Monitoring cardiac autonomic function and sleep duration in NCAA Division I football players during preseason and in-season using wearable tracking devices

Authors: Portia Resnick¹, Davis Hale², Roger Kollock², Tori Stafford², Erich Anthony³

¹Department of Kinesiology, California State University, Long Beach, Long Beach, CA
²Oxley College of Health Sciences, University of Tulsa, Tulsa, OK
³Department of Athletics, University of Tulsa, Tulsa, OK

Corresponding Author:
Portia B. Resnick, PhD, ATC, BCTMB
California State University, Long Beach
Department of Kinesiology
1250 Bellflower Boulevard
Long Beach, CA 90840
Portia.resnick@csulb.edu
908-812-9320

Portia Resnick is an assistant professor at California State University, Long Beach

Monitoring cardiac autonomic function and sleep duration in NCAA Division I football players during preseason and in-season using wearable tracking devices

ABSTRACT

Sleep duration (SD) is critical for exercise recovery, however collegiate student athletes are typically sleep deprived secondary to early morning workouts, class responsibilities, late day competitions, and travel.  As such, cardiovascular autonomic function (CAF), measured via heart rate variability (HRV) and resting heart rate (RHR), can help monitor athlete recovery.  PURPOSE:  The purpose of this study was to compare two six-week periods, preseason and in-season, on HRV, RHR, and SD in college football players.  METHODS:  Eight malecollege football players were fitted with WHOOP® wearable activity/recovery tracking devices that use photoplethysmography and accelerometry to determine HRV (RMSSD), RHR (bpm), and SD (hrs/day).  The devices were worn 24 hours a day over two six-weeks data collection periods during which the athletes participated in their normal day-to-day preseason conditioning and in-season practice sessions.  RESULTS:  A series of three, paired sample t-tests were performed to compare HRV, RHR, and SD between pooled data from preseason and in-season, reflecting the changes of the group and not the change of any individual participant.  Both HRV (preseason =100 ± 35 ms, in-season = 82 ± 34 ms, p = 0.002) and SD (preseason = 4.55 ± 1.49 hrs/day, in-season = 5.33 ± 1.55 hrs/day, p = 0.002) were different between the two six-week periods while RHR was not different (preseason = 56 ± 6 bpm, in-season = 58 ± 6 bpm, p = 0.201).  CONCLUSIONS: Athletes had higher HRV during the preseason period, indicative of greater parasympathetic activity, and had increased SD during the in-season period; however, RHR did not differ.  APPLICATIONS IN SPORT:  The examination of HRV, RHR, and SD during various periods of conditioning in collegiate football players found differences that could not be explained and therefore warrants further research.   

Key Words: heart rate variability, sleep, football

INTRODUCTION

Sleep is a crucial part of recovery from training and competition (7).  While sleep studies often focus on performance factors, sleep deprivation also leads to an increase in circulatory cortisol, and an increase in the sympathetic activity of the autonomic nervous system (ANS) (13, 16).  Sleep deprivation has the potential to impair the recovery process thereby inhibiting training adaptations which, when combined with heightened psychological stress, can lead to further increases in sympathetic activity (11).  For optimal recovery it is recommended that athletes get 7 to 9 hours of sleep per night; however, student athletes are often sleep deprived secondary to early morning practices, late night competitions, and travel, along with the requirements of maintaining a full-time academic load (9, 16).  Male athletes, in particular male field sport athletes, have a lower sleep efficiency than any other group (16). 

The time domain measure of heart rate variability (HRV), expressed as the root mean squared of the standard deviation (RMSSD), is the most reliable measure of cardiac autonomic function (CAF) (5).  Athletes with a resting heart rate (RHR) below 60 bpm, with higher HRV values may have greater parasympathetic activity (5).  While there are no established HRV norms, continuous monitoring allows for athletes and coaches to track the effects of training on the autonomic nervous system, which may relate to levels of recovery or preparedness for competition (5).  Devices worn on the wrist involve photoplethysmography to track pulse rate and pulse rate variability, which has been shown to be an effective measure of CAF in healthy subjects at rest (18).  The use of continuous monitoring through a wearable device allows for the most accurate data of HRV, RHR and sleep duration (SD) (7, 8). 

Investigating HRV in American football players allows for unique analyses as there is a mixture of player positions that rely on slightly different energy pathways.  Skill position players rely on both of the anaerobic energy pathways, ATP-PC and glycolytic, with more potential for disturbing the muscle’s buffering capacity than linemen, who focus more on the repeated bouts of peak anaerobic power with limited time for recovery between efforts.  These two types of athletes may experience varied HRV patterns based on these position differences although the reactivation of parasympathetic activity following a training session is unique to each individual (6, 12, 13, 19).  Regardless of position played, each athlete underwent a similar training protocol of early morning conditioning in the summer and late afternoon team practices and games during the fall seasons. 

