School-based mindfulness interventions in children and adolescents: an umbrella review of systematic reviews, with implications for physical education

Anxiety, depressive symptoms and stress-related difficulties have increased among school-aged children and adolescents, and the school has been identified as a priority setting for universal preventive strategies. Mindfulness-based interventions (MBIs) are increasingly implemented in schools, and Physical Education has been proposed as a suitable curricular space because it explicitly mobilises body awareness, attention and emotional self-regulation. A growing number of systematic reviews with meta-analysis are available, but they differ in populations, protocols, comparators and outcomes, and their conclusions are not uniform. This umbrella review aimed to (i) synthesise the evidence provided by systematic reviews with meta-analysis on the effects of school-based MBIs on emotional, behavioural and cognitive outcomes in children and adolescents aged 5–18 years; (ii) appraise the methodological quality of those reviews and the certainty of the evidence; (iii) examine the factors that moderate effectiveness; and (iv) identify gaps to guide future research. Umbrella review of systematic reviews with meta-analysis, conducted according to the Joanna Briggs Institute guidance for umbrella reviews and reported following the PRISMA 2020 statement. The protocol was prospectively registered in PROSPERO (CRD420261381514). PubMed, Cochrane Library and EBSCO were searched from January 2014 to 29 May 2026, with publications restricted to English and Spanish. Two reviewers independently screened records, extracted data and appraised methodological quality with AMSTAR-2; the overlap of primary studies across reviews was quantified using the corrected covered area. Because the included reviews overlapped and differed in populations, comparators and effect metrics, no new pooled analysis was conducted: findings were synthesised narratively and structured by outcome domain, and the certainty of the evidence for each domain was rated with the GRADE approach. Because the protocol-defined requirement for a quantitative synthesis was broadened during full-text screening (Sect. 2.4), reviews without a pooled effect size were also eligible; this deviation and its rationale are reported in the Methods. Nine reviews met the eligibility criteria, which were broadened during full-text screening to admit narrative and correlational syntheses alongside meta-analyses. Verification against the original publications showed that four reviews report a pooled quantitative effect [14, 8, 3, 6] and five synthesise their evidence narratively, correlationally, or through a levels-of-evidence rating [4, 7, 24, 25]; three reviews included wholly or partly clinical samples. The two meta-analyses most directly addressing anxiety disagreed: [14] reported a small significant effect (d = 0.16–0.30, attenuated under active comparators) while [8] reported a non-significant pooled effect (d = 0.013). Depressive symptoms improved in two reviews and were explicitly unimproved in the non-clinical subgroup of a third. Evidence for emotion regulation and emotional intelligence was mixed and partly correlational rather than experimental. The single review addressing cognitive functioning found a significant effect in fewer than half of its primary studies and could not isolate mindfulness from co-occurring intervention components. No review evaluated interventions delivered specifically within Physical Education. The evidence for school-based MBIs in children and adolescents is more uncertain, and less consistent across reviews, than earlier secondary citations of this literature suggest. At most, a small and inconsistent effect on anxiety is supported; evidence for other outcomes is weaker, indirect, or based on designs that cannot support a causal claim. No review provides direct evidence on Physical Education as a delivery context. These interventions should not be presented as a demonstrated solution to youth mental health difficulties on the basis of this evidence, and any recommendation to integrate them into Physical Education remains a hypothesis to be tested, not a conclusion to be drawn from the reviews synthesised here. Prospero CRD420261381514. Date 29/04/2026.