Laughter Yoga Intervention for People With Major Depressive Disorder

72 community dwelling people with co-morbid symptoms of depression, anxiety and stress who are diagnosed and treated for major depressive disorder (MDD) will be recruited into the study from the Community Psychiatric Service (CPS) of Castle Peak Hospital. After baseline measurements, 36 participants will be randomly allocated into the Laughter Yoga group (LY) or Treatment-as-usual group (TAU). The LY intervention will be delivered by a certified lead LY trainer and monitored by three Co-Is and/or Research Assistants. The LY participants will be asked to attend a total of eight 45-minutes sessions of group-based laughter yoga over a 4 week period. TAU participants will receive their usual routine community mental health care. All data collection will be conducted by a trained research assistant. Demographic data and all the relevant clinical/treatment data will be collected at baseline, whereas outcome data will be collected at three time points: Baseline, after the 4 weeks' intervention period, and at 3 months after finishing the intervention. The level of depression, anxiety and stress will be the primary outcome of the study. They will be measured by the Chinese version of the Depression Anxiety Stress Scale (DASS-21) developed by Lovibond and Lovibond (1995) . The secondary outcome will be quality of life (physical and mental health related), which will be assessed with the short Form 12 item (version 2) Health Survey (SF12v2) (Lam et al., 2014). The Chinese language version of the client satisfaction questionnaire (CSQ-8)(Attkisson and Zwick, 1982; Attkisson, 2012) will be used to measure patient's view about their satisfaction with the LY intervention immediately post intervention. To explore patient's experience on the LY intervention, a sample of 12 patients who received the LY intervention will be invited to a short individual interview with the research assistant at three months' follow-up. Trial feasibility data, such as refusal rate, response rate, drop out, attendance rate will also be recorded throughout the study. Descriptive statistics will be used to contextualize the demographic and clinical characteristics of the study population at baseline. The baseline variables will be compared between the groups so that potential confounders will be identified and where necessary adjusted for. For pre-test and post-test comparisons between the two groups, ANOVA test will be adopted if satisfactory normality of data is established. Otherwise, Kruskal-Wallis test will be used.