The diaries completed by participants indicated that they had performed their home practice a mean of 4.04 (SD: 1.17) days per week, a one-way ANOVA finding no statistical differences between the groups (F = 1.651, ns).
There were no baseline differences in the scores of the groups for the DASS, the PSQI, the SVS, the MAAS, the TAS, the VAS, the ISQ, or salivary cortisol, as disclosed by one-way ANOVAs (F ≤ 2.447, P ≥ 0.104). It was thus considered reasonable to compare the mean changes of the groups for each of these measures. There was a tendency though for the baseline scores of the 2-Back to statistically differ for the different groups (t = 3.095, P = 0.06), where a Post hoc Tukey test found a trend for the Iyengar participants to have significantly lower scores than the Mindfulness participants (Mean Difference: −17.444, P = 0.072), although the differences between the Iyengar and BWV groups (Mean Difference: −15.944, ns) and the BWV and Mindfulness groups (Mean Difference: −1.5, ns) were nonsignificant.
3.1. The Depression, Anxiety, and Stress Scale [
7]
Considering first the mood questionnaire used to select participants, the means and standard deviations for the sum total of DASS items, Total DASS, and for the subscales Depression, Anxiety, and Stress, are shown in . Two Mindfulness participants were excluded from the DASS analysis—one participant whom had a Pre Anxiety score that was 3.2 SDs above the sample mean, and one participant whom had a Pre Depression score 2.96 SDs above the sample mean.
| Table 1Means (SD)* of total dass, depression, anxiety, and stress**. |
Mood improved overall the whole sample, as disclosed by a mixed ANOVA which found a significant main effect of Session for the sum total of DASS items, Total DASS (
F(1, 28) = 19.617,
P < 0.001). The mean (SD) changes in the Total group scores were as follows: BWV decrease = 5.67 (6.21); Iyengar decrease = 6.44 (7.07); Mindfulness decrease =


3.5 (6.39)). Separate ANOVAs on the subscales indicated that the overall improvement was largely due to reductions in both Stress (
F(1, 28) = 23.596,
P < 0.001) and Depression (
F(1, 28) = 7.148,
P = 0.012), for reductions in Anxiety were not significant (
F(1, 28) = 2.223, ns). Subscale results are shown in .
While there were no Group x Session interactions, planned contrasts within groups with paired t-tests found for the BWV group a decrease in Total scores (t(1, 11) = 3.159, P = 0.009) and in the subscales of Depression (t(1, 11) = 2.24, P = 0.047) and Stress (t(1, 11) = 2.339, P = 0.039), although the change in Anxiety was nonsignficant (t(1, 11) = 0.904, ns). In the Iyengar group there was a significant decrease in the Total score (t(1, 8) = 2.733, P = 0.026) and in Stress (t(1, 8) = 2.596, P = 0.032) but not in the other subscales (Depression: t(1, 8) = 1.828, ns; Anxiety: t(1, 8) = 1.931, ns), while in the Mindfulness group there was a reduction in Stress (t(1, 9) = 3.932, P = 0.003), and no other effects (Total DASS: t(1, 9) = 1.759, ns; Depression: t(1, 9) = 0.402, ns; Anxiety: t(1, 9) = −0.874, ns).
In summary, there were improvements in Total scores following BWV and Iyengar, improvements in Stress for all groups, for Depression with BWV, while none of the groups improved in Anxiety.
3.2. The Pittsburg Sleep Quality Index [
12]
shows the mean Pre and Post scores for the sum total of PSQI items, Global Sleep, and for each of the PSQI subscales.
| Table 7 Mean (SD)* of Global Sleep and the PSQI subscales**. |
The sample as a whole showed no significant change in Global Sleep (Session:
F(1, 30) = 1.84, ns). Considering the groups separately with paired
t-tests also disclosed no significant changes (BWV:

t(1, 11) = 1.503, ns; Iyengar:
t(1, 8) = 0.223, ns; Mindfulness:
t(1, 11) = 0.639, ns).
