Instrumental variable analysis of primary outcomes at 1-year follow-up, with allocated group as an instrument for the implementation variables
| Outcome/implementation variables | N (K) | Coefficient | 95% CI | P value |
| Risk of depression | ||||
| Dose | 5508 (65) | 0.08 | −0.01 to 0.17 | 0.065 |
| Fidelity | 5667 (66) | 0.01 | −0.004 to 0.02 | 0.219 |
| Reach | 5673 (65) | 0.02 | −0.01 to 0.04 | 0.284 |
| Quality | 6139 (73) | 0.12 | −0.07 to 0.31 | 0.199 |
| Practice (postintervention) | 5960 (73) | 0.40 | −0.24 to 1.03 | 0.226 |
| Practice (1 year follow-up) | 6196 (73) | 0.56 | −0.33 to 1.44 | 0.218 |
| Social–emotional–behavioural function | ||||
| Dose | 5423 (65) | 0.05 | −0.001 to 0.09 | 0.056 |
| Fidelity | 5577 (66) | 0.01 | −0.001 to 0.01 | 0.109 |
| Reach | 5580 (65) | 0.01 | −0.01 to 0.02 | 0.185 |
| Quality | 6046 (73) | 0.07 | −0.03 to 0.58 | 0.150 |
| Practice (postintervention) | 5874 (73) | 0.18 | −0.14 to 0.49 | 0.274 |
| Practice (1-year follow-up) | 6106 (73) | 0.26 | −0.18 to 0.71 | 0.243 |
| Well-being | ||||
| Dose | 5484 (65) | −0.03 | −0.11 to 0.04 | 0.363 |
| Fidelity | 5643 (66) | −0.001 | −0.01 to 0.01 | 0.781 |
| Reach | 5649 (65) | −0.004 | −0.03 to 0.02 | 0.742 |
| Quality | 6116 (73) | −0.02 | −0.18 to 0.15 | 0.849 |
| Practice (postintervention) | 5936 (73) | −0.003 | −0.55 to 0.54 | 0.992 |
| Practice (1-year follow-up) | 6168 (73) | −0.02 | −0.76 to 0.73 | 0.967 |
Coefficient: unstandardised regression coefficient (slope) of the instrumental variable analysis (representing the increase in the outcome per 1 unit increase in the predictor) with cluster-robust maximum likelihood estimation, including schools (clusters) as random effects, and adjusted for the baseline levels of student mental health (ie, risk of depression, social–emotional–behavioural functioning, well-being), cohort, school size, school sex, and country. Dose is the number of SBMT sessions that students received. Quality is the teacher competency delivering the SBMT independently evaluated by using the Mindfulness-based Interventions Teaching Assessment Criteria. Fidelity is the independently rated percentage of the total original SBMT content delivered by the teacher. Reach is the percentage of school study students receiving ≥67% of SBMT sessions. Practice is the amount of home-based student mindfulness practice during/after the intervention.
K, number of clusters (schools) in analysis; n, number of students in analysis; p, p value for the slope; SBMT, school-based mindfulness training.