The present study demonstrated that long-term yoga practitioners have lower mental disturbance, tension-anxiety, anger-hostility, and fatigue scores in the POMS test in comparison to non-experienced participants, although there were no significant differences in the levels of urinary stress-related markers. This is the first study to demonstrate that long-term yoga practitioners have a better mental state than healthy participants who do not engage in yoga. These findings suggest that ongoing yoga training reduces the level of mental disturbance, anxiety, anger, and fatigue not only over the short- or intermediate-term, but also over a long term.
Yoga was found to be as effective as relaxation in reducing anxiety [
5]. A systematic review of the use of yoga to treat anxiety suggested that yoga may be beneficial for different anxiety disorders [
4]. Waelde et al. indicated that an intervention consisting of a six-session yoga program reduced the anxiety of dementia caregivers who did not have an anxiety disorder [
36]. In addition, a week of yoga practice prevented an increase in anxiety among flood survivors [
37]. Moreover, significant reductions were shown for anger and anxiety symptoms in participants who had been diagnosed with unipolar major depression in partial remission and who completed 20 yoga sessions [
9]. These reports and the present results suggest that the anxiolytic and relaxational effects of yoga may continue for an extended period.
Regarding the level of depression in the current study, there was no significant difference between the depression score in the long-term yoga group and that in the control group, which was unexpected. Previous studies of yoga for depression suggest that yoga may have beneficial effects not only among depressive patients [
9], but also non-patients with elevated symptoms of depression [
38,
39]. However, the current results indicated that the mean depressive scores of both the long-term yoga group and the control group were lower than 11, which is equivalent to 50 based on T-score conversion (the T-score table is attached to the Japanese version of the POMS). Therefore, it may be difficult for non-depressive healthy participants to reduce their depressive score in the POMS. Like the mean depressive scores, the mean confusion scores in both groups were lower than 8.5, which is equivalent to 50 based on the T-score conversion. Therefore, there was limited potential for there to be any change in the confusion score in the POMS.
On the other hand, we could not find a "dose" effect for yoga with regard to the mental disturbance, tension-anxiety, anger-hostility, and fatigue scores on the POMS in the long-term yoga group. The mean scores in the long-term yoga group were lower than those in the control group, but the scores in the control group were lower than the equivalent values of 50 based on the T-score conversion, suggesting the presence of a floor effect. In other words, it is speculated that the scores of mental disturbance, tension-anxiety, anger-hostility, and fatigue on the POMS may not decrease after a certain point within 2 years, even if practitioners continue to exercise yoga. A longitudinal study is needed to verify this speculation.
As for urinary 8-OHdG, previous studies have shown that a depressive state affects female 8-OHdG levels in leukocytes [
40,
41]. Forlenza and Miller also reported that severely depressed individuals had increased serum levels of 8-OHdG in comparison to healthy individuals [
42]. In the present study, the urinary 8-OHdG level in the long-term yoga group tended to be lower than that in the control group, although there was no significant difference between the depression score of the POMS in the long-term yoga group and that in the control group. Psychological distress, not depression, may be important to increasing the concentration of urinary 8-OHdG. Further studies are needed to clarify the relationship between 8-OHdG and psychological distress.
Concerning urinary biopyrrin, several studies have indicated that the urinary excretion of biopyrrins is higher in patients with psychiatric disorders, such as schizophrenia [
31,
32] and it increases due to psychological stress [
33]. However, the current results showed no significant difference between the urinary biopyrrin levels of the long-term yoga group and that of the control group. This may be because it is difficult for healthy people to reduce their concentration of urinary biopyrrin by yoga training because it would appear that their urinary excretion of biopyrrins, in healthy people, is already stable at a lower concentration.
Cortisol is an accepted objective stress-related biological marker, because dysregulation of the level of cortisol is related to pathologies associated with stress-related symptoms, such as anxiety, depression, and negative affect [
43-
45]. The decrease in cortisol soon after yoga practice shown by West et al. [
46] was similar to that seen in many types of interventions to reduce stress, such as meditation [
22,
47,
48] and progressive muscle relaxation [
49]. On the other hand, Vera et al. showed that a long-term yoga group displayed higher baseline cortisol levels than control participants [
25]. This result was inconsistent with the present result which showed that the baseline cortisol level of the long-term yoga group was not significantly different from that of the control group. As both Vera et al.'s study and our study are cross-sectional studies, further research is needed to conclusively determine how baseline cortisol levels change over time in long-term yoga practitioners.
There are several limitations in this study. First, a relatively small number of participants were examined in this study. However, we were still able to demonstrate significant differences in some POMS scores because we properly estimated the sample size. Second, the POMS scores, some of which seemed to be influenced by the menstrual cycle, and stress-related biochemical indices were assessed only once throughout the study. Third, it was impossible to discriminate whether the practice of yoga improved the participant's mental state, or whether practitioners who had a good mental state continued to practice yoga, because this study was a cross-sectional study. Although the results must be interpreted with caution, it is possible to conclude that the improvement of mental health with yoga may continue for an extended period, because there is ample evidence that short-term or intermediate-term practice of yoga improves the participants' mental status. The fourth limitation is that the participants were only females. The long-term practice of yoga may have little effect on males. However, mood improvements by yoga are probably not related to gender in healthy participants, because Lavey et al. demonstrated that mood improvement with yoga practice was not related to gender in psychiatric inpatients [
19]. Fifth, the comparison of baseline demographics is limited. We did not measure the current psychological stress levels except by using the POMS, or current exposure to situations likely to cause increased or decreased stress, such as lifestyle, work situation, any other forms of physical exercise or relaxation techniques. Also, we did not confirm whether the total hours of yoga exercise in the long-term yoga group were true, because this information was self-reported, and whether participants had actually avoided vigorous exercise and heavy psychological stress for 24 hours prior to testing. The sixth limitation is that we defined more than 2 years of experience as long-term in this study. Further research is needed to elucidate the far longer-term effects of yoga practice. Finally, the current study measured urinary biopyrrin and baseline cortisol concentrations as stress-related biochemical indices, which are unsuitable or insufficient to assess the mental state of healthy individuals. A multifaceted approach combining the assessment of more suitable stress-related biological markers such as the immune, endocrine, or autonomic nervous systems, and brain function, are important to more accurately assess the mental state of healthy individuals.
Despite these limitations, the present findings suggest that long-term yoga training can reduce the score related to mental health indicators such as self-rated anxiety, anger, and fatigue. Long-term yoga training may affect mental well-being and be useful to prevent mental disorders over an extended period. Further studies that measure self-rated mental states, stress-related biomarkers of the immune, endocrine, autonomic nervous systems, and brain function simultaneously are needed to verify the beneficial effects of yoga and to elucidate the mechanism(s) responsible for the benefits that are associated with yoga.