Table S3.

Grading of confidence in the reported results for various outcomes

Number of studies	Certainty assessment						Effect			Certainty	Importance
	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	Number of events	Number of individuals	Rate (95% CI)
Anxiety
43	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Publication bias strongly suspected strong association all plausible residual confounding would reduce the demonstrated effectc	27,525	90,080	Event rate 26.59 per 100 (22.81-30.36)	⨁⨁◯◯ LOW	Important
Anxiety among frontline health care workers
10	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Strong association all plausible residual confounding would reduce the demonstrated effect	2065	6545	Event rate 27.2 per 100 (18.1-36.31)	⨁⨁⨁◯ Moderate	Important
Anxiety among second line health care worker
7	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Strong association all plausible residual confounding would reduce the demonstrated effect	3199	12,124	Event rate 26.89 per 100 (20.35-33.43)	⨁⨁⨁◯ moderate	Important
Anxiety among general population
24	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Publication bias strongly suspected strong association all plausible residual confounding would reduce the demonstrated effectc	21,778	68,742	Event rate 25.87 per 100 (20.51-31.24)	⨁⨁◯◯ low	Important
Mild anxiety
18	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Publication bias strongly suspected strong association all plausible residual confounding would reduce the demonstrated effectc			Event rate 16.69 per 100 (12.27-21.11)	⨁⨁◯◯ Low	Not important
Moderate anxiety
18	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Publication bias strongly suspected all plausible residual confounding would reduce the demonstrated effectc			Event rate 7.34 per 100 (4.4-10.27)	⨁◯◯◯ very low	Important
Severe anxiety
17	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Publication bias strongly suspected all plausible residual confounding would reduce the demonstrated effectc			Event rate 5.39 per 100 (3.19-7.59)	⨁◯◯◯ very low	Critical
Depression
43	Observational studies	Seriousd	Very seriousb	Not serious	Not serious	Publication bias strongly suspected strong association all plausible residual confounding would reduce the demonstrated effectc	21,019	77,932	Event rate 26.17 per 100 (21.84-30.5)	⨁⨁◯◯ Low	Critical
Depression among frontline health care worker
9	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Strong association all plausible residual confounding would reduce the demonstrated effect	1927	5958	Event rate 32.1 per 100 (17.97-46.22)	⨁⨁⨁◯ moderate	Critical
Depression among second line health care workers
5	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Strong association all plausible residual confounding would reduce the demonstrated effect	815	4625	Event rate 15.72 per 100 (11.28-20.15)	⨁⨁⨁◯ moderate	Critical
Depression among general population
25	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Publication bias strongly suspected strong association all plausible residual confounding would reduce the demonstrated effectc	17,496	62,195	Event rate 25.87 per 100 (20.22-31.53)	⨁⨁◯◯ low	Critical
Mild depression
12	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Publication bias strongly suspected strong association all plausible residual confounding would reduce the demonstrated effectc			Event rate 20.81 per 100 (13.94-27.69)	⨁⨁◯◯ Low	Not important
Moderate depression
13	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Publication bias strongly suspected all plausible residual confounding would reduce the demonstrated effectc			Event rate 7.41 per 100 (4.93-9.9)	⨁◯◯◯ very low	Important
Severe depression
12	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Publication bias strongly suspected all plausible residual confounding would reduce the demonstrated effectc			Event rate 4.61 per 100 (2.65-6.58)	⨁◯◯◯ very low	Critical
Stress
20	Observational studies	Seriousa	Very seriousb	Seriousd	Not serious	Strong association all plausible residual confounding would reduce the demonstrated effect	25,210	79,138	Event rate 26.16 per 100 (17.73-34.59)	⨁⨁◯◯ low	Important
Stress among frontline health care worker
3	Observational studies	Seriouse	Very seriousb	Seriousd	Seriousc	Strong association all plausible residual confounding would reduce the demonstrated effect	1165	1903	Event rate 55.63 0.36 110.9 per 100 (0.36-100)	⨁◯◯◯ very low	Important
Stress among second line health care worker
3	Observational studies	Seriousa	Very seriousb	Seriousd	Seriousc	All plausible residual confounding would reduce the demonstrated effect	86	1244	Event rate 7.01 per 100 (3.11-10.92)	⨁◯◯◯ very low	Important
Stress among general population
13	Observational studies	Seriousa	Very seriousb	Seriousd	Not serious	Strong association all plausible residual confounding would reduce the demonstrated effect	23,892	73,506	Event rate 25.54 per 100 (17.8-33.29)	⨁⨁◯◯ low	Important
Insomnia
11	Observational studies	Seriousa	Very seriousb	Not serious	Not serious	Strong association all plausible residual confounding would reduce the demonstrated effect	7490	27,775	Event rate 32.87 per 100 (23.94-41.8)	⨁⨁⨁◯ moderate	Important
Insomnia among frontline health care worker
3	Observational studies	Seriousa	Seriousb	Not serious	Seriousd	Publication bias strongly suspected strong associationc	1362	3828	Event rate 34.44 per 100 (32.54-36.34)	⨁◯◯◯ very low	Important
Insomnia among second line health care worker
2	Observational studies	Very seriousb, d	Seriousb	Not serious	Seriousd	Strong association all plausible residual confounding would reduce the demonstrated effect	1270	3505	Event rate 33.96 per 100 (32.48-35.43)	⨁⨁◯◯ low	Important
Insomnia among general population
6	Observational studies	Seriousb	Seriousb	Not serious	Not serious	Strong association all plausible residual confounding would reduce the demonstrated effect dose response gradient	4858	20,442	Event rate 29.36 per 100 (17.02-41.69)	⨁⨁⨁⨁ high	Important

^aNonrandom sampling was used for most of the studies, ^bHeterogeneity index measured using I2 statistic is very high, ^cFew number of studies, ^dThere was wide variety in the scales and definition of stress among studies, ^eThe quality assessed by new Castel Ottawa scale is poor. CI: Confidence interval