Reliability of Intestinal Temperature Using an Ingestible Telemetry Pill System During Exercise in a Hot Environment : The Journal of Strength & Conditioning Research

Secondary Logo

Journal Logo

Original Research

Reliability of Intestinal Temperature Using an Ingestible Telemetry Pill System During Exercise in a Hot Environment

Ruddock, Alan D.; Tew, Garry A.; Purvis, Alison J.

Author Information
Journal of Strength and Conditioning Research: March 2014 - Volume 28 - Issue 3 - p 861-869
doi: 10.1519/JSC.0b013e3182aa5dd0
  • Free



Exercise in a hot environment challenges the function of cardiovascular (17), metabolic (14), and thermoregulatory systems (24); thus, exposure to prolonged, intense exercise that raises body temperature increases the risk of heat illness (1) and impairs athletic performance. Subsequently, strategies have been designed to help athletes perform in hot environments, all of which are designed to regulate the body temperature. A heat tolerance test is useful for assessing the ability to regulate the body temperature and measurements of body temperature responses are integral to interpretation. Where improved heat tolerance is required, decisions regarding the most appropriate method and their specific derivatives are determined; these are usually grouped into (a) heat acclimation or acclimatization protocols and (b) cooling methods. The most appropriate method is context specific, but the choice is driven by whether body temperature responses to a strategy are sufficient to improve safety and performance. Consequently, it is important that practitioners use valid and reliable methods to monitor body temperature (6) to make well-informed and clear decisions about the success of the chosen method. Esophageal and rectal sites are most commonly used for assessing core body temperature (6,30); however, ingestible telemetry pill systems that measure intestinal temperature are being increasingly used to overcome the impracticalities associated with traditional methods (5). Several studies provide evidence to suggest that telemetry pill systems are valid tools for the assessment of core temperature (12,13,15,16,25,26); however, there are little data regarding the reliability of intestinal temperature measurement during exercise in a hot environment.

Ingestible telemetry pills are often used when environmental temperature is high (≥25° C) to monitor core body temperature or detect treatment effects after the application of an intervention (e.g., cooling). For example, the reported magnitude of intestinal temperature change after a treatment in a hot environment ranges from 0.30 to 1.8° C (27,33,35). As such, ingestible telemetry pill systems must demonstrate a small test-retest error for practitioners to be confident that these changes are true and unlikely because of measurement error. Incorrect interpretation of intestinal temperature because of measurement error or failing to understand the limits of the system could place athletes at an increased risk of heat illness and impair athletic performance. To improve confidence when interpreting a change in a dependent variable, it is important to identify the magnitude of the measurement error within repeated measurements. Performing such an assessment assists in interpreting whether a change has occurred because of a treatment or because of the inherent measurement error (biological or technical) and is also important for calculating sample size (3,11,20,21,29).

To date only 2 studies have published data regarding the interday test-retest reliability of intestinal pill telemetry systems. Goosey-Tolfrey et al. (18) conducted a small reliability study (n = 5) using ingestible telemetry pills (CorTemp, HQinc, Palmetto, FL, USA) in wheelchair athletes exercising intermittently at a submaximal intensity (50% peak power output [Wpeak]) for 60 minutes in a hot environment (30.8° C, 60.1% relative humidity [RH]). The authors reported a mean test-retest error of 0.30° C (95% confidence interval [CI], 0.20–0.40) and limits of agreement (LOA 95%) of ±1.2° C. In a cooler environment (15.0° C, 60% RH), Gant et al. (16) investigated the reliability of telemetry pills (CorTemp, HQinc) during intermittent shuttle running and reported excellent reliability with a near absent test-retest error of 0.01° C (CI, −0.02–0.05° C) and LOA 95% of ±0.23° C.

Given that ingestible telemetry pill systems are increasingly being used to monitor core body temperature responses and that important health, treatment-effect and performance decisions are made on these assessments, it is important to establish the test-retest reliability of this method. The purpose of this study is to investigate the interday reliability of intestinal core temperature using an ingestible telemetry pill system during exercise within a hot environment.