Therefore, the purpose of this study was to compare two six-week periods, preseason and in-season, on HRV, RHR, and SD in college football players.  It was hypothesized that the in-season schedule would lead to an increase in SD and HRV and a decrease in RHR, indicative of increased parasympathetic activity.

METHODS

This study used a quasi-experimental interrupted time-series design.  Following IRB approval, football players identified to be significant contributors to the season were recruited to participate in data collection.  Significant contributors were those athletes identified by the coaching staff as being a starter or having a potential role that related to substantial playing time during games.  None of the athletes had any known cardiac conditions such as premature ventricular contractions nor were diagnosed with a sleep condition such as sleep apnea.  All athletes underwent informed consent, yearly pre-participation physical exams, and were cleared by the Sports Medicine staff for participation in all activities.  Following prescreening, a total of 22 athletes were selected for study participation.  For statistical comparisons, athlete data was only retained if the athlete wore the device for the study duration (15).  A final number of eight athletes participating in both time periods completed data collection for six-weeks of preseason conditioning and six weeks of in-season practices and were included for data analyses.  Completion criteria were defined as wearing the device for a minimum of five days during each of the six-week time frames.  The athletes participated in their normal day-to-day conditioning and practice sessions during the preseason and in-season data collection periods.  The two six-week periods compared were preseason early morning (6:00 am) summer conditioning and in-season fall afternoon (3:30 pm) practices.  No alterations were made to conditioning and/or practice sessions based on the data collection process. 

Athletes were instructed to wear the WHOOP® wearable activity tracking device system 24 hours a day, seven days a week during both six-week periods.  The WHOOP® device utilizes three techniques to assess HRV, HR, and SD: a tri-axial accelerometer, an optical sensor, and a touch/ambient temperature sensor at a sampling rate of 100 Hz.  Wrist based resting assessment of HRV via photoplethysmography is a valid method of measuring R-R intervals and wrist based accelerometry has been accepted as a reliable and valid measure of sleep time, staging, and wakefulness (2, 10, 14, 17, 18). A recent study by Berryhill and colleagues (2020) reported strong evidence of the WHOOP® device’s ability to consistently measure sleep and cardiac autonomic function when compared to ECG and self-reported sleep logs (3).  With the WHOOP® device, recovery metrics of HRV and RHR are collected during the last cycle of sleep, the preferred time to measure CAF.  The data collected during this cycle was used for analysis.  In addition to the HRV, RHR and SD measures, the device also uses accelerometry to measures exercise which is given in a score termed strain.  While this measure was collected during the data collection period, the data were not analyzed for this study.  The WHOOP® device is worn on the wrist about 2 cm from the distal end of the radius and ulna and is connected via Bluetooth technology to an application on a smartphone.  Data were collected continuously throughout the day and downloaded to the application on a smartphone via Bluetooth® technology by the wearer.   The smartphone application displayed physiological data in real time as well as provided a historical overview.  The researchers were able to view and analyze the outputs of the raw data provided by the manufacturer.  The participants had no access to the metrics provided by the device to limit bias.

For the purpose of this study, each time period was defined as a 42-day training cycle (preseason and in-season).  The mean of the three variables across both six-week training periods (42-days per athlete) for those who contributed a minimum of five days of data per week for both cycles were calculated for statistical comparisons.  Thus, data from the athletes during both the preseason and in-season periods were analyzed for the impact on HRV (RMSSD reported in ms), RHR (bpm) and SD (reported in hrs/day).  The Statistical Package for the Social Sciences SPSS (v24, IBM Corp., Chicago, IL, USA) was used for the statistical analyses.  Prior to data comparisons, a Kolmogorov-Smirnov test of normality was performed for each of the three variables and data were determined to be from a normal distribution.  A series of paired sample t-tests were performed to compare mean preseason and in-season data for each of the three variables (HRV, RHR, and SD) with the significance level set at p = 0.05.  Effect sizes were calculated utilizing Cohen’s d (1).

RESULTS

Preseason HRV was significantly greater than in-season HRV (see table 1) indicating greater parasympathetic activity during the preseason with a medium effect size (d = 0.516).  Sleep duration was significantly longer (medium effect size: d = .0513) in-season when athletes had mandatory breakfast at 7:00am and practice at 3:30pm (see table 1) compared to the preseason when training sessions started at 6:00am.  The correlation between HRV and SD, while positive, was not significant during either the preseason (r = 0.67, n = 80, p = 0.278) or in-season (r = 0.028, n = 81, p = 0.402).  Resting HR was not different between the two time periods (see Table 1) and reported a small effect size (d = 0.214). 

