| dc.description.abstracteng | Background The prevalence of diabetes mellitus is very high among Asian Indians and especially among those residing in urban areas. In the face of this, there is much scope for research in understanding the mechanisms and management of diabetes through indigenous methods such as yoga. Methods 55 diabetics attended a 1-hour daily yoga class for 4 months. A matched-pair control group of 47 diabetics undertook cardiovascular exercises for the same time period. Insulin, cortisol, lipid profile, glucose, and HbA1c were measured at 3 time points during the study. Psychosocial parameters were measured with the HADS and the SF-36 questionnaire Results A significant change from baseline to 4 months was found for salivary cortisol, fasting insulin, and triglycerides in the yoga group but not for the exercise group. However, the exercise group showed a significant drop in salivary cortisol after 2 months, which then increased again to yield no significance between 0 and 4 months. In the yoga and exercise group, the fasting and post-prandial blood glucose values, as well as glycosylated hemoglobin values, largely stayed the same, with no significant change. A significant improvement from 0 to 4 months was seen in the both groups for several HADS and SF-36 psychosocial variables. However, the exercise group's improvement in these variables was more marked. The yoga group’s fasting insulin values decreased while the exercise group’s values increased between 2 and 4 months dramatically. The cortisol values decreased overall in both the yoga and the exercise groups. Conclusion:Insulin levels in the yoga group dropped significantly with corresponding steady blood glucose and HbA1c values. The exercise group's insulin values increased dramatically also with steady blood glucose and HbA1c values. It seems that the yoga group was able to keep steady blood glucose values with significantly less insulin, indicating improved insulin sensitivity. The exercise group seems to have benefitted more on psychosocial well-being than the yoga group, although both showed significant improvement. However, the non-randomised design of our study and marked baseline differences between both groups limit the generalisabilty of our findings and call for future randomized trials of yoga and exercise or their combination in diabetic patients. | de |