Experimental Approach to the Problem

The subjects were asked to visit the exercise physiology laboratory on 3 separate occasions. Each visit was separated by a minimum of 3 days and maximum of 10 days. During visit 1, the subjects performed an incremental exercise test to maximal volitional exertion for the assessment of peak oxygen uptake (V[Combining Dot Above]O2peak) and associated power output (Wpeak) at room temperature (20° C, 40% RH) followed by a 30-minute exercise-heat stress accustomization trial within an environmental chamber (35° C dry bulb temperature, 50% RH). During visits 2 and 3, the subjects rested for 60 minutes in a temperature-controlled environmental chamber set at 35° C, 50% RH before cycling at a power output equivalent to 50% Wpeak for 60 minutes. The combination of environmental conditions and intensity of exercise was chosen to induce moderate-to-high heat strain within an ethically acceptable limit of (39.5° C intestinal temperature). All exercise trials were performed at the same time of the day and at least 3 hours after waking and 3 hours before sleeping to minimize the circadian rhythm impact on body temperature and gastrointestinal (GI) function (28). All the exercise trials were conducted using an externally verified, electromagnetically braked arm crank ergometer (Lode Angio, Groningen, the Netherlands) positioned for recumbent cycling.


Twelve non–heat acclimated physically active men (age 25 ± 4 years, stature 181.7 ± 7.0 cm, body mass 81.1 ± 10.6 kg) volunteered for the study, which was approved by the Ethics Committee of Sheffield Hallam University and conducted according to the principles of the Declaration of Helsinki. No subjects were under 18 years of age. The risks and experimental procedures were fully explained, and all the subjects provided written informed consent before the study. The subjects were instructed to refrain from performing strenuous exercise and consuming caffeine and alcohol 24 hours before each trial and from taking food 3 hours before each trial. The subjects recorded their diet 24 hours before the first heat stress trial and replicated their diet for the retest. Adherence to the standardized diet was verified by the investigator before each trial. All the experimental trials were conducted between October and January in the United Kingdom, where the mean maximum ambient temperatures ranged from 14 to 7° C.

Procedures: V[Combining Dot Above]O2peak Test and Accustomization Trial

The subjects undertook a 10-minute warm-up period before physiological testing. Peak oxygen uptake (V[Combining Dot Above]O2peak) and associated power output (Wpeak) were assessed using a continuous incremental exercise test to maximal volitional exertion. The initial intensity of exercise was set at 50 W and was increased by 25 W·min−1 in a stepped manner. Pedaling frequency was self-selected in the range of 60–80 rpm, and the subjects were encouraged to continue to maximal volitional exertion. Breath-by-breath pulmonary oxygen uptake was measured continuously using a calibrated low dead space (20 ml) bidirectional differential pressure pneumotach and rapid response galvanic O2 and nondispersive infrared CO2 analyzers (Ultima CardiO2, Medgraphics, St. Paul, MN, USA). Peak oxygen uptake was calculated as the highest 30-second mean average value recorded before the termination of the test. The power output at the last completed stage was used to determine Wpeak. Heart rate (HR) was recorded continuously during exercise using short-range telemetry (RS400, Polar Electro, Kempele, Finland).

After 60 minutes of rest in a temperate environment, the subjects undertook 30 minutes of recumbent cycling within the environmental chamber (35° C, 50% RH) at an intensity of 50% Wpeak to accustomize with the experimental procedures. Nude body mass (kilograms) was recorded before and after the accustomization trial along with the volume of water ingested ad libitum during the trial. Sweat rate (milliliters per minute) was estimated from the change in the body mass adjusted for fluid intake and urine output. Fluid requirements (which matched sweat rate) were calculated for each subject and were closely adhered to in the main experimental trials. During the exercise session, the HR, rating of perceived exertion (RPE), and thermal perception (TP) (32) were assessed every 5 minutes.

