Table 2.

Subgroup analyses and relative findings performed in the included meta-analyses.

Subgroup analysis	Study	Findings from subgroup analysis
Country	Deng et al., 2020 [23]	The prevalence in 20 studies conducted in China was 45% (36–55%). The prevalence for Italy, Ecuador, and Iran was 38% (29–47%), 60% (55–66%), and 38% (28–49%), respectively, as reported by one study in each subgroup
	Dragioti et al., 2021 [30]	Prevalence was higher in low/middle income countries
	Liu C. et al., 2021 [26]	The prevalence reported in 13 studies conducted in China was 39% (25–54%), the prevalence reported in 2 studies conducted in Iran was 65% (0–100%, 95% CI). The prevalence for South Korea, Ecuador, Jordan, Turkey, and Italy was 39% (23–57%), 23% (18–28%), 44% (32–56%), 32% (17–50%), and 11% (6–18%), respectively, as reported by one study in each subgroup
Sex	Deng et al., 2020 [23]	Stratified data for gender were available in 9 studies, prevalence was higher in females 50% (38–62%) than in males 39% (26–53%)
	Dorri et al., 2021 [25]	Stratified data for gender were available in 2 studies, prevalence was higher in females 19% (15–22%) than in males 12% (9–15%)
	Liu C. et al., 2021 [26]	Stratified data for gender were available in 11 studies, prevalence was higher in females 46% (32–60%) than in males 32% (17–47%)
Study design	Deng et al., 2020 [23]	The prevalence of one cohort-study (74%, 62–83%) was higher than the prevalence of 22 cross-sectional studies (44%, 36–53%)
	Dorri et al., 2021 [25]	The prevalence was similar between 4 retrospective (12%, 5–18%) and 3 prospective (12%, 9–15%) cohort studies
	Liu C. et al., 2021 [26]	The prevalence of 3 cohort-studies (88%, 44–100%) was higher than the prevalence of 17 cross-sectional studies (34%, 21–46%)
Severity of depression	Deng et al., 2020 [23]	The prevalence of mild, moderate, and severe depression was 33% (26–39%, 11 studies), 14% (11–16%, 11 studies), and 7% (4–10%, 12 studies), respectively.
	Lao et al., 2020 [27]	The prevalence of mild, moderate, and severe depression was 31% (19–43%, 5 studies), 13% (11–15%, 4 studies), and 5% (2–8%, 4 studies), respectively.
	Liu C. et al., 2021 [26]	The prevalence of mild, moderate, and severe depression was 29% (24–34% 9 studies), 17% (11–22% 9 studies), and 10% (2–20% 11 studies), respectively.
Depression screening tool	Deng et al., 2020 [23]	The prevalence according to different depression screening tools were PHQ-9 (n = 9, 52% (45–59%)), HADS (n = 2, 20% (16–23%)), ZSDS (n = 6, 53% (42–65%)), and SCL-90 (n = 2, 19% (17–22%)). The remaining studies used unvalidated custom questionnaires or interviews (n = 4, 47% (15–80%)).
	Dong et al., 2021 [4]	The prevalence according to different depression screening tools were PHQ-9 (n = 9, 33%), ZSDS (n = 6, 22%), HADS (n = 4, 15%), SCL-90 (n = 4, 15%), HAMD (n = 2, 7%), and PHQ-2 (n = 2, 7%).
	Dorri et al., 2021 [25]	The prevalence according to different depression screening tools were PHQ-9 (n = 2, 16% (13–18%)), HADS (n = 2, 16% (13–20%)), ZSDS (n = 1, 31% (26–36%)), Self-reported questionnaire (n = 1, 4% (3–6%)), DSM-IV (n = 1, 10% (7–13%)), BDI-13 (n = 1, 11% (8–15%))
Disease stage	Liu C. et al., 2021 [26]	17 studies had reported the prevalence of 42% (29–56%) in patients who were experiencing SARS-CoV-2 infection, and 3 studies conducted reported a prevalence of 14% (0–48%, 95% CI) in patients who were at the recovery stage
	Lao et al., 2020 [27]	The prevalence of depressive symptoms in discharged patients was higher (55%, 34–77%, 2 studies) than the prevalence in hospitalized patients was (40%, 28–52%, 6 studies)
COVID-19 severity	Dong et al., 2021 [4]	The prevalence of depression in patients with severe COVID-19 was 66% (16–117%, 2 studies), higher than 31% (7–55%, 4 studies) with clinically stable COVID-19. prevalence for discharged patients was 52% (25–79%, 2 studies).
	Dorri et al., 2021 [25]	The prevalence of depression in patients with severe COVID-19 was 22% (16–28%, 2 studies), higher than 15% (11–18%, 2 studies) with moderate and 13% (8–18%, 2 studies) mild forms.
Setting of care for COVID-19	Deng et al., 2020 [23]	The prevalence was higher in 12 study reporting prevalence for inpatients (48%, 35–61%) than for the single study reporting prevalence for outpatients (35%, 22–48%)
	Premraj et al., 2022 [29]	The prevalence was higher in studies reporting prevalence for outpatients (25%) than for the studies reporting prevalence for inpatients (14%)
Quality	Liu C. et al., 2021 [26]	7 high-quality studies had reported a prevalence of 38% (17–60%) and the remaining 13 low-quality studies reported a prevalence of 34% (16–54%)
Follow-up duration	Dorri et al., 2021 [25]	The prevalence was similar between studies with mean/median follow-up duration ≤31 days and those with longer follow-up time (>31): 16% (13–18%, 3 studies) vs. 15% (2–28%, 4 studies) respectively