Responses of Driver-Athletes to Repeated Driving Stints

Abstract

Having overcome old societal stereotypes, the race car driver is now recognized as a true athlete. Research in driver science continues to contribute to this change by expanding its scope to examine multiple forms of racing, such as endurance racing, which challenges the driver-athlete beyond that of a traditional race.

Purpose 

The purpose of this study was to examine and quantify the effect of repeated driving stints on the physiologic, metabolic, and hormonal responses of three professional endurance driver-athletes.

Methods 

Core body temperature, HR, and physiological strain index were recorded during the Rolex 24 Hours of Daytona endurance race using the Equivital Life Monitor system. Blood glucose was monitored continuously during the event using a FreeStyle Libre Pro (Abbott, Alameda, CA). Alpha-amylase and cortisol were sampled immediately before the beginning of a stint and immediately after.

Results 

First-stint overall and individual driver-athlete responses were similar to those reported in the literature. Later-stint responses diverged from the literature. Reductions in initial core temperature, absence of increases in HR and physiological strain index, and altered glucose and hormonal responses were each observed in the later stint.

Conclusion 

The data support previous research showing that motorsports has a measurable physiological, metabolic, and hormonal effect on the driver-athlete. This study also shows that multiple stints elicit responses that deviate from the published literature on single-stint events. This study is also particularly interesting in that it represents one of the first times that longitudinal data have been gathered on endurance racing driver-athletes.

The scientific literature has established that the race car driver is a true athlete. Evidence for this statement can be found in the manuscripts that have been published on the topic of driver science in recent years. Where once there may have been only 30 or so peer-reviewed articles examining the physical challenges of motorsport, there is now a growing number of articles that directly address the topic of the driver-athlete and driver science (^1–23). In addition, motorsport competition as a whole is recognized within the scientific literature as being an extremely challenging physical endeavor (^1–3), and scientists continue to learn more about what the driver-athlete endures while racing (^{4–6,10,11,15–20,22}). For example, Potkanowicz (⁶) reported core body temperatures as high as 38.78°C and 38.42°C in two elite level drivers competing in the Petit Le Mans. Similarly, Barthel et al. (²²) reported core body temperatures of 38.4°C and 38.2°C during competition in professional and amateur sports car drivers, respectively. Barthel and colleagues also observed that compared with professional drivers, amateur drivers had a reduced glycolytic capacity, which has the potential to compromise driving performance in this particular group. Brearley and Finn (¹⁷) offer that in long-form V8 Supercar competition, physiological strain index (PSI) levels can rise to levels of 8, or “high,” on the 0–10 PSI scale. Moreover, in one of the more recent papers, Ferguson et al. (⁴) offer data that contradict the long-held belief that female driver-athletes experience greater physiological fatigue while driving because of their menstrual cycle then their male counterparts. Rather, the authors contend that the type of car being driven (i.e., open vs closed cockpit) and the years of driving experience influence thermal stress and performance more than the menstrual cycle alone.

Although these studies have expanded our understanding of the driver’s response to shorter-duration races and have laid to rest the fallacy that the driver is not an athlete, what is lacking in the scientific literature is an examination of how these physiological responses change during an endurance race. Endurance racing is one of the more demanding forms of motorsport competition, the intention of which is to challenge the driver (and the car) for much longer periods of time and across repeated driving stints.

Endurance racing in motorsports has been around for over 100 yr. One of the earliest recorded endurance events was the Coppa Florio, a 540-km event first contested in 1900 on the Circuit de Bologna in Italy (²⁴). The world’s first formally organized 24-h race was contested in 1905 at the Driving Park, a 1-mile oval in Columbus, Ohio (²⁵). Since then, endurance events have been run on 1000-mile courses, like the Mille Miglia, over multiple days and even in the deserts between Paris and Dakar. Today, years after the first Coppa Florio, driver-athletes still compete for a variety of titles and trophies, the most prestigious of which is endurance racing’s “Triple Crown,” which involves winning the 24 Hours of Le Mans, the 12 Hours of Sebring, and the 24 Hours of Daytona.