Heat Stress Trials 1 and 2

These sessions were used to assess the reliability of intestinal temperature measurements. The subjects were instructed to replicate their diet 24 hours before each trial and to consume 500 ml of noncaffeinated fluid in the preceding 2 hours to promote euhydration (2). On arrival, the subject's nude body mass (kilograms) and urine osmolality (milliosmoles per kilogram of H2O; Advanced Model 3320 Micro-Osmometer, Advanced Instruments Inc, Norwood, MA, USA) were assessed. A urine osmolality of ≤700 mOsmol·kg−1 H2O was used to verify pretrial and posttrial euhydration status (8). The subjects drank 100 ml of cold water (4° C) to verify the positioning of the telemetry pill in the GI tract. We observed changes in the pill temperature >0.1° C for 2 subjects; in these instances, the tests were rescheduled. For all the trials, there was no change in pill temperature >0.03° C after cold water ingestion. After these initial measures, the subjects rested for 60 minutes in the environmental chamber, which was set at 35° C, 50% RH to stabilize intestinal and skin temperature before commencing exercise. At rest and during exercise, the subjects were encouraged to drink room temperature water (35° C) to meet their individual fluid requirements. Measures of HR, skin temperature (Tsk), intestinal temperature (Tint), RPE, and TP were recorded immediately before exercise (0 minutes) and at 5-minute intervals throughout the exercise trial. The exercise test was terminated when 1 of the following criteria was met: (a) the subjects voluntarily stopped exercising, (b) when the intestinal core temperature reached 39.5° C, or (c) the subjects completed 60 minutes of exercise.

Intestinal Temperature

Intestinal temperature was assessed using a telemetric monitoring system consisting of an ingestible temperature sensor and a data logger (CorTemp, HQinc). Before ingestion, the temperature measurement of each pill was verified by immersion in a water bath at 4 temperatures (37, 38, 39, and 40° C) according to recommended guidelines (22). The water temperature was verified using a calibrated thermometer. A linear regression relationship between measured (pill temperature) and actual (water bath) temperatures was derived, and the resulting regression equation was used to convert the measured temperature during exercise to the actual temperature. The coefficient of determination of this relationship was r2 = 0.99, and the 2 methods demonstrated excellent agreement (LOA 95%−0.01 ± 0.09° C). The subjects ingested temperature sensors at the same time of the day, 6 hours before each visit (5). This ingestion time has been reported to be sufficient to allow the telemetry pill to pass into the GI tract and produce valid intestinal temperature measurements (13).

Skin Temperature

Four skin thermistors (Grant Instruments, Shepreth, United Kingdom; Tegaderm 3M Healthcare, Bracknell, United Kingdom) were attached to the left side of the body at the medial calf, anterior midthigh, anterior midforearm, and chest using acrylic dressing (Tegaderm; 3M Healthcare, USA) and secured in place using hypoallergenic surgical tape (Transpore; 3M Healthcare). Weighted mean skin temperature (Tsk) was calculated using the equation of Ramanathan (34). The mean body temperature (Tbody) was calculated as Tbody = 0.66(Tint) + 0.34(Tsk) at rest and Tbody = 0.79(Tint) + 0.21(Tsk) during exercise (10).

Statistical Analyses

The reliability of GI temperature, which in this study refers to the reproducibility of day-to-day measurements at identical time points during exercise, was assessed through a number of statistical analyses following the guidelines of Atkinson and Nevill (3). Bland–Altman plots (4) were generated to investigate systematic and random error trends. The presence or absence of heteroscedasticity was formally investigated by plotting a regression line through the data points of the Bland–Altman plots. A paired t-test was used to assess the systematic bias between trials, with the statistical significance set at p ≤ 0.05. Coefficient of variation (CV), standard error of measurement, and 95% LOA were used to assess the absolute reliability. Relative reliability was assessed using Pearson's correlation coefficients and intraclass correlation coefficients (ICCs). Cohen's d was used as a measure of effect size, and data were evaluated according to small (0.2), medium (0.5), and large (0.8) effects (9). A paired t-test was used to assess between-trial differences in environmental conditions, urine osmolality, skin temperature, sweat rate, fluid intake, and HR. Statistical significance was set at p ≤ 0.05. The Gaussian distribution of data was assessed using Kolmogorov–Smirnov test and was accepted when p ≥ 0.05.

