodness of fit” of the data was very weak. Thus, cockpit temperature had a minimal relationship with indicators of thermal strain among race car drivers.
The analyses of cockpit temperature and physiological responses as a function of time allow for the evaluation of the cumulative effects of racing on thermal strain. Percent of HRmax showed a positive moderate relationship with driving time (slope = 0.1987 ± 0.0041, r2 = 0.505, P < 0.001; Fig. 2A), indicating that percent of HRmax increased by 2% after every 10 min of the driving stint. Tskin had a positive, moderate relationship with driving time (slope = 0.03549 ± 0.00071, r2 = 0.528, P < 0.001; Fig. 2B), suggesting that Tskin rose by 0.5°C for every 14 min of driving stint time. Consistent with Tskin, Tcore had a positive, moderate relationship with driving time (slope = 0.01395 ± 0.0003, r2 = 0.495, P < 0.001; Fig. 2C). Tcore increased 0.5°C after every 36 min of driving stint time. PSI had a positive, moderate relationship with driving time (slope = 0.04155 ± 0.00083, r2 = 0.526, P < 0.001; Fig. 2E), suggesting that every 24 min of driving results in PSI increasing by 1 U. Interestingly, cockpit temperature had a negative and weak relationship with driving time (slope = −0.01853 ± 0.0013, r2 = 0.224, P < 0.001; Fig. 2E), suggesting that cockpit temperature decreases by 0.5°C for every 54 min of driving time. Thus, thermal strain increased over 60 min of driving, whereas cockpit temperature decreased over 60 min of driving.
Recovery from thermal strain
The ANOVA analysis with the main effects of driving stint number (1–3) and whether the race car driver was in the race car (in or out) was run to determine whether race car driver indicators of thermal strain recovered after a driving stint. Percent HRmax (P < 0.001; Fig. 3A) was reduced when the drivers were outside the race car, and there was no difference in percent HRmax between driving stints 1, 2, and 3 (P = 0.877). Driving stint and being in or out of the car did not significantly interact to influence Tskin (Fig. 3B, interaction P = 0.5706, stint P = 0.7907 or in car vs out of car P = 0.4467) or Tcore (Fig. 3C, interaction P = 0.861, stint P = 0.810, or in vs out P = 0.651). PSI (Fig. 3D) was lower when drivers were out of the car after stints 1 and 3 (P < 0.001).
Race car drivers must pilot vehicles at high speeds while being exposed to physical stressors such as vibration, g forces, and heat (4). Thermal strain is potentially the most pronounced stressor placed on race car drivers and results from the heat generated by the drive train of the race car, the ambient temperature of the race venue, and the metabolic heat produced by the driver (12). The fire protective suit worn by the drivers is of particular importance in creating thermal strain, as the three-layer suit prevents evaporative heat loss. The clinical significance of thermal strain is that it is not uncommon for racing drivers to lose 3.5 kg of sweat during a 4-h race, placing them at risk of heat illness (18).
To prevent heat illness during endurance races, the FIA recently issued regulation 901-1, which (i) limits the maximum cockpit temperature to 32°C and (ii) limits driving time to 80 min during races with ambient temperature above 32°C (Table 1). Although this regulation is like protocols in other popular sports such as football and rugby, the FIA’s indirect measurement of driver heat strain via cockpit temperature does not account for the insulating fire suit. The present investigation independently evaluated the relationship between cockpit temperature and variables, indicating racing driver thermal strain. The cross-sectional analysis (Fig. 1) indicated that cockpit temperature was not an appropriate predictor of thermal strain. In fact, the results suggest that as cockpit temperature increases, driver Tcore decreases (Fig. 1C). Cockpit temperature is often the highest when the car completes a pit stop. The stationary car does not dissipate heat because of the limited airflow through the cockpit. In addition, during a pit stop, the driver is not working to pilot the vehicle (metabolic heat production decreases), consuming water, and raising his/her helmet visor to allow some airflow and convective cooling. These observations are consistent with the findings of this study—specifically the fact that as the driving stint continued, cockpit temperature decreased (Fig. 2E), yet driver thermal strain increased (Fig. 2A–D) because of the metabolic work to drive the car, the protective clothing that inhibits driver cooling and may lead to potential dehydration (2–4,19). Thereby, data from this investigation indicate that cockpit temperature is not a valid determinant of race car driver thermal strain.
