Table 1: Summary of clinical studies of mindfulness meditation in diabetes.

Author, year	Study design	Results
Rosenzweig et al., 200734	Prospective observational study, 14 adults with T2D participating in MBSR program over 1 month.	11/14 patients completed the intervention. HbA1c reduced by -0.48% and mean arterial pressure by -6 mmHg, no change in weight at 1 month. Significant improvement in depression, anxiety and general psychological distress.
Kopf and Hartmann et al., HEIDIS trial, 201435,36	In an open-label, randomised study, 110 patients with T2D and early diabetic kidney disease (microalbuminuria) were randomised to 8-week MBSR (n=53) or standard care (n=57). Participants met once weekly for 8 weeks followed by a booster session after 6 months and long-term effects were assessed at 1, 2 and 3 years. Patient health questionnaire PHQ-9 was used to assess depression score and stress scores. Subjective health was assessed using short form health survey (SF-12).	All patients were assessed after 1 year. Albumin-creatinine ratio decreased from 44 to 39 mg/g creatinine at 1 year in intervention group, while it increased from 47 to 59 mg/g creatinine in control group (p=0.05). There was no difference in the depression score in the immediate post-intervention period (PHQ-9 depression score difference of adjusted means (d), d=0.03, p=0.9) but a significant reduction was seen at 1-year assessment (PHQ-9 depression score d=-0.71, p=0.007) in the immediate post-intervention period and at 1 year. Subjective health – a significant improvement was noted in the SF-12 mental composite score (d=-0.54, p=0.033) but not in physical composite score at 1 year. PHQ-9 Stress Score was not affected. There was a trend towards reduction of diastolic blood pressure immediate post-intervention and this became statistically significant at 1 year (d=0.78, p=0.004). However, HbA1c and systolic blood pressure were not significantly affected. Sensitivity analyses that included imputed data yielded similar results. Reduction in stress levels, reduced catecholamine levels, improved mean 24-hour arterial and maximum systolic blood pressure and intima media thickness at 1 year in intervention group. Intervention group had reduced catecholamine levels (p<0.05), improved 24-hour mean arterial pressure (p<0.05) and maximal systolic blood pressure (p<0.01) at 1 year. MBSR group also had a reduction in intima media thickness (p<0.01). 89 patients (42 in control group and 47 in MBSR group) were available for 3 years’ follow-up. The beneficial effects were not sustained at 2- and 3-year follow-up.
Rungreangkulkij et al., 201137	A quasi-experimental study using intervention group and matched controls. 62 patients with T2D and with depressive symptoms were assigned to Buddhist group therapy and meditation practices (n=32) versus control group which received standard treatment only (n=32). Intervention group underwent weekly 2-hour sessions for 6 weeks plus home meditation practices.	After 6 months, there was improvement in depression score in the meditation group. In the intention-to-treat analysis, relative risk of depressive symptoms was 6.5 (95% CI 1.4, 30.60) between experimental and control groups. Qualitative data reported greater acceptance of current living condition in the meditation group.
Keyworth et al., 201438	Pilot study. 40 adults with diabetes and coronary heart disease underwent a 6-week meditation and mindfulness intervention program. The Penn State Worry Questionnaire (PSWQ) was used to assess pathological worry and the White Bear Suppression Inventory (WBSI) was used to measure thought suppression. Qualitative effects on acceptability, feasibility and user experience were assessed using a focus group (n=11) and in-depth interviews (n=16).	Intervention was highly acceptable (>90% completed ≥5 sessions). There was significant improvement in PSWQ before and after intervention: mean = 48.55 ± 14.14 and 42.84 ± 12.05, respectively (p<0.001). Similar improvements were noted in WBSI scores before and after intervention: mean = 49.53 ± 12.44 and 46.34 ± 11.47 (p=0.015). In the qualitative assessment, meditation was associated with improved sleep, greater relaxation, better acceptance of illness and illness experience.
Teixeira et al., 201039	20 participants with painful diabetic neuropathy, mindfulness meditation versus standard care.	No difference in painful neuropathy. But improvement in pain perception and symptom-related quality of life.