As defined by the Federation Internationale De L’Automobile World Endurance Championship Regulations, an endurance race is a motorsport competition lasting no less than 4 h and requires a team of minimally two drivers and, in some longer races, as many as four (²⁶). For perspective, the Indy 500 is a single-driver race, with the longest 500 of the modern-era lasting 3:43:051 in 1992. Along with being a mechanism for testing the durability of race cars, endurance racing also tests the endurance of the driver-athlete. For example, according to the Technical and Sporting Regulations that govern the 24 Hours of Le Mans (²⁶), a driver must be behind the wheel “for at least six hours but no more than 14 hours in all.” In addition, a driver cannot drive “for more than four hours within any six-hour period,” extended periods of time where both driver-athlete and car are competing at their maximal physical and mechanical limits, respectively. In addition to the challenge of time, there are several issues known to affect human performance that must be managed throughout an endurance race in order for the driver to be successful. For example, adequate hydration, changing environmental conditions, and a feeding strategy that provides sufficient calories and nutritional density for adequate recovery each represent individual and combined challenges that must be addressed over the course of the race. Given these types of regulations, and given these challenges to physical performance, it becomes evident that the challenges that an endurance driver-athlete encounters, although similar in nature to those of a shorter-duration race, far exceed what he or she might encounter in a “traditional” race event (i.e., races of lesser total mileage or time).

Examining endurance racing, therefore, expands the scope of driver science investigations beyond the driver-athlete’s response to a single-stint or shorter-duration event by considering its additional challenges, including multiple long-duration driving stints, long stretches of time without meaningful sleep, and concern for proper feeding and hydration for both preparation and recovery. By quantifying a driver’s physiological, metabolic, and hormonal responses over the course of multiple long-duration stints, we can uncover possible performance deficits and additional safety risks in races such as the 12 h of Sebring, the 24 Hours of Le Mans, or, as is the focus of this study, the 24 Hours of Daytona. Therefore, the purpose of this investigation was to examine and quantify the effect of repeated driving stints on the physiologic, metabolic, and hormonal responses of endurance driver-athletes during the running of the 2018 24 Hours of Daytona.

METHODS

The study was approved by the participating institutions’ institutional review boards before data collection. Before participating in the study, the participants had the protocol explained to them, and each participant was provided with a health history questionnaire to complete and a written informed consent to read and sign. Informed consent was secured in the presence of the principal investigator.

As a natural experiment, and given the uniqueness of the competitive racing environment, the authors did not intervene or attempt to establish controls over race-related variables in the study design, such as changing environmental conditions or the drivers’ preferred feeding and sleeping routines. Rather, the authors consider racing-related variables necessary to elicit a true competitive physiological response from each of the drivers. Simply put, in-race competition needs to be evaluated exactly as it is happening, as there is a differential physiological response between competitive and practice environments shown in the scientific literature (^4,7,22). It is important to understand the mechanisms of change in competitive driver physiology to provide evidence-based insight and to promote driver safety. The methodology presented here is similar to that used by the authors in previous work and, therefore, will be described here in a similar fashion (^4–6,22).

Participants

The volunteer research participants were three licensed professional driver-athletes competing in the Daytona Prototype International class of the International Motor Sports Association series at the “Rolex 24 Hours of Daytona” sports car race. Driver demographics are presented in Table 1.

TABLE 1 - Driver demographics, stint order, stint duration, and accumulated driving time.

