Table 2

Summary of evidence—observational studies for the effectiveness of face mask wearing in the community

Observational studies	OR (95% CI)	Setting	Limitations
Lau et al 24	0.36 (0.25 to 0.62)	Population wide	Misclassification: WHO’s case definition for probable severe acute respiratory syndrome (SARS) cases was used. Some cases may not have laboratory confirmation. Reporting bias.
Wu et al 25	Sometimes wore a mask: 0.40 (0.02 to 0.9) Always wore a mask: 0.3 (0.1 to 0.6)	Population wide	Misclassification: China Ministry of Health’s definitions for probable SARS cases was used. Some cases may not have laboratory confirmation. Reporting bias.
Uchida et al 26	0.859 (0.778 to 0.949)	Population wide	Misclassification less likely: 96.4% of diagnoses by rapid influenza diagnostic kits. Reporting bias.