Gainey et al., 201640	23 patients with T2D were randomised to traditional walking exercise or Buddhism-based walking meditation exercise. Both groups followed a 12-week exercise program that involved walking on a treadmill at exercise intensity of 50–70% maximum heart rate, for 30 min/session, three times/week. Meditation group meditated on the sound ‘Budd-Dha’ while exercising.	Decrease in fasting blood glucose and maximal oxygen consumption in both groups at 12 weeks (p<0.05). Significant decrease in HbA1c, systolic and diastolic blood pressure seen only in walking meditation group. Flow-mediated vasodilatation improved in both groups (p<0.05), but arterial stiffness improved only in walking meditation group. Blood cortisol levels reduced (p<0.05) in walking meditation group.
Chaiopanont et al., 200841	Quasi-experimental study that included 50 patients with T2D (11 males, 39 females). All participants were trained to practice sitting breathing meditation after breakfast for 2 weeks.	Significant reduction was seen in postprandial plasma glucose in the second week visit (-19.26 ± 30.99 mg/dl, p<0.001) and third week visit (-17.64 ± 25.48 mg/dl, p<0.001). Systolic blood pressure was also significantly reduced in second visit (-6.49 ± 11.77 mgHg, p<0.001) and diastolic blood pressure was reduced in the third visit (-3.04 ± 9.79, p<0.05).
Jung et al., 201542	Cluster randomised trial design. 56 adults with diabetes were randomised to patient education, MBSR or walking exercise groups over 8 weeks. Diabetes-related distress was assessed by Diabetes Distress Scale. Psychological response to stress was measured by Perceived Stress Response Inventory. Other outcomes included fasting glucose, serum cortisol, serum plasminogen activator inhibitor-1 (PAI-1) and tissue plasminogen activator (t-PA).	No difference between the three groups in relation to diabetes distress, psychological response to stress, glycaemic control or vascular inflammation (t-PA). MBSR group and walking exercise groups had statistically significant reductions in plasma cortisol (-2.53 ± 3.96 μg/dl and -4.51 ± 5.69 pg/dl) from baseline. Significant reductions in PAI-1 levels were seen in all three intervention groups (-8.94 ± 10.84 in MBSR group, -11.74 ± 16-07 in walking exercise group and -15.73 ± 10.86 ng/ml in patient education group).
Rogers et al., 201743	Meta-analysis of 15 randomised controlled trials or prospective cohort studies measuring outcomes of mindfulness-based interventions in 560 individuals who were overweight or obese.	Average weight loss 4.2 kg. Overall effect size was large for improvement in eating behaviours (likelihood ratio as calcuated in G-test [g], g=1.08) and medium for eating attitudes (g=0.57), depression (g=0.64) and anxiety (g=0.62). Effect on body mass index (g=0.47) and metacognition (g=0.38) was small. Similar benefits persisted when only higher quality randomised controlled trials were included.
Abbott et al., 201444	Meta-analysis of eight randomised controlled trials, effect of MBSR or MBCT, 578 participants with prehypertension/ hypertension (three trials), diabetes (two trials), heart disease (two trials) and stroke (one trial).	Significant reduction in stress (-0.36, 95% CI -0.67 to -0.09; p=0.01), depression (-0.35; 95% CI -0.53 to -0.16; p=0.003) and anxiety (-0.50; 95% CI -0.70 to -0.29; p<0.001). Mixed effects on physical outcomes (blood pressure, albuminuria, stress hormones).
Anderson et al., 200845	Meta-analysis of nine randomised controlled trials comparing blood pressure response to transcendental meditation technique with control group. Three studies rates as high quality and three as acceptable quality. Effect of transcendental meditation on blood pressure assessed by random-effects meta-analysis model.	Transcendental meditation was associated with significant reduction in systolic blood pressure (-4.7 mmHg, 95% CI -7.4 to -1.9) and diastolic blood pressure (-3.2 mmHg, 95% CI -5.4 to -1.3). Subgroup analysis of high quality studies showed similar reductions.

CI = confidence interval; d = difference of adjusted means; g = likelihood ratio as calculated in G-test; HbA1c = glycated haemoglobin A1c; MBCT = mindfulness-based cognitive therapy; MBSR = Mindfulness-based stress reduction; PAI-1 = serum plasminogen activator inhibitor-1; PHQ-9 = Patient Health Questionnaire 9; T2D = type 2 diabetes.