Table 2

The country, subjects, test/population size, key variable, key comparator and key outcome of included studies

Author	Subject	N (tests)	Key dependent variable	Key test group	Key outcome
Laboratory studies
Lee et al 22 Taiwan	Human	30	PF to NaCl	FRSMs vs FFP2/FFP3	Respirators provide up to 16× higher PF in measured conditions
Gawn et al 6 UK	Dummy+human	19	PF to NaCl and to live influenza virus	FRSMs vs FFP1/FFP2/FFP3	Respirators provide up to 17× higher PF in measured conditions. Live influenza virus penetrates/circumnavigates FRSMs
He et al 23 USA	Dummy	20	PF to NaCl	FRSMs vs N95	Respirators provide up to 108× higher PF in measured conditions, but is reduced to 13.4× at high MIF
Clinical studies
Radonovich et al 24 USA	7 US hospitals	2862	Laboratory-confirmed influenza	FRSM vs N95	No significant difference in infection rates between the two groups
Loeb et al 25 Canada	8 Canadian hospitals	446	Respiratory symptoms Laboratory-confirmed influenza	FRSM vs N95	No significant difference in infection rates between the two groups
MacIntyre et al 26 China	19 Chinese hospitals	1669	Respiratory symptoms Laboratory-confirmed influenza	FRSM vs N95 Targeted vs continuous use	Symptom rates were highest in FRSMs (17%), followed by respirators (targeted, 11.8%), followed by respirators (continuous, 7.2%)
MacIntyre et al 27 China	15 Beijing hospitals	1441	Respiratory symptoms Laboratory-confirmed influenza	FRSM vs N95 vs no mask	Respirators were statistically more protective than control group
Ng et al 28 Singapore	1 Singaporean hospital	41	Laboratory-confirmed SARS-CoV-2	FRSM vs N95	No confirmed SARS-CoV-2 cases—efficacy difference between two groups not confirmed
Loeb et al 29 Canada	2 Canadian hospitals	43	Retrospective recall of SARS symptoms	FRSM vs N95 vs no mask	80% reduction in risk of infections using mask vs no mask, but no difference between mask types

FFP, filtering face piece; FRSM, fluid repellent surgical mask; MIF, mean inspiratory flow; PF, protection factor.