Characteristic/Driver	Driver 1	Driver 2	Driver 3
Age (yr)	32	38	27
Height (m)	1.77	1.77	1.83
Weight (kg)	71.2	86.6	71.7
Stint 1—start	14:30	16:25:01	18:28:03
Stint 1—end	16:23:40	18:26:44	21:35:38
Stint 1—duration	1:53:40	2:01:43	3:07:35
Wt. in/wt. out/Δ (kg)	71.0/70.8/−0.2	86.3/86.20/−0.1	71.5/70.2/−1.3
Stint 2—start	21:36:57	00:09:17	02:40:55
Stint 2—end	00:07:53	02:39:33	05:14:05
Stint 2—duration	2:30:56	2:30:16	2:33:10
Wt. in/wt. out/Δ (kg)	71.0/71.09/+0.09	86.2/86.5/+0.3	71.6/71.3/−0.3
Stint 3—start	05:15:26	No 3rd Stint	No 3rd Stint
Stint 3—end	06:46:48	No 3rd Stint	No 3rd Stint
Stint 3—duration	1:31:22	No 3rd Stint	No 3rd Stint
Wt. in/wt. out/Δ (kg)	Not Collected	No 3rd Stint	No 3rd Stint
Total accumulated driving Time a	5:55:58	4:31:59	5:40:45

The team did not finish the entire 24-h race. Because of mechanical problems, the car was withdrawn from the race on the morning of 1/28/2018 after 16:08:42 of competition. There were 07:51:18 remaining in the race.

^aMinus any time in the pits.

Physiological Responses to Racing

Core body temperature, HR, PSI, and body mass

Core body temperature (Tgi) and HR were monitored continuously throughout the race. Core body temperature and HR were monitored using the Equivital Life Monitor® (Hidalgo, Cambridge, United Kingdom) paired with the VitalSense® core temperature capsule (Philips Respironics, Philips, NV). This device is approved for motorsport as it does not interfere with the electronics of the race car. In addition, it has been deemed “safe for competition” by the Federation Internationale De L’Automobile and has been used in motorsports (^4,7,27). Four hours before the start of the race, drivers ingested the core temperature pill. This was done to ensure that the pill was in the small intestine of the driver during the race. Before putting on their driving suit, drivers were fitted with the Equivital life monitor, and data were collected every 15 s.

PSI was calculated using the Tgi and HR data collected during the race. Because one of the participants exhibited an HR response above 180 bpm during his driving stints, PSI for all three drivers was calculated using the equation of Tikuisis et al. (²⁸). Tikuisis et al. modified the original equation of Moran et al. (²⁹) to allow for HR above the 180 bpm. Tikuisis et al. calculates PSI as follows:

$\mathrm{PSI}=5\left(T_{\mathrm{ret}}–T_{\text{re0}}\right){\left(39.5–T_{\text{re0}}\right)}^{-1}+5\left({\mathrm{HR}}_{\mathrm{t}}–60\right){\left({\mathrm{HR}}_{\max }–60\right)}^{-1}$

where HR_max represents the maximal observed HR and 60 represents the arbitrarily chosen resting HR value proposed by Tikuisis et al.

Body mass was assessed pre- and poststint for monitoring fluid changes. Each driver was weighed before the beginning of each stint and as soon as reasonably possible after the completion a stint. Drivers were weighed separately from their fire suits and safety gear wearing only their underwear using a Sunbeam Health-O-Meter 349KLX digital scale. Changes in body mass pre- to poststint are presented in Table 1. As is seen in Table 1, no driver had a greater than 2% loss in body mass, and as such, saliva measures were not influenced by dehydration.

Metabolic Responses to Racing

Blood glucose

Four hours before the start of the race, drivers had a continuous blood glucose monitor (FreeStyle Libre Pro, Abbott) secured to their preferred arm, superficial to the belly of the triceps. Blood glucose data were collected at 15-min intervals throughout the entirety of the race.

Hormonal Responses to Racing

Salivary alpha-amylase and cortisol

Just before entering the race car and immediately upon exiting the race car at the conclusion of a driving stint, an approximately 200-μL saliva sample was obtained from each driver. The saliva was stored on ice for later analysis of alpha-amylase (an epinephrine surrogate) and cortisol. Saliva was analyzed using enzyme-linked immunosorbent assay (LifeSpan BioSciences, Inc., Seattle, WA) according to established techniques (^30–32). Briefly, samples were vortexed in Eppendorf tubes and serial diluted according to manufacturer instructions before being loaded onto a 96-well plate with specificity for cortisol or alpha-amylase. A microplate reader (Bio-Rad Industries, Hercules, CA) compared samples (in triplicate) against a manufacturer’s supplied standard. Only samples within the linear range of the instrument were analyzed.