Table 1. Training Cycle Comparisons: Preseason & In-Season Paired Samples t-test

		Mean	Std. Deviation	p	d	Effect Size
HRV (ms)	HRV Preseason	100	35	0.002*	0.516	medium
	HRV In-season	82	34
RHR (bpm)	RHR Preseason	56	8	0.201	0.214	small
	RHR In-season	58	6
SD (hrs)	SD Preseason	4	1.49	0.002*	0.513	medium

* Significantly different at p ≤ 0.05.

Table 1. compares the variables Heart Rate Variability (HRV) in milliseconds (ms) and Resting Heart Rate (RHR) in beats per minute (bpm) taken during the lasts sleep cycle and Sleep Duration (SD) in hours (hrs) as averaged over 6-week time frames during preseason and in-season for 8 NCAA Division I Collegiate Football players.

DISCUSSION

During preseason period the athletes averaged less than five hours of sleep per night, consistent with previous research on athletes with practices that began between 5:00am and 6:00am (16).  The increase of an average of about one hour in sleep during the in-season period coincided with one hour later report time (7:00am mandatory breakfast) indicating that the athletes were going to sleep at about the same time during the preseason and in-season, gaining the hour with the later wake-up time.  Based on circadian rhythms, it is probable that the athletes were not just unable to go to sleep earlier during the summer conditioning period based on their schedule, but also based on their own biorhythms (9, 16).  In both time periods the athletes were sleeping less than the recommended seven to nine hours per night.  While the easy solution to gain sleep would be to recommend later practices, this may not be a viable solution in all cases.  For best practices, the coaching staff may choose to schedule conditioning during the mornings to avoid the warmer ambient temperatures of the summer afternoons that place increased heat stress on the athletes.  Therefore, from a clinical standpoint, the better solution might be to educate the athletes about the need for more sleep or for the utilization of naps when early morning practice or training is scheduled.  Taking a 20-minute nap has proven to be beneficial in endurance runners and provides a more practical solution than either an earlier bedtime or a later wake-up (4). 

Even with the later practice schedule and the increase in sleep, the athletes showed higher sympathetic activity in the in-season data collection period as indicated by the lower HRV values.  The pre-season schedule consisted of conditioning and weight training with limited non-contact football activities while the in-season schedule added practices with contact, games and travel.  It is possible that the increase in contact activity as well as the sympathetic activity associated with performance in games led to the decrease in HRV (16, 19).  Based on NCAA regulations, athletes are limited to 8 hours of conditioning during the summer which increases to 20 hours per week of activity in-season, therefore the number of hours per week of activity increased from pre-season to in-season in addition to the type of activity changing.  The demands of an in-season schedule also include travel and games that fall outside of the activity limit, potentially adding additional stressors.  Therefore, it is considerable that the type and length of activity, contact practices and games, led to the increase in sympathetic activity which was not offset by the significant increase in SD or the later wake-up time (9, 19). 

Although during these time periods none of the monitored athletes had an injury, sleep deprivation is also associated with neurocognitive deficits therefore making sleep monitoring important in athletes who have a suspected or diagnosed concussion (13).  Diminished sleep can also lead to increased bodily discomfort and an amplification of pain (11).  Sleep is suggested to play a vital role in synaptic plasticity as well as having multiple functions for physical recovery, thereby making sleep an important part of injury prevention and treatment (11).

Even though the results associate a later wake-up time for collegiate football players with an increase in SD, the study had its limitations.  First, athlete compliance was reported at an average of 5 days/week which is not totally inclusive of all time during the 24/7 monitoring recommended by WHOOP®.  There was inconsistency in the days during which the athletes wore the device making it difficult to examine specific days of the week, for example how the response on a Monday differed from a Thursday, or to track changes in one individual over a consistent five-day period consistently throughout either the preseason or in-season.  In addition, the preseason and in-season periods contained different activities due to the nature of collegiate sports and the training cycle.  Therefore, many factors that were not measured during these time periods may have contributed to changes in HRV, RHR, and SD.  Few studies have investigated sleep duration and cardiac autonomic function in collegiate football players during different training periods.  The need to establish individualized trends in football athlete adaptation and recovery is still needed in the sport studies literature.  Future research in this area may include the investigation of the impact of workload on these aspects of recovery.   

CONCLUSIONS

Both SD and HRV differed between pre-season and in-season in college football players.  Athletes had higher HRV, indicative of greater parasympathetic activity, during the preseason period, while SD increased during the in-season period.  The RHR did not change between the two time periods.

APPLICATIONS IN SPORT

Optimal training requires the balance of physiological demands with appropriate recovery.  This study indicates that the time of year or type of training may result in differing SD and HRV levels in college football players as recorded by the WHOOP® device.  The ability to translate this into usable clinical data in regards to recovery requires further investigation. 