The required rest time mandated by the FIA regulation during a hot race is partially supported through the results of this investigation. During driving stints, PSI was over 6, which is associated with increased risk of heat illness (13). By reducing driving time, heat illness risk is potentially limited. PSI and percent of HRmax (Fig. 3A and D) were significantly lower when the drivers were out of the car, suggesting lower thermal stress once the driver was removed from the vehicle and a lowered risk of heat illness. However, the absence of the metabolic work required to drive the car would account for the decrease in HR and, consequently, PSI. In fact, there was no difference in Tskin and Tcore (Fig. 3B and C) when drivers were in or out of the car, suggesting that recovery from thermal strain is not accomplished with 60 min of recovery from driving. It should be noted that the recovery period that drivers experienced in this study (60 min) is twice as long as that mandated by the FIA (30 min). It should also be noted that the drivers in this study attempted to cool themselves during recovery periods by removing the fire suit, sitting in an air-conditioned environment (~15°C), and consuming cold beverages. These cooling methods are common practice across the racing paddock, and more research is needed to determine the effectiveness of each method (2,4,7,11).
In the event that cockpit temperature exceeds the predetermined threshold for driver safety, the FIA regulation states that cockpit temperature must be returned to a predetermined safe level within 8 min or the car will have to stop on pit road to cooldown. The data from this investigation undermine the logic of this regulation by showing that cockpit temperature is not a good indicator of a driver’s thermal strain. In the interests of ensuring driver health, our study suggests it would be more appropriate to measure and base safety regulations on driver temperature during competition rather than cockpit temperature. Secondary in importance to driver health (but still considerations in the world of motorsport racing) are the implications of the FIA’s regulation for finishing position in a race, championship points, and financial investment from the team and sponsors. For example, the heat dissipation metric for the Corvette C7.R indicates 54 min is required to lower the cockpit temperature by 0.5°C, meaning it is unlikely for this vehicle to return to predetermined safe levels in the allotted 8 min. Thus, if the cockpit temperature of the C7.R exceeded predetermined levels, it is likely that the car would have to stop on pit road. An unnecessary pit stop could be costly in terms of performance outcomes and, as our data indicate, may not be achieving the FIA’s goal of reducing thermal strain on drivers.
The data presented suggest that cockpit temperature is not an appropriate method to monitor driver physiology. As such, we advocate for the use of a more reliable method to monitor driver thermal strain. Tcore is regarded as the best predictor of thermal strain (19). A direct Tcore measurement can be obtained with a variety of methods. The present study used an ingestible pill sensor that was linked with a data collection system, separate from the car, via Bluetooth technology (16). Although pill sensors are an accurate measure, they are single use (16) and could prevent the driver from immediately undergoing an MRI if there was a medical emergency. Both interphalangeal and rectal temperature sensors cannot be used in racing because of the fire suit and helmet that must always be worn (1,4). An alternate method, and the recommendation of the authors, is that Tcore is measured through a tympanic (ear) thermometer. Although not as accurate as rectal or interphalangeal measurements, a tympanic temperature sensor could easily be integrated into the driver’s racing set up using custom molded earpieces. Tympanic temperature sensors would provide a reliable Tcore measurement for the driver that could be feasibly incorporated with the established and mandatory equipment. A tympanic temperature sensor could also be integrated into the existing robust and required data systems, allowing teams and sanctioning bodies to monitor driver temperature during races with greater accuracy.
The present investigation is the first to independently address the FIA regulations regarding driver thermal strain. Although the physical fitness parameters and physiological responses to racing are consistent with those in the literature (2,20), there are some limitations of this investigation. A primary limitation is that data were only collected on four participants at a single racing event. Driver fluid intake and loss (hydration status) was not evaluated as the nature of competition did not allow for feasible data collection. The present investigation used the Equivital Life Monitor for data collection, which includes a single site Tskin measure. Typically, multiple skin sites are used to estimate Tskin, and the use of a single site has limitations, but it does provide a good estimate of the microclimate within the race suit. Further, data were only collected from the Corvette C7.R, which is one of several types of race cars (Ford GT, BMW M8, and Porsche 911) that compete in endurance racing. Different cars and their thermal properties should be evaluated using similar methods to those used in this investigation, to determine whether car construction influences driver thermal strain. The race itself was not an FIA sanctioned race, meaning that the cockpit temperature regulation was not enforced but encouraged to be followed. However, the extreme heat in this race allowed for the modeling of driver thermal strain and cockpit temperature, providing the information necessary to inform best practices for driver safety.
