• athlete;
  • fatigue;
  • overtraining syndrome;
  • cortisol;
  • recovery

The objective of this paper is to measure the clinical course (months) in young athletes with persistent fatigue and to identify any covariates affecting the duration of recovery. This was a prospective longitudinal study of 68 athletes; 87% were elite (42 males, 26 females), aged 20.5±3.74 years (SD), who presented with the symptom of persistent fatigue. The collective duration to full clinical recovery was estimated using Kaplan–Meier product–limit curves, and covariates associated with prolonging recovery were identified from Cox proportional hazard models. The median recovery was 5 months (range 1–60 months). The range of presenting symptom duration was 0.5–36 months. The covariates identified were an increased duration of presenting symptoms [hazard ratio (HR), 1.06; 95% confidence interval (CI), 1.02–1.12; P=0.005] and the response of serum cortisol concentration to a standard exercise challenge (HR, 1.92; 95% CI, 1.09–3.38; P=0.03). Delay in recovery was not associated with categories of fatigue that included medical, training-related diagnoses, or other causes. In conclusion, the fatigued athlete represents a significant clinical problem with a median recovery of 5 months, whose collective clinical course to recovery can be estimated by Kaplan–Meier curves and appears to be a continuum.