We acknowledge that practitioners prefer to use different approaches when assessing reliability and that the acceptable levels of error may differ among researchers. Therefore, we have presented a range of reliability statistics and avoided using stringent and predefined acceptable levels of error. To provide a real-world practical context to reliability data, we present hypothetical changes in the intestinal temperature (when used in a comparison study this is an approach used to determine the effectiveness of a treatment) and used the approach of Hopkins (21) to interpret the magnitudes of change. This analysis determined the chance (probability) that a change was harmful, trivial, or beneficial in context to the error (reliability) of the test and smallest worthwhile change. It also enabled the identification of the minimum change in intestinal temperature that is deemed to be a likely beneficial change (76% probability), which in the example was a decrease in the intestinal temperature.


The subjects' mean V[Combining Dot Above]O2peak elicited during the incremental exercise test was 36.5 ± 5.2 ml·kg−1·min−1 corresponding to a mean peak power of 293 ± 36 W. The mean duration of heat stress trials was 55.4 ± 9.4 minutes completed at 147 ± 18 W. There were no statistically significant differences for environmental temperature (p = 0.38) and RH (p = 0.74) between trial 1 (35.0 ± 0.2° C, 49 ± 3% RH) and trial 2 (35.1 ± 0.4° C, 49 ± 5% RH). Urine osmolality at the start of trial 1 (317 ± 179 mOsmol·kg−1 H2O) was not significantly different (p = 0.95) from the start of trial 2 (313 ± 194 mOsmol·kg−1 H2O). The intestinal temperature at the start of trial 1 was similar to that at the start of trial 2 (37.26 ± 0.23 vs. 37.26 ± 0.25° C; p = 0.99). However, the skin temperature was marginally lower at the start of trial 1 (35.43 ± 0.35 vs. 35.70 ± 0.37° C, p = 0.01). Sweat rate did not differ markedly between trials 1 and 2 (29 ± 5 vs. 28 ± 5 ml·min−1; p = 0.63). Ad libitum fluid intake helped restrict body mass deficits in trial 1 to 0.82 ± 0.58% and trial 2 to 0.84 ± 0.52%, which were not significantly different (p = 0.87). Heart rate and skin and body temperatures progressively increased throughout the heat trials reaching peak values at the end of exercise as evidenced in Table 1 and Table 2. There was no statistically significant difference in the mean HR between trial 1 and trial 2 (p = 0.97).

Table 1:
Physiological and perceptual responses during exercise averaged over 10 minutes (mean ± SD).*

The mean intestinal temperature after exercise (Table 2) indicates that the intensity and the duration of exercise were sufficient to induce moderate-to-high levels of heat strain in both trials. The mean intestinal temperature bias between trials 1 and 2 was −0.07 ± 0.31° C indicating a small but statistically significant systematic bias (p = 0.02). Visual inspection of Figure 1 illustrates that the intestinal temperature in trial 2 was consistently higher than that in trial 1 from 10 minutes onward until the termination of exercise.

Table 2:
Mean intestinal temperature (±SD) data for the entire trial and 10-minute blocks of exercise.*
Figure 1:
Mean intestinal temperature (degrees centigrade) during exercise. Error bars represent the SD.
Table 1
= trial 1;
Table 1
= trial 2.

The LOA for intestinal temperature are presented in Figure 2. The scattering of data points indicated a degree of both systematic and random errors proportional to the measurement range of intestinal temperature. Although the slope of the relationship was close to zero (−0.16), a statistically significant (p < 0.01) small correlation coefficient (r = 0.26) confirmed the presence of heteroscedasticity.

Figure 2:
Bland–Altman plot exhibiting variations in the intestinal temperature (degrees centigrade) measurement recorded every 5 minutes during exercise (n = 145). The solid line represents the mean intestinal temperature bias. Dashed lines represent the 95% limits of agreement (random errors).

Reliability data for the entire trial and each 10-minute segment of the trial can be found in Table 3. Using data from the entire set indicates a good overall reliability for all statistical methods. However, when data were grouped by time (Table 3), small changes in error were evident over the exercise duration. The error displays a trend that increases from the onset of exercise, before peaking midexercise and decreasing toward the end of exercise.