Data Analysis

An exploratory data analysis was conducted to aid in the interpretation of the observed data and to highlight trends that occurred within the individual drivers and across all the drivers during the race. No between-driver comparisons were made because of the potential interindividual differences caused by race logistics in the uncontrolled nonlaboratory environment and the observed sample size.

Mixed linear models were used to evaluate overall driver trends during driving and between driving stints. Two-way models (stint: 1 or 2 × time: continuous variable, minutes) were used to test for changes in blood glucose, core body temperature, HR, and PSI. Random effects for the slope and intercept of each driver within each stint were included in the models to account for the repeated-measures design. Stint 3 was excluded from the models as it was only completed by one driver. Contrasts were used to test if the slope of the dependent variable over time within the driving stint was different from zero, if the slope of stint 2 was different from stint 1, and if the initial values (intercept) of stint 2 were different from stint 1. Two-way (stint: 1 or 2 × time: categorical variable, pre or post) models were used to test for differences in alpha-amylase and cortisol. These models included random effects for driver within time and stint to account for the repeated-measures design. Contrasts were used to test pre- to postdriving differences, differences between stint 1 and stint 2 predriving, and the absolute change from pre- to postdriving values between stints 1 and 2.

To evaluate individual driver trends, three-way (stint × time × driver) multiple linear regression models were used. Where significant main effects were observed, contrasts were used to determine whether the slope of the dependent variable over time was different from zero, if the slopes of stint 2 or 3 were different from stint 1, and if the initial values (intercept) of stint 2 or 3 were different from stint 1. All post hoc contrasts were completed using the Bonferroni correction method for multiple comparisons. All statistical analyses were conducted using R version 4.0.1 (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was set to α = 0.05.

RESULTS

The results for the physiological, metabolic, and hormonal responses are presented below. The stint order, stint duration, and accumulated driving time for each driver are presented in Table 1. Because of mechanical problems, the team was unable to finish the race and the car was withdrawn after 16:08:42 of competition. Driver 1 completed three stints, whereas both drivers 2 and 3 drove two stints.

Physiological Responses to Racing

Core body temperature

Average core temperature data across drivers during driving stints 1 and 2 are presented in Figure 1A. During stint 1, an increase in core body temperature over time was observed (P = 0.057). The slope of core body temperature over time during stint 2 was not different to stint 1 (P = 0.315). At the initiation of stint 2, core body temperature was decreased compared with stint 1 (P = 0.032).

FIGURE 1: Overall driver trends for (A) Tgi, (B) HR, (C) PSI, and (D) blood glucose data during driving (with linear regression lines and SE).

Individual driver core temperature data are presented in Figure 2A. During driving stints 1 and 2, drivers 2 and 3 showed a significant increase in core body temperature across time (P < 0.001). Driver 1 did not show any changes in core body temperature during his driving stints (P ≥ 0.10). Driver 2 showed a significant decrease in the rate of change in core body temperature in stint 2 relative to stint 1 (P < 0.001). All other stints showed a similar slope of core temperature relative to the driver’s initial driving stint (P ≥ 0.325). Relative to core temperature at the beginning of stint 1, drivers 1 and 3 showed a decreased core temperature in their subsequent driving stints (P ≤ 0.04).

FIGURE 2: Individual driver (A) core body temperature (Tgi), (B) HR, (C) PSI, and (D) blood glucose data during driving (with linear regression lines and SE).

HR

HR data averaged across drivers during driving are presented in Figure 1B. A significant decrease in initial HR at the start of stint 2 was observed compared with stint 1 (P < 0.001). During driving, no changes in the slope of HR over time were observed in either of the driving stints (stint 1: P = 0.252; stint 2: P = 0.394; stint 1 vs stint 2: P = 0.544).