ACKNOWLEDGEMENTS

The authors would like to acknowledge The University of Tulsa Athletic Performance and Sports Medicine staff for their cooperation on the project.  The project was supported by the American Athletic Conference Academic Consortium Research Grant Program.

REFERENCES

1. Batterham AM and Hopkins WG. Making meaningful inferences about magnitudes. International journal of sports physiology and performance 1: 50-57, 2006.
2. Beattie Z, Oyang Y, Statan A, Ghoreyshi A, Pantelopoulos A, Russell A, and Heneghan C. Estimation of sleep stages in a healthy adult population from optical plethysmography and accelerometer signals. Physiological measurement 38: 1968-1979, 2017.
3. Berryhill S, Morton CJ, Dean A, Berryhill A, Provencio-Dean N, Patel SI, Estep L, Combs D, Mashaqi S, Gerald LB, Krishnan JA, and Parthasarathy S. Effect of wearables on sleep in healthy individuals: a randomized crossover trial and validation study. J Clin Sleep Med 16: 775-783, 2020.
4. Blanchfield AW, Lewis-Jones TM, Wignall JR, Roberts JB, and Oliver SJ. The influence of an afternoon nap on the endurance performance of trained runners. Eur J Sport Sci 18: 1177-1184, 2018.
5. Buchheit M. Monitoring training status with HR measures: do all roads lead to Rome? Frontiers in physiology 5: 73, 2014.
6. Chen JL, Yeh DP, Lee JP, Chen CY, Huang CY, Lee SD, Chen CC, Kuo TB, Kao CL, and Kuo CH. Parasympathetic nervous activity mirrors recovery status in weightlifting performance after training. Journal of strength and conditioning research / National Strength & Conditioning Association 25: 1546-1552, 2011.
7. Flatt AA, Esco MR, Allen JR, Robinson JB, Earley RL, Fedewa MV, Bragg A, Keith CM, and Wingo JE. Heart rate variability and training load among National Collegiate Athletic Association Division 1 college football players throughout spring camp. Journal of strength and conditioning research / National Strength & Conditioning Association 32: 3127-3134, 2018.
8. Flatt AA, Esco MR, Nakamura FY, and Plews DJ. Interpreting daily heart rate variability changes in collegiate female soccer players. The Journal of sports medicine and physical fitness, 2016.
9. Fullagar HH, Skorski S, Duffield R, Hammes D, Coutts AJ, and Meyer T. Sleep and athletic performance: the effects of sleep loss on exercise performance, and physiological and cognitive responses to exercise. Sports medicine 45: 161-186, 2015.
10. Gil E, Orini M, Bailon R, Vergara JM, Mainardi L, and Laguna P. Photoplethysmography pulse rate variability as a surrogate measurement of heart rate variability during non-stationary conditions. Physiological measurement 31: 1271-1290, 2010.
11. Halson SL. Nutrition, sleep and recovery. European Journal of Sport Science 8: 119-126, 2008.
12. Heffernan KS, Kelly EE, Collier SR, and Fernhall B. Cardiac autonomic modulation during recover from acute endurance versus resistance exercise. European Journal of Cardiovascular Prevention and Rehabilitation 13: 80-86, 2006.
13. Kirschen GW, Jones JJ, and Hale L. The impact of sleep duration on performance among competitive athletes: A systematic literature review. Clinical journal of sport medicine : official journal of the Canadian Academy of Sport Medicine, 2018.
14. Marino M, Li Y, Rueschman MN, Winkelman JW, Ellenbogen JM, Solet JM, Dulin H, Berkman LF, and Buxton OM. Measuring sleep: accuracy, sensitivity, and specificity of wrist actigraphy compared to polysomnography. Sleep 36: 1747-1755, 2013.
15. Plews DJ, Laursen PB, Kilding AE, and Buchheit M. Evaluating training adaptation with heart-rate measures: a methodological comparison. International journal of sports physiology and performance 8: 688-691, 2013.
16. Roberts SSH, Teo WP, and Warmington SA. Effects of training and competition on the sleep of elite athletes: a systematic review and meta-analysis. British journal of sports medicine 53: 513-522, 2019.
17. Sargent C, Lastella M, Halson SL, and Roach GD. The validity of activity monitors for measuring sleep in elite athletes. Journal of science and medicine in sport / Sports Medicine Australia 19: 848-853, 2016.
18. Schafer A and Vagedes J. How accurate is pulse rate variability as an estimate of heart rate variability? A review on studies comparing photoplethysmographic technology with an electrocardiogram. International journal of cardiology 166: 15-29, 2013.
19. Stanley J, Peake JM, and Buchheit M. Cardiac parasympathetic reactivation following exercise: implications for training prescription. Sports medicine 43: 1259-1277, 2013.