Table 3.

Certainty of evidence for the risk factors associated with adverse mental health outcomes on health and care staff during the COVID-19 pandemic

No of studies	Design	Risk of bias	Additional considerations	Certainty (overall score)a
Factor: Frontline staff/Close contact with COVID-19 patients [30, 31, 34, 42]
4	2	2	Inconsistency: Higher burnout reported in non-frontline staff (Cancer hospital, Wuhan) [43]. Frontline nurses reported lower vicarious trauma scores [32]. No difference between frontline and non-frontline staff reported (This finding was not statistically significant) [33].	Moderate
Factor: Nurse [24, 26, 31]
3	2	2	Inconsistency: Doctors were found to have more sleep disturbances than nurses (This finding was not statistically significant) [42]. Not all confounding factors were dealt with in the three studies reporting nurses to be at a higher risk for adverse psychological outcomes. No studies compared nurses to primary care or social staff.	Moderate
Factor: Clinical healthcare workers [34, 47]
2	2	2	Inconsistency in these findings [39].	Moderate
Factor: Heavy workload [26, 35, 37]
3	2	2	No serious inconsistencies.	High
Factor: Lack of personal protective equipment (PPE) [24, 28, 40, 46]
4	2	2	No serious inconsistencies.	High
Factor: Point of outbreak [37, 58]
2	2	1	Only two studies - one was limited to the sample of a surgical department where confounding factors were not dealt with and one was a qualitative study.	Low
Factor: Rural location [47]
1	2	0	Only one study reported findings on effect of rurality.	Very low
Factor: Fear of infection [21, 26, 28]
3	2	2	No serious inconsistencies.	High
Factor: Concern about family [24, 26, 37, 40]
4	2	1	This theme was predominantly raised in qualitative literature.	Moderate
Factor: Younger age [24, 29, 33]
3	2	2	Age was found to be a complex risk factor where the focus of anxiety depended on the age group assessed [24].	Low
Factor: Gender – Female [31, 47]
2	2	1	Inconsistencies were found – for example: a large global survey of dentists found no differences based on gender [21]. Furthermore, confounding factors assessing gender in both included studies were not satisfactorily dealt with.	Low
Factor: Organic illness [36, 47]
2	2	1	No serious inconsistencies.	Moderate
Factor: Being an only child [35, 42]
2	2	0	No serious inconsistencies.	Low

^a 4 High = This research provides a very good indication of the likely effect. The likelihood that the effect will be substantially different** is low

3 Moderate = This research provides a good indication of the likely effect. The likelihood that the effect will be substantially different** is moderate

2 Low = This research provides some indication of the likely effect. However, the likelihood that it will be substantially different** is high

1 Very low = This research does not provide a reliable indication of the likely effect. The likelihood that the effect will be substantially different** is very high

** Substantially different = a large enough difference that it might affect a decision