Table 3:
Reliability statistics for entire trial and 10-minute blocks of exercise.*

Because the Bland–Altman plot suggested the presence of heteroscedasticity, the CV (which assumes the largest test-retest error occur at higher values) was used to determine the probabilities of a change in intestinal temperature in relation to the smallest worthwhile change using the approach of Hopkins (21) (Table 4). This analysis suggests that a change in the intestinal temperature as a result of a treatment or intervention (e.g., cooling) of 0.34° C is required to be interpreted as a likely beneficial change (≥76% chance).

Table 4:
Probabilities of hypothetical changes in the intestinal temperature.*


The main finding of this study is that the measurement of intestinal temperature using intestinal telemetry pills demonstrated a good test-retest reliability during submaximal recumbent cycling exercise in the heat. In accordance with the recommendations suggested by Atkinson and Nevill (3), we used a range of reliability statistics to assess intestinal temperature. Inspection of Bland–Altman plots identified a small degree of systematic and random error proportional to the measured range of intestinal temperature and suggested the presence of heteroscedasticity. Assessment of intestinal temperature is often used to determine the success of an intervention (e.g., prevent increases in “core” temperature). To provide a practical context to the reliability data and investigate whether hypothetical changes in intestinal temperature were likely greater than the smallest worthwhile change, we used the approach of Hopkins (21). We identified the smallest magnitude of the change in the intestinal temperature required to detect a likely beneficial change (76% chance) to be 0.34° C. Changes <0.34° C decrease the confidence in interpretation and augment the uncertainty of the true effect of an intervention. This information might be useful for scientists interpreting changes in intestinal temperature because of an intervention and should be considered when decisions regarding body temperature need to be made from health, treatment-effect, and performance perspectives. For example, a decrease in intestinal temperature of 0.25° C as a result of an intervention has a 67% chance that it is a beneficial change, 19% chance trivial, and 14% chance harmful (Table 4). When the decrease in the intestinal temperature is 0.5° C, there is an 88% likely probable chance that this is a beneficial change and a 4% very unlikely chance that it is a harmful change. In this example, a decrease in core temperature may not always be a true decrease when considering the magnitude of change in context to the error of the system and smallest worthwhile change. It should be noted that to improve confidence in interpretation, a smaller error (better reliability) would be required.

We observed a systematic error of −0.07° C and LOA (95%) ± 0.61° C, which is lower than the mean bias of 0.30° C and LOA (95%) of ±1.2° C reported by Goosey-Tolfrey et al. (18) despite a similar intensity of exercise, duration, and environmental conditions in both studies. The differences may be explained by a smaller sample size (n = 5) and a large variation in intestinal temperature of 1 subject in the Goosey-Tolfrey et al. (18) investigation. The mean bias in this study is greater than that reported by Gant et al. (16) who observed a near absent test-retest systematic error of 0.01° C and LOA of ±0.23° C during intermittent running in a cool environment. Furthermore, the values reported for correlation coefficients, ICC, and CV indicate better reliability in the study by Gant et al. (16). This observation of larger systematic and random errors in the intestinal temperature during exercise in hot environments is in agreement with data of Jette et al. (23) who studied the reliability of the rectal temperature during exercise in 2 different conditions in men wearing chemical protective clothing. At 20° C, no statistically significant difference in the mean rectal temperature change between trials was observed (−0.01° C). However, at 40° C, the authors reported a statistically significant mean difference of −0.05° C, which is similar to the mean bias observed within the current investigation.

A potential explanation for the increased variation in the intestinal temperature might be the GI function. The GI tract is a complex organ and the many potential interactions within it could serve to increase the variability observed in GI pill temperature in heat stressed humans. The recommended pill ingestion time of 6 hours before intestinal temperature measurement reflects a balance between sensor gastric emptying and expulsion time (5). We observed no changes in pill temperature >0.03° C when the subjects ingested 100 ml of cold water 1 hour before exercise, confirming that the pill ingestion time of 6 hours provided sufficient time for gastric emptying. However, 2 tests out of 24 were rescheduled because of a change in the observed pill temperature of >0.1° C. Wilkinson et al. (37) demonstrated that drinking cold water (5–8° C) influenced the pill temperature by up to 2° C 8 hours after pill ingestion. The authors suggested that this temperature variation may be caused by the pill residing in areas of the GI tract in close proximity to the stomach (e.g., duodenum and transverse colon). To reduce the potential for pill temperature fluctuations in this study, the subjects ingested water at room temperature (35° C) at frequent intervals throughout the trial, and as a result, we observed no large variations in the intestinal temperature.