Individual driver HR data are presented in Figure 2B. Drivers 1 and 2 in stints 1 and 2 showed a significant decrease in HR while driving (P ≤ 0.037). The changes in HR during all other driving stints were not different from zero (P ≥ 0.264). Relative to stint 1, no differences in the slope of HR over time were observed (P ≥ 0.999). However, compared with the driver’s initial HR during stint 1, all drivers showed a decrease in HR in all subsequent stints (P ≤ 0.013).

PSI

Average stint PSI data while driving are presented in Figure 1C. On average, drivers showed a significant decrease in PSI at the start of stint 2 relative to initial values from stint 1 (P < 0.001). No changes in PSI over time were observed in either driving stint (stint 1: P = 0.212; stint 2: P = 0.493; stint 1 vs stint 2: P = 0.456).

Individual driver PSI data while driving are presented in Figure 2C. Changes in PSI during the driving stint were observed in driver 1 (stint 2), driver 2 (stint 1), and driver 3 (stints 1 and 2) (P ≤ 0.001). In addition, driver 2 showed a decrease in the change of PSI over time during stint 2 relative to stint 1 (P < 0.001). All other driving stints were not different from stint 1 (P ≥ 0.205). All drivers showed a decrease in initial PSI at the start of stint 2 or 3 relative to their initial values during stint 1 (P < 0.001).

Metabolic Responses to Racing

Blood glucose

Blood glucose data were available for drivers 1 and 2; data for driver 3 are unavailable because of device malfunction. Average driver blood glucose data are presented in Figure 1D. On average, drivers showed an increase in blood glucose concentrations during stint 1 (P = 0.008), but no changes in blood glucose concentrations were observed during stint 2 (P = 0.903). Therefore, the rate of change in blood glucose during driving was different in stint 1 vs stint 2 (P = 0.008). There was no difference in the initial starting blood glucose concentrations between stints 1 and 2 (P = 0.542).

Individual driver blood glucose data are presented in Figure 2D. Slopes of blood glucose over time were not different than 0 for any of the driving stints (P ≥ 0.067). However, because the slope of the stint 1 trial of driver 1 was positive and stint 2 was negative, a difference between trials was observed (P = 0.016). Driver 1 showed a decrease in initial blood glucose concentrations before stint 3 compared with stint 1 (P = 0.011). In both drivers, initial blood glucose concentrations during stint 2 were not different from initial values during stint 1 (P ≥ 0.129).

Hormonal Responses to Racing

Alpha-amylase

Pre- and postdriving alpha-amylase concentrations across all drivers are presented in Figure 3A. After driving, alpha-amylase concentrations were increased in both stint 1 (P < 0.001) and stint 2 (P < 0.001) compared with predriving concentrations. There was no difference in the absolute change in alpha-amylase concentration from pre- to postdriving between stints (P = 0.393). Before stint 2, drivers displayed a decrease in alpha-amylase compared with values taken before stint 1 (P < 0.001). Individual driver pre– and post–alpha-amylase data are presented in 4A.

FIGURE 3: A, Pre– and post–alpha-amylase observed across stints, presented as mean ± SD. B, Pre- and postcortisol observed across stints, presented as mean ± SD.

Cortisol

Average pre- and postdriving cortisol data are presented in Figure 3B. Cortisol concentrations were significantly increased after driving in stint 1 (P < 0.001). Before stint 2, cortisol concentrations remained elevated compared with values before stint 1 (P = 0.039). However, during stint 2, concentrations were not significantly different after driving compared with predriving values (P = 0.076). The absolute change in cortisol from pre- to postdriving was significantly greater in stint 1 compared with stint 2 (P = 0.033). Individual driver pre- and postcortisol data are presented in Figure 4B.

FIGURE 4: Individual driver (A) alpha-amylase and (B) cortisol data while driving, presented as mean ± SD.