Significant changes were found for certain PSQI subscales however. The mean changes of the three groups for each of the subscales are shown in , except for Sleep Disturbance, as no participant showed any change in this subscale. Sleep Latency was found to significantly decrease in the whole sample (Session: F(1, 30) = 5.315, P = 0.028). Paired t-tests found a significant decrease in the BWV group (t(1, 11) = 2.321, P = 0.04), while neither of the other groups showed significant changes (Iyengar: t(1, 8) = 1.512, ns; Mindfulness: t(1, 11) = 0.692, ns).
There was also a marked improvement in the subscale Daytime Dysfunction in the whole sample, where the main effect of Session was highly significant (F(1, 30) = 9.19, P = 0.005). Both the BWV and Mindfulness groups were found by paired t-tests to have significantly lower mean scores following the intervention (BWV: t(1, 11) = 2.569, P = 0.026; Mindfulness: t(1, 11) = 2.569, P = 0.026), although there was no notable change in the Iyengar group (t(1, 8) = 0.686, ns).
No main effects of Session were found for the other PSQI subscales however (Sleep Quality: F(1, 30) = 1.446, ns; Asleep-to-bed: F(1, 30) = 1.348; ns, Sleep Disturbance: F(1, 30) = 0, ns; Sleep Duration: F(1, 30) = 0.056, ns; Sleep Efficiency: F(1, 30) = 1.292, ns). There were also no Session x Group effects found for the total PSQI score or for any of the subscales (F(2, 30) ≤ 1.078, P ≥ 0.353).
In summary, the BWV group showed the greater benefit in time to fall asleep and a benefit shared with Mindfulness with residual daytime sleepiness.
3.3. The Subjective Vitality Scale [
13]
The mean Pre and Post scores for the sum total of SVS items, Total SVS, are shown in . There was an increase in vitality in the sample overall, as can be seen by the change in the Total SVS score in (F(1, 30) = 7.208, P = 0.012). Paired t-tests disclosed marked improvements in the BWV (t(1, 11) = −2.931, P = 0.014) and Iyengar (t(1, 8) = −2.881, P = 0.02), but little change in the Mindfulness group (t(1, 11) = −0.343, ns).
3.4. The Mindfulness Attention Awareness Scale [
14]
shows the mean Pre and Post scores of the three groups and the sample overall for the mean of MAAS items, Mean MAAS.
There was an increase in Mindfulness for sample as a whole (F(1, 30) = −2.39, P = 0.036). Paired t-tests indicated significant effects for the Mindfulness (t(1, 11) = −2.688, P = 0.028) and Iyengar (t(1, 8) = −2.688, P = 0.028) groups and an effect approaching significance in the BWV group (t(1, 11) = −2.082, P = 0.062). These increases can be seen in .
As with the previous scales, though, there were no statistical differences in the comparative changes of the three groups, and the Group x Session interaction was nonsignificant (F(2, 30) = 0.075, ns).
3.5. The Tellegen Absorption Scale [
15]
The mean and standard deviation Pre and Post scores for the sum total of TAS items, Total TAS, are shown in .
There was a marginal increase in absorption in the sample overall, marked by a slight tendency towards a main effect of Session (F(1, 30) = 3.254, P = 0.081). As can be seen in the Mindfulness group accounted for most of the improvement (t(1, 11) = −4.322, P = 0.001; BWV: t(1, 11) = −0.896, ns; Iyengar: t(1, 8) = 0.833, ns), supported by a significant Session x Group interaction (F(2, 30) = 4.992, P = 0.013).
3.6. The Visual Analogue Scale of BWV Benefits
The mean and standard deviations of the Pre and Post scores for each of the items of the VAS are shown in .
| Table 8 Mean (SD) of each VAS item (%). |
In accordance with the increase in vitality seen with the SVS, the VAS also indicated that vitality increased in the whole sample, disclosing a main effect of Session for the item “I have had high vitality levels” (F(1, 30) = 6.764, P = 0.014). Separate paired t-tests indicated that this increase approached significance in both the BWV (t(1, 11) = −1.985, P = 0.078, mean increase: 15.3% (SD: 26.6%)) and Mindfulness (t(1, 11) = −2.018, P = 0.069, mean increase: 14.5% (SD: 28.7%)) groups, but not in the Iyengar group (t(1, 8) = −0.989, ns, mean increase: 9.4% (SD: 29.6%)).