There is a paucity of scientific investigations on race car driver physiology, health, and safety risks (21). Although changes have been made to the safety equipment (HANS device, seat technology, and SAFER barrier) to protect the driver during a crash, to our knowledge FIA regulation 901-1 is the first to address the thermal stress placed on race car drivers. Therefore, the FIA should be commended for their forward thinking to protect racing drivers. Based on the results from this investigation, cockpit temperature is a weak predictor of driver thermal strain, and the authors recommend incorporating a driver metric instead of a car metric to evaluate the risk of heat illness in race car drivers.
The authors thank all members of the Corvette Racing and Pratt and Miller Engineering for their constant support and unprecedented access to the team for this project. The authors also want to thank all members of the Spartan Motorsports Performance Lab and the Human Energy Research Lab at Michigan State University for their valuable feedback and support of the project.
This project was funded by startup funds from Michigan State University, and the authors have no conflicts of interest to report. The results of this study are presented clearly, honestly and without fabrication, falsification, or any inappropriate data manipulation. The authors do not have any conflicts of interests related to this study. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Potkanowicz ES, Mendel RW. The case for driver science in motorsport: a review and recommendations. Sports Med
2. Ferguson DP, Barthel SC, Pruett ML, Buckingham TM, Waaso PR. Physiological responses of male and female race car drivers during competition. Med Sci Sports Exerc
3. Ferguson DP, Myers ND. Physical fitness and blood glucose influence performance in IndyCar racing. J Strength Cond Res
4. Ferguson DP. The Science of Motorsport Racing
. New York: Routledge; 2018.
5. Jacobs PL, Olvey SE, Johnson BM, Cohn K. Physiological responses to high-speed, open-wheel racecar driving. Med Sci Sports Exerc
6. Schwaberger G. Heart rate, metabolic and hormonal responses to maximal psycho-emotional and physical stress in motor car racing drivers. Int Arch Occup Environ Health
7. Carlson LA, Ferguson DP, Kenefick RW. Physiological strain of stock car drivers during competitive racing. J Therm Biol
8. Kenny GP, Jay O, Journeay WS. Disturbance of thermal homeostasis following dynamic exercise. Appl Physiol Nutr Metab
9. Cramer MN, Jay O. Biophysical aspects of human thermoregulation
during heat stress. Auton Neurosci
10. Jay O, Gariepy LM, Reardon FD, et al. A three-compartment thermometry model for the improved estimation of changes in body heat content. Am J Physiol Regul Integr Comp Physiol
11. Potkanowicz ES. A real-time case study in driver science: physiological strain and related variables. Int J Sports Physiol Perform
12. Carlson LA, Lawrence MA, Kenefick RW. Hydration status and thermoregulatory responses in drivers during competitive racing. J Strength Cond Res
13. Kohl HW, Gibbons LW, Gordon NF, Blair SN. An empirical evaluation of the ACSM guidelines for exercise testing. Med Sci Sports Exerc
14. Peyer KL, Pivarnik JM, Eisenmann JC, Vorkapich M. Physiological characteristics of National Collegiate Athletic Association Division I ice hockey players and their relation to game performance. J Strength Cond Res
15. Byrne C, Lim CL. The ingestible telemetric body core temperature sensor: a review of validity and exercise applications. Br J Sports Med
16. Moran DS, Shitzer A, Pandolf KB. A physiological strain index to evaluate heat stress. Am J Physiol
17. Moore D, Notz W, Flinger M. The Basic Practice of Statistics
. 6th ed. New York (NY): W. H. Freeman and Company; 2013.
18. Brearley MB, Finn JP. Responses of motor-sport athletes to v8 supercar racing in hot conditions. Int J Sports Physiol Perform
19. Sawka MN, Young AJ, Latzka WA, Neufer PD, Quigley MD, Pandolf KB. Human tolerance to heat strain during exercise: influence of hydration. J Appl Physiol (1985)
20. McKnight PJ, Bennett LA, Malvern JJ, Ferguson DP. V˙O2peak
, body composition, and neck strength of elite motor racing drivers. Med Sci Sports Exerc
21. Reid MB, Lightfoot JT. The physiology of auto racing. Med Sci Sports Exerc