In a further attempt to standardize GI function, the subjects refrained from taking food 3 hours before each test, and their diet was replicated in the preceding 24 hours. However, exercise speeds intestinal peristaltic velocity (36), and it is conceivable that peristaltic velocity might have been different between trials. Subsequently, this variation could have changed the position of the pill within the GI tract. Indeed, we observed an increase in temperature variability midexercise. As eluded to by Gant et al. (16), this midexercise variability may have been caused by peristalsis advancing the pill along the GI tract before reaching more compact fecal matter toward the end of exercise, thereby reducing temperature variability.

Several human studies provide an indirect evidence to support the notion of a temperature gradient along the GI tract because the GI pill temperature is consistently higher than the rectal temperature (5). Further evidence is available from animal studies that demonstrate a temperature gradient along with the GI tract because duodenum and ilium temperatures were significantly higher than in the stomach, large intestine, and rectum (19). As a result, one possible explanation for the observed variation in intestinal temperature is that variability in the intestinal peristaltic velocity advances the pill to areas of the GI tract, which may exhibit different temperatures. This interpretation suggests that although a pill ingestion time of 6 hours before exercise is sufficient for gastric emptying, perhaps a longer period is required to limit the effects of peristaltic velocity.

Several studies have presented data indicating that each pill has a bias from certified thermometry; as such, it is recommended that each pill is individually calibrated. Following the recommendations by Hunt and Stewart (22), we noted excellent agreement (−0.01 ± 0.09° C) between a verified electronic thermometer and intestinal pills. This agreement was similar to that reported within previous studies (7,22,37) and within the guidelines set by Moran and Mendel (30) who suggest that random errors between thermometers should not be >±0.1° C. To reduce potential systematic and random bias, we applied a linear regression equation to each pill determined after verification. Therefore, it is unlikely that the variation in the intestinal temperature during exercise was because of invalid temperature measurement by the system. Furthermore, the variation in intestinal temperature was not likely caused by a difference in the total energy expenditure between the 2 trials because there was no statistically significant difference between trials for mean HR (p = 0.97).

Our findings are only applicable to pill ingestion times of 6 hours. It is likely that different ingestion times will change pill location in the GI tract and influence the reliability of the method. Although skin temperature, TP, and RPE were high, the mean end-exercise intestinal temperature was approximately 38.60 ± 0.4° C, which is around 1–1.5° C lower than the reported voluntary exercise termination core temperature. As such, our results should be applied with caution to temperatures >39° C; however, during self-paced exercise, which is common in most sports, intestinal temperature may be regulated to be around 39° C (31). We also applied pill-specific regression equations to account for bias, and as such, our findings might only be applicable after corrections have been employed. We recommend following the verification guidelines of Hunt and Stewart (22) before dispensing telemetry pills for ingestion as conducted in this study. Recumbent cycling was chosen as the mode of exercise as laser-Doppler flowmetry, which requires a stable upper body, was used to assess forearm skin blood flow (data not presented). Generalizations to other modes of exercise should be applied with caution; however, it is likely that our results are applicable to other modes of exercise that produce similar energetic and thermoregulatory demands.

