What diagnostic tools are available to detect overtraining?
Overtraining and overreaching are very difficult to diagnose in athletes let alone predict before they happen. Currently, the focus of many coaches and researchers for monitoring the early signs of overreaching in athletes seems to be on heart rate variability (HRV). However, there are other options.
This fascinating study tested the ability of four other diagnostic tools for assessing non-functional overreaching in a small number of athletes.
The study: Different Diagnostic Tools in Nonfunctional Overreaching, by Nederhof, Zwerver, Brink, Meeusen and Lemmink, in International Journal of Sports Medicine, 2008
How can overtraining be defined?
Athletic training involves overload followed by recovery. Such overload can lead to feelings of fatigue and acute decreases in performance either directly after a single intense training session, or following an intense period of training. However, in the normal course of training and recovery, this overload then leads to a positive training response, adaptation and a consequent improvement in performance.
It is thought that if the balance between overload and recovery is not managed properly, however, that a positive training response does not occur and performance is not enhanced. The investigation of this undesirable response is the focus of researchers studying overtraining. Such researchers have produced consensus statements and guidance documents (Meeusen, 2006, Kreher, 2012, Meeusen, 2013) in which the following definitions have been proposed:
- Overtraining is a process of intensified training with possible outcomes of functional (or short-term) overreaching, non-functional (or extreme) overreaching, or overtraining syndrome.
- Functional overreaching is a process of increased training leading to a temporary performance decrement and with improved performance following rest.
- Non-functional overreaching is a process of intense training leading to a longer performance decrement but with full recovery after rest, accompanied by certain psychological and/or neuro-endocrinological symptoms.
- Overtraining Syndrome – a condition consistent with non-functional overreaching but with (1) a longer duration of performance reduction (>2 months), (2) more severe symptomatology and maladapted physiology (psychological, neurological, endocrinological, immunological), and (3) an additional stressor not explained by other disease.
As noted above, since other conditions can cause similar cases of unexplained underperformance, in order to reach a true diagnosis of overtraining syndrome, it is necessary to exclude the presence of non-communicable diseases (e.g. disorders involving the thyroid or adrenal gland, diabetes, iron deficiency and anemia), infectious diseases (e.g. myocarditis, hepatitis and glandular fever) and other major disorders or feeding behaviors (e.g. anorexia nervosa and bulimia).
What are the characteristics of individuals with overtraining syndrome?
Many studies have assessed the characteristics of overtraining syndrome but not all measure the same parameters and those that do measure the same variables do not always find the same results. A recent meta-analysis (Wyatt, 2013) was performed to provide summary quantitative findings of blood biomarkers and physiological and psychological measurements that have been associated with overtraining syndrome. Their findings were as follows:
- Blood biomarkers – the reviewers reported on studies that assessed glutamine, glutamate, cortisol, IL-6, testosterone, cholesterol, glucose, leptin, hematocrit, hemoglobin, norepinephrine, epinephrine, and creatine kinase. They found that only the effect sizes for glutamine, glutamate, cortisol, IL-6, and glucose were large (in fact, the effects sizes for glutamine, glutamate and IL-6 were extremely large), while the effect sizes for testosterone and creatine kinase were moderate (all reductions apart from creatine kinase).
- Physiological measurements – the reviewers reported on studies that assessed resting heart rate, resting systolic blood pressure and resting diastolic blood pressure but not heart rate variability. They noted that the studies produced conflicting results, although the effect sizes for both resting heart rate and systolic blood pressure were large and indicated that these variables are reduced in the overtrained athlete. However, this finding conflicts with the conclusions of previous reviews.
- Psychological measurements – the reviewers found that mood states for tension, fatigue, confusion, vigor, anger and depression were often considerably altered in the studies but with no obvious trend for direction. Such variances may reflect a highly individual nature of the psychological response to overtraining. However, they did observe a trend for impaired sleep patterns, increased wakefulness, decrements and stability in sleep quality and increased levels of stress.
In summary, it is likely that athletes with overtraining syndrome would likely display increased self-perception of stress, reduced sleep quantity and quality, disturbances in sleep, self-perception and mood, immune suppression and altered sympathetic nervous system activity.
For some reason, the reviewers in this systematic review did not consider heart rate variability (HRV). Just as for resting heart rate, studies have pointed towards both increases and decreases in various measures of beat-to-beat variability in overtrained athletes. However, researchers have proposed that this reflects different pathways to an overtrained state.
Makivic (2013) performed a systematic review and concluded that overtraining associated with increased exercise volume leads to sympathetic predominance, while overtraining associated with increased exercise intensity leads to parasympathetic predominance. Additionally, they note that non-functional overreaching phases may also be characterized by sympathetic dominance. Since modern smartphones can now be used in conjunction with heart rate monitors to measure HRV very easily and accurately, this makes HRV a very attractive tool for coaches.
What causes overtraining syndrome?