DISCUSSION

The purpose of this investigation was to quantify the effect of repeated stints on select physiological, metabolic, and hormonal responses in driver-athletes. In doing so, the study sought to identify overall time-related trends and individual driver responses while competing in the 2018 24 Hours of Daytona.

Using past research findings as a guide, one might intuitively assume that the drivers’ responses to racing in this study would be similar. Although the data are similar in some instances, the results of this study also suggest that drivers exhibit responses specific to endurance racing that we were not previously aware of and are not presented in the driver science literature. For example, where authors have previously spoken of increases in core body temperature (^{1,2,4–6,11,15,17}) while driving, the current investigation saw a similar increase in core temperature in overall driver trends for the two stints that were evaluated. However, what the data also show that is new is that these endurance drivers started their second stint with a significantly lower core body temperature compared with the first (Fig. 1A). Examining the core temperature data of individual drivers (Fig. 2A), the current investigation demonstrates a similar divergence from the published literature that may be the result of multiple, long-duration driving stints. For example, drivers 2 and 3 showed significant increases in core body temperature across time in both stint 1 and 2. However, although driver 2 had a very similar rate of change in core temperature in both of his stints, driver 3 exhibited a significant decrease in the rate of change in core body temperature in stint 2 relative to stint 1. In addition, driver 1 is of particular individual interest as he exhibited no change in core temperature across time in any stint.

Past research (^{1,2,4–6,11,15,17}) has suggested that drivers exhibit increases in HR and PSI while driving. That was not the case in the current investigation for either variable. Examining the overall driver trends (Fig. 1B), HR was observed to decrease over time. Although this response differs from the driver science literature, two particular overall responses stand out in the context of endurance racing. First, like core temperature, there was a significant decrease in the drivers’ initial HR at the beginning of stint 2 compared with stint 1. Second, there was no change in the slope of HR over time in either stint 1 or stint 2. A similar HR response is seen in the individual driver data as well (Fig. 2B), save for driver 2 who did exhibit an increase in HR over time in each stint, although his stint 2 HR response was lower than his stint 1 response.

Examining PSI, there is a similar divergence from the driver science literature. The overall driver trends for PSI in this study showed no change over time in either of the driving stints (Fig. 1C). However, like core temperature, there was a significantly lower overall initial PSI value at the beginning of stint 2 compared with stint 1. This same kind of difference in starting PSI values is also seen in the individual driver data (Fig. 2C).

Regarding the metabolic responses of the drivers, the observation by others (^22,33) that blood glucose increases in response to a driving stint is supported by this investigation, at least during stint 1. During stint 1, there was a significant increase in average driver blood glucose concentration over time, similar to what has been reported in the scientific literature. However, as it relates to the possible influence of additional stints, this same increase in blood glucose associated with driving was not seen during stint 2 (Fig. 1D). The individual blood glucose data for one driver also suggest that endurance racing may elicit a response different than what has been reported. Looking at driver 1, although his blood glucose increases over time during stint 1, there is a decrease in blood glucose during stint 2, regardless of the fact that he started each stint with a relatively similar blood glucose level (Fig. 2D).

Hormonally, the overall driver trends for alpha-amylase exhibited the expected increase over time associated with increases in physical activity during both stints. There was, however, a significant decrease in the drivers’ initial alpha-amylase level at the beginning of stint 2 compared with the initial levels of stint 1, similar to the response observed in core temperature and other study variables (Fig. 3A). Individually, the alpha-amylase response of each driver is similar to the overall trend (Fig. 4A). Although each of the three drivers, regardless of stint, showed a significant increase in pre- to poststint alpha-amylase levels, drivers 2 and 3 had significantly lower stint 2 prelevels compared with their respective stint 1 prelevels. The difference in driver 1 was observed between his stint 1 and stint 3 prelevels, with alpha-amylase being lower at the start of stint 3.