Practical Applications

We have assessed the reliability of intestinal temperature sensors using a range of statistical approaches, and practitioners are encouraged to interpret these results using their preferred method. By using hypothetical data we have demonstrated that practitioners can be confident that the observed changes in intestinal temperature >0.34° C as a result of an intervention in a hot environment are likely beneficial and less influenced by error associated with the method. This interpretation is important from a safety perspective for monitoring athletes within hot environments, when determining the effect of a treatment on intestinal temperature responses, and when making performance-based decisions. For example, understanding the reliability of a method and obtaining reliable temperature measurements are important for the prescription of appropriate training intensities in hot environments, especially at high intensities where the body temperature can rise to induce considerable strain on cardiovascular, metabolic, and thermoregulatory systems and place athletes at risk of heat illness. Heat tolerance tests are often used to determine whether athletes can regulate body temperature in hot environments or require acclimation and acclimatization before competition. Reliable body temperature measurements are needed to ensure that athletes who are heat intolerant are not considered tolerant and vice versa because incorrect decisions because of measurement error can place athletes at risk of heat illness and waste vital training and adaptation time and financial resources. In instances where athletes require acclimation, intestinal temperature is often used to determine the effectiveness of the intervention. It is important that practitioners can be confident that changes in the body temperature are the result of an adaptation to heat and not because of measurement error. Similarly, where cooling methods are required to lower core temperature before, during, or after exercise, scientists require reliable measurements of core temperature to elucidate the most appropriate method and to monitor the health of athletes. We have demonstrated that intestinal temperature measurement assessed using an ingestible telemetry pill system is capable of providing scientists with reliable temperature measurements so that health, treatment-effect, and performance decisions can be made with confidence when accounting for the error within the method.


The authors wish to acknowledge the Academy of Sport and Physical Activity, Sheffield Hallam University, for the funding of this research. The authors have no conflict of interest and declare that the findings of this study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association.


