Sports Medicine and Orthopaedic


Effects of Overtraining Status on the Cortisol Awakening Response—Endocrine and Metabolic Responses on Overtraining Syndrome (EROS-CAR)

Cortisol is the predominant circulating glucocorticoid released in response to stressful stimuli. In addition to acute stress responses, cortisol is one of the major circadian regulating hormones, and it exhibits a robust and distinct circadian and diurnal profile.

Cortisol concentrations are typically greatest in the early morning, before decreasing across the daytime period and reaching its nadir in the late evening. In response to waking from nocturnal sleep, cortisol concentrations will diverge from the normal linear early morning increase and will increase by 50% to 100% in the 30 to 45 minutes after waking. This rise occurs independently from the cortisol circadian rhythm. The phenomenon has been termed the cortisol awakening response (CAR), and has been associated with a variety of psychological and physiological conditions, particularly fatigue states and compromised sleep quality.


Despite the paucity of CAR research within the exercise science literature, both exercise training and competition appear to impact the CAR. CAR is a distinct component of the typical cortisol profile and may be a useful indicator of the development of syndromes in response to excess exercise training. Given the proposed role of the CAR as a resource-mobilizing phenomena, a blunted CAR may be viewed as an undesirable compensatory effect, while an elevated CAR may represent a necessary adjustment to the normal training-recovery process during times of substantial training stress.

The overtraining syndrome (OTS) has been investigated for many years, and yet no single biomarker is currently available for the indication or diagnosis of the condition. CAR is a promising candidate, as it represents both the underlying basal adrenal output of cortisol and the reactivity of the HPA axis. The cortisol concentration at awakening is most sensitive to poor sleep quality, and its impairment is indicative of non-repairing sleep, one of the remarkable findings in OTS.

While the utility of the CAR to detect OTS is promising, there are presently no studies that have systematically investigated whether the CAR is affected by OTS or whether this biomarker, or the diurnal cortisol slope, is altered by the development of OTS.

The aim of this study was to investigate the differences in the CAR and diurnal slope between healthy athletes(ATL), athletes diagnosed with OTS, and healthy sedentary controls (SED) from a post hoc re-analysis of the Endocrine and Metabolic Responses on Overtraining Syndrome study.

Athletes (ATH; n = 25) were males aged 18–50 years old, participating in at least 4 sessions and a total of 300 minutes of moderate to intense exercise training per week, for at least 6 months. Participants suspected of OTS (OTS; n = 14) were diagnosed with OTS if reductions in performance of 10% or more were confirmed by a sport specialist and were unexplained by illness or overt medical, endocrine, or psychological conditions. Furthermore, the OTS athletes were required to report persistent fatigue, increased perceived exertion during exercise training, and reduced sleep quality. Approximately 90% of athletes initially considered as candidates for OTS were excluded during the participant screening.

Within the ATL group, 96% of the athletes competed in high-intensity functional training. Athletes in ATL reported 550.0 (±180.2) training minutes per week. Of the OTS athletes, 78.6% (11 of 14) also participated in high-intensity functional training full time.

The participants self-collected saliva samples in provided kits at 4 time points: immediately after waking (S1), 30 minutes after waking (S2), at 16:00 hours, and at 23:00 hours. Salivary cortisol concentration was determined by electrochemiluminescence assay.

Separate simple linear regressions tested whether there were systematic relations between cortisol concentrations at S1 and S2 with waking time. Mixed-effects models were used to assess the conditional effect of group (SED, OTS, and ATL) on the change in cortisol over time. Separate models were fit for the awakening samples (S1 and S2) and for the diurnal samples (S1, 16:00 h, and 23:00 h).

The model demonstrated a significant time-by-group interaction for OTS for the cortisol concentrations at S1 and S2 (β = −9.33, P < 0.001) as compared with ATL. The participants in the SED group did not demonstrate a change over time that was statistically different from the ATL group (P = 0.469). The CAR was not associated with self-reported sleep duration or sleep quality.

There was no effect of group assignment on diurnal cortisol slopes (β = 0.02, P = 0.80). The diurnal slope was also not associated with self-reported sleep duration or sleep quality.

The CAR appears to be blunted in athletes diagnosed with OTS, relative to both ATH and SED. Well-controlled intervention studies are needed to verify this effect and to confirm at what point in the development of OTS the CAR metrics begin to deviate from ATL.


Anderson T et al. (2021) Effects of Overtraining Status on the Cortisol Awakening Response—Endocrine and Metabolic Responses on Overtraining Syndrome (EROS-CAR). International Journal of Sports Physiology and Performance 16, 965-973.

Link to Paper:


Share this post with your friends

Share on facebook
Share on twitter
Share on linkedin
Share on pinterest