Lin 2015.

Study characteristics
Methods	Study design: randomised controlled trial Study grouping: parallel group
Participants	Baseline characteristics Yoga Age in years (mean ± SD): 32.1 ± 7.5 Sex (N (% female)): 26 (86.7%) Sample size: 30 Years of experience (mean ± SD): NR Control (no intervention) Age in years (mean ± SD): 29.8 ± 6.9 Sex (N (% female)): 22 (73.3%) Sample size: 30 Years of experience (mean ± SD): NR Overall Age in years (mean ± SD): NR Sex (N (% female)): 48 (80%) Sample size: 60 Years of experience (mean ± SD): NR Included criteria: the inclusion criteria consisted of mental health professionals who were not involved in a formal exercise program and who were willing to participate in this study. Excluded criteria: exclusion criteria included pain due to injuries to shoulders, waist, or lower back, and musculoskeletal diseases such as muscle strains, that made participants unsuitable to participate in this study. Pretreatment: the demographic characteristics of the two groups, including gender, marital status, religious reference, educational status, job title, and age, showed no significant differences. The total scores of pretest stress adaptation between yoga and control groups did not reach statistical significance. The total scores of pretest work‐related stress between the two groups reached statistical significance. Thus, we took the total scores of pretest work‐related stress as a covariate to control for possible confounding. Type of healthcare worker: mental health professionals in a teaching hospital, but 38% non‐medical nor nursing staff Response rate: 80% Compliance rate: 100%
Interventions	Intervention characteristics Yoga Type of the intervention: Intervention type 2 ‐ to focus one’s attention away from the experience of stress Description of the intervention: The yoga class regularly began with slower warm‐up exercises: Abdominal breathing, cooling breath, and bellows breath, followed by forced abdominal breathing, meditation, and bodily stretching positions. The number of sessions: 12 Duration of each session on average: 60 minutes Duration of the entire intervention: 12 weeks Duration of the entire intervention short vs long: Long Intervention deliverer: The fidelity of the intervention was monitored and directed by two qualified teachers. Intervention form: NR Control (no intervention) Type of the intervention: NA Description of the intervention: A free tea time during which they watched television and did not exercise. The number of sessions: 12 Duration of each session on average: 60 minutes Duration of the entire intervention: 12 weeks Duration of the entire intervention short vs long: Long Intervention deliverer: NA Intervention form: NA
Outcomes	Work‐related stress scale Outcome type: ContinuousOutcome Reporting: Fully reported Stress adaptation scale Outcome type: ContinuousOutcome Reporting: Fully reported
Identification	Sponsorship source: Funding for this study was provided by a grant RA12042 from Changhua Show Chwan Memorial Hospital and a grant MOST‐103‐2314‐B‐166‐003 from the Minister of Science in Taiwan Country: Taiwan Setting: A teaching hospital Comments: NR Authors name: Shu‐Hui Yeh Institution: Central Taiwan University of Science and Technology–Nursing Email: yehshuhui@gmail.com ; 107514@ctust.edu.tw Address: No.666 Buzih Road, Beitun District, Taichung City 40601, Taiwan Time period: NR
Notes	Stress adaptation scale (higher is better) included in analysis 2.1
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	This study was a single‐blind, parallel‐arm randomized controlled trial in which the analyzer was unaware of which group was the experimental or control group. The intervention consisted of a series of weekly, 60‐minute yoga classes over a 12‐week period (Figure 1). Those who were assigned to the control group participated in a free tea time during which they watched television and did not exercise. The participants each signed an informed consent prior to enroling in the study. Then, the participants signed the informed consent form and were randomly assigned to yoga or control groups by drawing lots. There were 30 participants each in the yoga and control groups. It was expected that the two groups were homogeneous through drawing lots of random allocation.
Allocation concealment (selection bias)	High risk	Then, the participants signed the informed consent form and were randomly assigned to yoga or control groups by drawing lots. There were 30 participants each in the yoga and control groups. It was expected that the two groups were homogeneous through drawing lots of random allocation.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants not blinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Participants were not blinded whereas outcomes are self‐reported.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No loss to follow‐up.
Selective reporting (reporting bias)	Unclear risk	No trial registration, nor did we find one online.
Other bias	Unclear risk	Statistically significant baseline differences on stress.