The exact causes of overtraining syndrome are unclear, mainly because it is very difficult to study, for two main reasons. Firstly, it is not ethically possible to induce overtraining in athletes and therefore any study must by definition be retrospective. Secondly, since overtraining can currently only be diagnosed following an extensive period of observation and after ruling out many other possible causes, and, athletes who are persistently underperforming after many months sometimes choose to retire, making it difficult to assess whether non-functional overreaching or overtraining was in fact attained.
Nevertheless, there are several mechanisms that have support among various groups of researchers, which Kreher (2012) has summarized under the following headings:
- Glycogen hypothesis – in this model, it is believed that reduced glycogen causes fatigue and consequently decreased performance. However, Snyder (1995) found that it was possible for athletes to become overtrained despite normal levels of muscle glycogen, making this hypothesis very hard to support.
- Central fatigue hypothesis – in this model, it was original proposed that increased tryptophan uptake in the brain leads to increased levels of the neurotransmitter serotonin (5-HT) and that this produces adverse mood symptoms. Meeusen (2006) has explained the revised version of the hypothesis, in which an increase in the ratio of serotonin to dopamine is thought to bring about feelings of tiredness and lethargy.
- Glutamine hypothesis – in this model, it has been proposed that reduced glutamine causes immune dysfunction and increased susceptibility to infection. However, it is thought that overtraining syndrome can occur without the presence of an infection, making this hypothesis unattractive.
- Oxidative stress hypothesis – in this model, researchers have suggested that excessive oxidative stress leads to muscle damage and fatigue.
- Autonomic nervous system hypothesis – in this model, it is thought that parasympathetic predominance causes various symptoms. However, studies using heart rate variability to measure cardiac autonomic balance have noted increases in sympathetic as well as parasympathetic predominance following increased training loads (e.g. Hynynen, 2006) and it seems that in most cases the effects of intensified training on cardiac autonomic balance can be redressed within one week (e.g. Pichot, 2000).
- Hypothalamic hypothesis – in this model, it has been proposed that dysregulation of the hypothalamic-pituitary-adrenal and/or hypothalamic-pituitary-gonadal axes may cause various symptoms of overtraining syndrome, with obvious targets being cortisol or testosterone. However, studies are inconclusive regarding what happens to these hormones during periods of overtraining (e.g. Lehrmann, 1992, Hooper, 1993, Urhausen, 1998, Mackinnon, 1997, and Uusitalo, 1998), with studies showing either increased, decreased or unchanged levels of adrenal hormones. That being said, post-traumatic stress disorder (PTSD) researchers have been facing the exact same problems with the HPA axis for many years (e.g. see further within this full-text review by Pitman, 2012).
- Cytokine hypothesis – in this model (see further Smith, 2000) it has been proposed that inflammation and cytokine release causes most of the known effects and symptoms of overtraining syndrome. The strength of this model is that it can be related to most of the other proposed mechanisms. The main weakness is that few studies have assessed the prevalence of elevated cytokines in overtrained athletes and such studies have not produced positive results (e.g. Halson, 2003).
In summary, there is no currently accepted mechanism by which overtraining occurs. This makes it quite hard to assess whether overtraining is happening, until after the event.
What tools can be used to diagnose overtraining syndrome?
Despite the lack of an understanding regarding the mechanism of overtraining, coaches and athletes are keen to implement measures that may help them avoid it, even if these are only partially tested. Urhausen (2002) listed and reviewed the currently available diagnostic tools that may be helpful, although several studies have since been performed that were not available for that analysis, and Nederhof (2006) has proposed an additional tool in the form of psychomotor performance. Here’s a summary list for those variables that can be tested while resting:
- Resting heart rate
- Heart rate variability (HRV)
- Profile of mood states (POMS)
- Metabolic markers in blood
- Immunological parameters
- Psychomotor performance (i.e. reaction time)
The following additional variables have been proposed as markers of overtraining during exercise:
- Performance reduction
- Metabolic markers in blood
- Heart rate
- Rating of perceived exertion
Many of these markers are useful but none appear to be conclusive. Overall, it is likely that useful indicators of non-functional overreaching or overtraining include altered mood states (using the POMS scale), supressed immune markers, reduced reaction time, a performance reduction, either increased or decreased HRV, and reduced submaximal lactate concentrations during similar levels of exercise.
What did the researchers do?
As noted above, based on the outcomes of studies in overtrained athletes, it is likely that athletes displaying non-functional overreaching or overtraining syndrome will display increased self-perception of stress, reduced sleep quantity and quality, disturbances in sleep, alterations in self-perception and mood, immune suppression and altered sympathetic nervous system activity. In this study, the researchers compared the following tools in respect of their ability to perform an assessment of non-functional overreaching via their ability to measure one or more of those features, as follows:
- The Recovery Stress Questionnaire for Athletes (RESTQ-sport) – this is a structured questionnaire in 19 sections with 4 questions per section and with each question being answered on a 7-point scale that seeks to establish the frequency of experienced stressors and regeneration related activities. It has previously been validated in respect of monitoring athlete reactions to changes in training load.
