Table 2

Summary of findings

Question	Unique primary studies	Outcome	Result	Magnitude of effect	GRADE
1: What is the effect of school closures on community COVID-19 transmission, morbidity and mortality?	132	Transmission	Reduced: 6 SRs20–23 25 27 (closures) Uncertain: 2 SRs18 19 (closures) No effect: 2 SRs17 18 (re-openings)	Transmission from pupils less likely than from adults (OR 0.26, 95% CI (0.11 to 0.63))19 (Meta-analysis) Pupils may be less likely to be infected than adults (OR 0.6, 95% CI (0.25 to 1.47))19 (Meta-analysis)	Very low* ⨁◯◯◯
		Hospitalisation	Reduced: 1 SR26	45% reduction in paediatric hospital admissions26	Very low† ⨁◯◯◯
		Mortality	Reduced: 1 SR24	2%–4% reduction in deaths24	Very low‡ ⨁◯◯◯
2: What is the effect of school closures on children’s health?	123	Learning	Worse: 4 SRs26 28–30	Learning loss of −0.005 to −0.05 SD, per week of closures28	Low ⨁⨁◯◯
	164	Mental health	Worse: 8 SRs26 31–37	Prevalence of anxiety during COVID-19: 24% (95% CI (20% to 29%))31(Meta-analysis) Anxiety in outbreak phase: 25% (95% CI (17% to 34%))31 (Meta-analysis) Anxiety in diffusion attenuation phase: 42% (95% CI (35% to 50%))31(Meta-analysis) Prevalence of mental health problems during COVID-19: 28% (95% CI (22% to 34%))37 (Meta-analysis)	Very low§ ⨁◯◯◯
	22	Physical health	Worse: 3 SRs26 35 39	BMI increased by 0.77 units (95% CI (0.33 to 1.2)) p=0.0006), obesity prevalence by 1.23-fold (95% CI (1.1 to 1.37)) p=0.0002), and bodyweight by 2.67 kg (95% CI (2.12 to 3.23)) p<0.0000139 (Meta-analysis)	Very low¶ ⨁◯◯◯
	16	Sleep	Worse: 2 SRs35 40 No effect: 1 SR26	54% (95% CI (50% to 57%)) had sleep disturbance40 (Meta-analysis) 49% (95% CI (39% to 58%)) did not achieve recommended sleep quantities40 (Meta-analysis)	Very low** ⨁◯◯◯
	7	Domestic violence	Increased: 1 SR41 Uncertain: 1 SR35	67% reduction in reported cases of domestic violence involving children41 41% reduction in police reports for domestic violence involving children41 Estimated increase in victims of violence against children: 11 488 186–18 381 098 in Africa, 3 577 839–5 724 542 in Asia, 2 921 466–4 674 345 in Latin America, 7 59 600–1 215 360 in Europe, 2 009 722–3 215 554 in North America, and 32 010–51 216 in Oceania41	Very low†† ⨁◯◯◯
3: What is the effect of in-school mitigations on community COVID-19 transmission, morbidity or mortality?	69	Transmission	Reduced: 3 SRs17 42 43	Mask usage in schools associated with reduction in R in community by 0.011 (95% CI (0.008 to 0.0127))42	Very low‡‡ ⨁◯◯◯
		Hospitilisation	Reduced: 1 SR42	Mask usage in schools associated with reduction in excess hospitalisation per 10 000 of population in teachers to 4.2 (95% CI (−47.39 to 48.09)) from 40.5 (95% CI (−46.95 to 146.64))42	Very low§§ ⨁◯◯◯
		Mortality	Reduced: 1 SR42	Mask usage in schools associated with reduction in community mortality by ratio of 1.5 (95% CI (1.5 to 1.6))42	Very low¶¶ ⨁◯◯◯
4: What is the effect of in-school mitigations on children’s health?	0	Children’s health	0 SRs	Not available	–

Number of unique studies in individual SRs. For question 1 - 20: 2, 21: 3, 22: 5, 23: 4, 25: 15, 27: 3, 26: 1, 24: 4, 17: 14, 18: 25, 19: 34. For question 2 (learning) - 28: 1, 30: 77, 29: 29, 26: 3. For question 2 (mental health) - 31: 7, 32: 0, 33: 3, 34: 103, 35: 15, 36: 8, 26: 1, 37: 5. For question 2 (physical health) - 35: 7, 26: 2, 39: 12. For question 2 (sleep) - 40: 6, 35: 6, 26: 0. For question 2 (domestic violence) - 41: 4, 35: 2. For question 3 - 17: 17, 42: 38, 43: 14.
Data from meta-analysis were prioritised to represent effect measures for each outcome, where available.
*Downgraded for risk of bias and inconsistency.
†Downgraded for imprecision.
‡Downgraded for imprecision and risk of bias.
§Downgraded for risk of bias.
¶Downgraded for risk of bias.
**Downgraded for inconsistency and imprecision.
††Downgraded for indirectness.
‡‡Downgraded for indirectness and imprecision.
§§Downgraded for risk of bias, indirectness and imprecision.
¶¶Downgraded for risk of bias, indirectness and imprecision.
BMI, body mass index; GRADE, Grading of Recommendations, Assessment, Development and Evaluations.