For the item “I have felt tense” there was tendency towards a reduction overall (Session: F(1, 30) = 3.398, P = 0.061). There was a substantive decrease in the BWV group (t(1, 11) = 3.427, P = 0.006, mean decrease: 31.3% (SD: 31.7%)) not seen in the other groups (Iyengar: t(1, 8) = −0.106, ns, mean increase: 1.67% (SD: 47.4%); Mindfulness: t(1, 11) = −0.701, ns, mean decrease: 5.83% (SD: 44.5%)).
There was a general increase for the item “I have felt calm” (F(1, 30) = 4.449, P = 0.043), attributable to the Mindfulness group (t(1, 11) = −2.204, P = 0.05, mean increase: 18.8% (SD: 44.5%)) (BWV: t(1, 11) = −1.733, ns, mean increase: 15.3% (SD: 30.7%); Iyengar: t(1, 8) = −0.171, ns, mean increase: 2.22% (SD: 38.9%)).
The sample as a whole also tended to feel more grounded following the intervention, as evinced by the item “I have felt more grounded” (F(1, 30) = 3.882, P = 0.058), but with paired t-tests again disclosing a significant effect only for the Mindfulness group (t(1, 11) = −2.62, P = 0.024, mean increase: 17.9% (SD: 23.7%)) (BWV: t(1, 11) = −0.599, ns, mean increase: 7.67% (SD: 44.3%); Iyengar: t(1, 8) = −0.947, ns, mean increase: 14.9% (SD: 47.2%)).
Separate Mixed ANOVAs for each of the other VAS items found a significant main effect of Session for the item “I have had good circulation” (F(1, 30) = 4.192, P = 0.049) and a tendency towards a main effect of Session for the item “My head has felt clear and not foggy” (F(1, 30) = 3.659, P = 0.065), where for both items an increase was seen in the sample as a whole with no group differences. There were no main effects of Session for the remaining 18 VAS items (F(1, 30) ≤ 2.927, P ≥ 0.097). In addition, no Group x Session interactions were found for any subscale (F(2, 30) ≤ 2.117, P ≥ 0.138).
In summary, for the sample as a whole there were improvements in feelings of vitality, tension, calmness, groundedness, circulation, and clear-headedness, with advantages following BWV for vitality and tension, and following Mindfulness for vitality, calmness, and groundedness.
3.7. The Illness Symptoms Questionnaire [
8,
9] and the Dual-Back Task [
10]
The means and standard deviations for the Pre and Post scores for the sum total of Illness Symptoms Questionnaire items, Total ISQ, and for the Dual Back are shown in .
| Table 5 Mean (SD)* of the Total ISQ and the 2-Back**. |
The whole sample showed no change in either illness symptoms or working memory, and the main effects of Session for Total Illness and 2-Back were both nonsignificant (Total Illness:
F(1, 30) = 0.456, ns; 2-Back:
F(1, 30) = 0.072, ns), nor was there a Session x Group interaction for either of these variables (Total Illness:

F(2, 30) = 0.047, ns; 2-Back:
F(2, 30) = 0.196, ns).
3.8. Salivary Cortisol
shows the mean and standard deviation Pre and Post Salivary cortisol scores of the three groups and the sample overall. There was a slight increase in salivary cortisol in the whole sample, disclosed by a tendency towards a main effect of Session (
F(1, 30) = 3.373,
P = 0.073). However, paired
t-tests indicated that the increase was not significant for any individual group (BWV:
t(1, 11) = −1.022, ns; Iyengar:
t(1, 8) = −1.635, ns; Mindfulness:

t(1, 11) = −0.749, ns). The Session x Group interaction was also nonsignificant (
F(2, 30) = 0.093, ns).
| Table 6Mean (SD)* salivary cortisol. |