- The Profile of Mood States (POMS) – this is another structured questionnaire in 5 sections with a total of 32 items assessing depression (8 items), anger (7 items), fatigue (6 items), tension (6 items) and vigor (5 items), with each question being answered on a 5-point scale.
- Reaction time – some researchers (e.g. Nederhof, 2006) have proposed that psychomotor function may be impaired in athletes with non-functional overreaching or overtraining. This could be revealed by reduced or impaired reaction time. And the nice thing is that this is a very simple hard metric to measure multiple times and record.
- Hypothalamic-pituitary-adrenal (HPA) axis function – as noted above, it is thought that in athletes with overtraining syndrome, there is a dysfunctional response of cortisol and adrenocorticotrophic hormone (ACTH) reactions to a specific type of maximal exercise test performed twice in a single day. For each test, cortisol and ACTH concentrations were measured before and after an incremental graded exercise test on a cycle ergometer (starting at 120W and increasing 30W every 3 minutes until exhaustion).
These four tests therefore assessed self-perception of stress, alterations in self-perception and mood, altered sympathetic nervous system activity, and changes in psychomotor performance. For subjects, the researchers obtained access to three female speed skaters, one of whom was a healthy control subject and two of whom were later diagnosed with varying stages of non-functional overreaching (NFO), with one currently non-functionally overreaching and one in recovery from non-functional overreaching. The NFO athlete had refrained from training for 2 weeks and the athlete who was recovering from NFO had ceased training for 12 weeks.
The Recovery Stress Questionnaire for Athletes (RESTQ-sport)
The researchers noted that the control subject displayed low scores on the general and sports-specific stress scales and high levels on the general and sport specific regeneration scales. The athlete with NFO displayed high scores on the general and sport specific stress scales, low levels on the general regeneration scale, and low levels on some of the sport-specific sub-scales. The athlete in recovery from NFO displayed values that were midway between these two extremes.
Profile of mood states (POMS)
The researchers reported that the control subject displayed low scores on the negative mood scales and a high score for vigor. The athlete with NFO showed a high score on the fatigue scale and moderate-to-low scores on the other mood scales, including vigor. The athlete in recovery from NFO showed moderate scores on all scales.
Reaction time task
The researchers found that the athlete in recovery from NFO displayed the fastest reaction times while the athlete who currently with NFO displayed the longest reaction times.
The researchers reported that the control subject displayed a small decrease in cortisol concentrations after both exercise tests. The athlete recovering from NFO displayed a small decrease after the first test but a small increase after the second test. The athlete suffering from NFO showed a small increase after the first test and a large increase after the second test.
What did the researchers conclude?
The researchers concluded that the RESTQ-sport (which measures general and sports specific self-perceived stress), a reaction time test (which measures psychomotor speed) and the cortisol response to a double exercise protocol (which measures the HPA-axis response to exercise) are promising tools for monitoring the presence of non-functional overreaching.
However, they noted that since the POMS test (which measures mood states) was not able to differentiate effectively between the athlete with NFO and the athlete in recovery from NFO, this test was less useful.
What were the limitations?
The main limitation of the study was that there were only three subjects, of which one was a control. Therefore, individual differences could lead to a completely different picture being observed in other athletes.
Another key limitation is that the data gathered were single points and only reflect information gathered during the period in which diagnosis of non-functional overreaching was made. If the researchers had been able to gather data for the same athletes before they became non-functionally overreaching, the POMS test might have been more useful in that it could have demonstrated large swings for the individual athletes.
What are the practical implications?
For coaches and sports medicine physicians
Unexplained performance reductions are the gold-standard measurement for non-functional overreaching or overtraining. Athletes who persistently display reductions in performance despite engaging with prescribed training programs and recovery should be assessed for non-functional overreaching or overtraining.
In arriving at a diagnosis of overtraining, it is necessary to exclude the presence of non-communicable diseases (e.g. disorders involving the thyroid or adrenal gland, diabetes, iron deficiency and anemia), infectious diseases (e.g. myocarditis, hepatitis and glandular fever) and other major disorders or feeding behaviors (e.g. anorexia nervosa and bulimia).
It is likely that athletes with overtraining syndrome will display increased self-perception of stress, reduced sleep quantity and quality, disturbances in sleep, self-perception and mood, immune suppression and altered sympathetic nervous system activity.
Potentially useful early-warning indicators of non-functional overreaching or overtraining include altered mood states (potentially using the POMS scale), supressed immune markers, reduced reaction time, a performance reduction, either increased or decreased HRV, and reduced submaximal lactate concentrations during similar levels of exercise.
The RESTQ-sport (which measures general and sports specific self-perceived stress), a reaction time test (which measures psychomotor speed) and the cortisol response to a double exercise protocol (which measures the HPA-axis response to exercise) are promising tools for monitoring the presence of non-functional overreaching.