1. American College of Sports Medicine, Armstrong LE, Casa DJ, Millard-Stafford M, Moran DS, Pyne SW, Roberts WO. American College of Sports Medicine position stand. Exertional heat illness during training and competition. Med Sci Sports Exerc 39: 556–572, 2007.
2. American College of Sports Medicine, Sawka MN, Burke LM, Eichner ER, Maughan RJ, Montain SJ, Stachenfeld NS. American College of Sports Medicine position stand. Exercise and fluid replacement. Med Sci Sports Exerc 39: 377–390, 2007.
3. Atkinson G, Nevill AM. Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Med 26: 217–238, 1998.
4. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1: 307–310, 1986.
5. Byrne C, Lim CL. The ingestible telemetric body core temperature sensor: A review of validity and exercise applications. Br J Sports Med 41: 126–133, 2007.
6. Casa DJ, Becker SM, Ganio MS, Brown CM, Yeargin SW, Roti MW, Siegler J, Blowers JA, Glaviano NR, Huggins RA, Armstrong LE, Maresh CM. Validity of devices that assess body temperature during outdoor exercise in the heat. J Athl Train 42: 333–342, 2007.
7. Challis GG, Kolb JC. Agreement between an ingestible telemetric sensor system and a mercury thermometer before and after linear regression correction. Clin J Sport Med 20: 53–57, 2010.
8. Cheuvront S, Sawka MN. Hydration assessment of athletes. GSSI Sports Science Laboratory 18: 2005.
9. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates, 1988.
10. Colin J, Timbal J, Houdas Y, Boutelier C, Guieu JD. Computation of mean body temperature from rectal and skin temperatures. J Appl Physiol 31: 484–489, 1971.
11. Currell K, Jeukendrup AE. Validity, reliability and sensitivity of measures of sporting performance. Sports Med 38: 297–316, 2008.
12. Darwent D, Zhou X, van den Heuvel C, Sargent C, Roach GD. The validity of temperature-sensitive ingestible capsules for measuring core body temperature in laboratory protocols. Chronobiol Int 28: 719–726, 2011.
13. Domitrovich JW, Cuddy JS, Ruby BC. Core-temperature sensor ingestion timing and measurement variability. J Athl Train 45: 594–600, 2010.
14. Galloway SDR, Maughan RJ. Effects of ambient temperature on the capacity to perform prolonged cycle exercise in man. Med Sci Sports Exerc 29: 1240–1249, 1997.
15. Ganio MS, Brown CM, Casa DJ, Becker SM, Yeargin SW, McDermott BP, Boots LM, Boyd PW, Armstrong LE, Maresh CM. Validity and reliability of devices that assess body temperature during indoor exercise in the heat. J Athl Train 44: 124–135, 2009.
16. Gant N, Atkinson G, Williams C. The validity and reliability of intestinal temperature during intermittent running. Med Sci Sports Exerc 38: 1926–1931, 2006.
17. Gonzalez-Alonso J, Calbet JA. Reductions in systemic and skeletal muscle blood flow and oxygen delivery limit maximal aerobic capacity in humans. Circulation 107: 824–830, 2003.
18. Goosey-Tolfrey V, Swainson M, Boyd C, Atkinson G, Tolfrey K. The effectiveness of hand cooling at reducing exercise-induced hyperthermia and improving distance-race performance in wheelchair and able-bodied athletes. J Appl Physiol 105: 37–43, 2008.
19. Grayson J, Durotoye AO. Effect of environment on temperatures in the viscera of the dog. Int J Biometeorol 15: 176–180, 1971.
20. Hopkins WG. Measures of reliability in sports medicine and science. Sports Med 30: 1–15, 2000.
21. Hopkins WG. How to interpret changes in an athletic performance test. Sports Sci 8: 1–7, 2004.
22. Hunt AP, Stewart IB. Calibration of an ingestible temperature sensor. Physiol Meas 29: N71–N78, 2008.
23. Jette M, Quenneville J, Thoden J, Livingstone S. Reproducibility of body temperature response to standardized test conditions when assessing clothing. Ergonomics 38: 1057–1066, 1995.
24. Johnson JM. Exercise in a hot environment: The skin circulation. Scand J Med Sci Sports 20(Suppl 3): 29–39, 2010.
25. Kolka MA, Quigley MD, Blanchard LA, Toyota DA, Stephenson LA. Validation of a temperature telemetry system during moderate and strenuous exercise. J Therm Biol 18: 203–210, 1993.
26. Kolka MA, Levine L, Stephenson LA. Use of an ingestible telemetry sensor to measure core temperature under chemical protective clothing. J Therm Biol 22: 343–349, 1997.
27. Leicht AS, Sinclair WH, Patterson MJ, Rudzki S, Tulppo MP, Fogarty AL, Winter S. Influence of postexercise cooling techniques on heart rate variability in men. Exp Physiol 94: 695–703, 2009.
28. Manfredini R, Manfredini F, Fersini C, Conconi F. Circadian rhythms, athletic performance, and jet lag. Br J Sports Med 32: 101–106, 1998.
29. Marino FE, Kay D, Cannon J, Serwach N, Hilder M. A reproducible and variable intensity cycling performance protocol for warm conditions. J Sci Med Sport 5: 95–107, 2002.
30. Moran DS, Mendel L. Core temperature measurement: Methods and current insights. Sports Med 32: 879–885, 2002.
31. Morante SM, Brotherhood JR. Autonomic and behavioural thermoregulation in tennis. Br J Sports Med 42: 679–685, 2008; discussion 685.
32. Nielsen R, Gavhed DC, Nilsson H. Thermal function of a clothing ensemble during work: Dependency on inner clothing layer fit. Ergonomics 32: 1581–1594, 1989.
33. Ozgunen KT, Kurdak SS, Maughan RJ, Zeren C, Korkmaz S, Yazici Z, Ersoz G, Shirreffs SM, Binnet MS, Dvorak J. Effect of hot environmental conditions on physical activity patterns and temperature response of football players. Scand J Med Sci Sports 20(Suppl 3): 140–147, 2010.
34. Ramanathan NL. A new weighting system for mean surface temperature of the human body. J Appl Physiol 19: 531–533, 1964.
35. Schlader ZJ, Raman A, Morton RH, Stannard SR, Mundel T. Exercise modality modulates body temperature regulation during exercise in uncompensable heat stress. Eur J Appl Physiol 111: 757–766, 2011.
36. van Nieuwenhoven MA, Brouns F, Brummer RJ. The effect of physical exercise on parameters of gastrointestinal function. Neurogastroenterol Motil 11: 431–439, 1999.
37. Wilkinson DM, Carter JM, Richmond VL, Blacker SD, Rayson MP. The effect of cool water ingestion on gastrointestinal pill temperature. Med Sci Sports Exerc 40: 523–528, 2008.

core temperature; heat stress; gastrointestinal tract; reproducibility; monitoring; thermoregulation

Copyright © 2014 by the National Strength & Conditioning Association.