Cortisol also exhibited an expected, but slightly dissimilar, increase in response to the increased physical and psychoemotional stress of driving (Fig. 3B). Although average cortisol levels rose over time during stint 1, they did not return to prestint levels when the stint was over. Rather, cortisol levels remained elevated and the drivers began stint 2 with a higher average cortisol level. Although elevated at the start of stint 2, cortisol did not exhibit a significant increase over time pre- to poststint. Both the elevated post–stint 1 cortisol and the absence of a change in cortisol during stint 2 are interesting observations in the context of the possible influence of additional stints on driver responses. Individually, the cortisol response of each driver is similar to the overall trend, with only driver 2 having a lower pre–stint 2 value compared with his pre–stint 1 value (Fig. 4B).

The results of this study show that the changes observed in the endurance racing driver-athlete are initially similar in many ways to those observed in driver-athletes competing in more traditional races. However, the observed differences begin to diverge from previous investigations as the time of exposure is extended (i.e., multiple stints). What the data presented here cannot pinpoint is the exact reason for the observed changes. As a natural experiment conducted during an actual sanctioned race event with the team competing for series points, the current investigation was limited in its ability to control variables that may have influenced the observed changes. As was mentioned by Ferguson et al. (⁴), “The nature of automobile racing makes controlled studies difficult as the race venue, mechanical problems with the car, and incidence of a crash can hinder data collection.” Expanding on this statement, considerations such as environmental conditions, the appropriateness of the driver’s feeding and hydration, or the quality and length of his sleep had the potential to affect the observed changes.

Using feeding as an example, consider that driver 1 started his third stint with a blood glucose level significantly lower than what was recorded for his first stint (Fig. 2D). Further, the rate of change of driver 1 was more severe during his third stint compared with his first. Comparatively, driver 2 was able to maintain a relatively stable glucose level preceding his two stints. It is reasonable to conclude that these differences in driver responses are the result of macronutrient concentration differences in their respective meals, or the timing of those meals during recovery, variables the study was not permitted to control.

Using core temperature as a second example (Fig. 2A), driver 1 had a significantly lower core temperature at the beginning of his third stint compared with his first stint, whereas driver 3 was observed to have had a significantly lower core temperature at the beginning of his second compared with his first. It stands to reason that the observed change in the core temperature of driver 1 was the result of his third stint beginning at 0515 h, a time very near to when we typically experience our lowest body temperature of the diurnal cycle.

Alternatively, an environmental perspective may provide a different explanation for the observed changes in core body temperature differences. Driver 1 began the race with an ambient temperature of 22.2°C and a relative humidity of 66%. When he got back in the car over 7 h later at 2136 h, the ambient temperature had dropped to 16.6°C and the humidity had risen by 18%. As a second example, driver 2 began his first stint with a cockpit temperature of 23.3°C. When he returned to the cockpit for his second stint just a little over 1.5 h later, the cockpit temperature had dropped to 21.2°C, a difference of 2.1°C between the two stints. Similarly, the reduction in core temperature of driver 3 may very well have been the result of his second stint starting at a time of day when the ambient conditions were more favorable (i.e., 00:09:17).

In conclusion, the data from the current investigation support previous research showing that participating in motorsport competition has a measurable effect on the driver-athlete. Beyond that, this study showed that, even in the face of numerous uncontrolled variables, there are consistent and observable physiological, metabolic, and hormonal changes during an endurance race that may be different from what has been observed and reported about traditional length and duration races. This study is also particularly interesting in that it represents one of the first times that longitudinal data have been gathered on endurance racing driver-athletes.

In addition, the current investigation highlights the complexity of endurance racing and the challenges associated with conducting natural experiments in this environment. A variety of variables led to the changes that were observed in each of the driver-athletes, and it is difficult to measure the exact contribution of each variable because of the inherently dynamic nature of endurance racing. Because an endurance driver is behind the wheel for varying lengths of time, during both day and nighttime stints, and in changing environmental conditions, each new stint is, to a great degree, a new race and only one of multiple exposures. However, it is this varied nature and the almost constantly changing conditions of endurance racing that are reflected in the data and that argue for the continuation of this work in the future.