Table 2.

Quality Assessment Details of Included Studies

Question	Wang et al. (2014)49	Yuan (2019)43	Yeung et al. (2008)25	Zhang et al. (2013)50	Ye et al. (2020)44	Chen et al. (2013)46	Du et al. (2017)48
Overall risk of bias	Low	Indeterminate: unknown interval between test and reference standard	Low	Low	Indeterminate: no description of exclusion criteria, unknown blinding protocol	Low	Low
Was a consecutive or random sample of patients enrolled?	Yes: probability proportionate to size sampling and simple random sampling	Yes: consecutive	Yes: consecutive	Yes: consecutive	Yes: consecutive	Yes: random	University student email list—unclear if truly random
Was a case/control design avoided?	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Did the study avoid inappropriate exclusions?	Yes	Yes: illness preventing completion, previous history of mental illness or psychotherapy, unwilling	Yes	Yes: disabling disease or reduced life expectancy, difficulty communicating	Information not found	Yes: patients with psychotic disorders	Yes
Were the index testa results interpreted without knowledge of the results of the reference standardb?	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Are the specificity and sensitivity recorded for multiple cutoff scores?	Yes	Yes	No: only one (15)	Yes	Yes	Yes	Yes
Was the index test administered in a standardized fashion?	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Was an appropriate version of the index test used?	Information not found	Information not found	Yes: translated by investigators	Yes: translation on Hong Kong government website	Information not found	Information not found	Information not found
Is the reference standard likely to correctly classify the target condition?	Yes: MINI	Yes: MINI	Yes: SCID	Yes: MINI	Yes: DSM-V	Yes: SCID	Yes: MINI
Were the reference standard results interpreted without knowledge of the results of the index test?	Information not found	Yes	No: SCID conducted for all patients with PHQ-9≥15 and 8% of patients with PHQ-9<15	Yes	Information not found	Yes	Yes
Was the reference standard administered in a standardized fashion?	Yes: one interviewer	Yes: standardized training	Yes: one interviewer	Yes	Yes: one interviewer	Yes: two psychiatrists with interrater reliability 0.81	Yes: two psychiatrists
Was there an appropriate interval between index tests and reference standard?	Yes: <2 weeks	Information not found	No: “about 4 weeks”	No: “within 2–4 weeks”	Yes: immediate	Yes: 2 weeks	Yes: within 48 h
Did all patients receive a reference standard?	Yes	Yes	No: subsample analysis	No: subsample analysis	Yes	No: subsample analysis	No: subsample analysis
Did patients receive the same reference standard?	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Were all patients included in the analysis without a significant percentage of dropouts?	Yes: 0% dropout rate (1045/1045)	Yes: 0% dropout rate (782/782)	Yes: 0% dropout rate (1940/1940)	Yes: 0% dropout rate (586/586)	Yes: 0% dropout rate (148/148)	Yes: 7% dropout rate (280/300)	Yes: 0% dropout rate (230/230)

Question	Peng et al. (2020)45	Liu et al. (2016)40	Xiong (2015)47	Zhang et al. (2013)50	Xia (2019)51	Chen (2010)53	Liu et al. (2011)41
Overall risk of bias	Indeterminate: unknown blinding protocol, interval between test and reference standard	Low	Indeterminate: unknown blinding protocol, interval between test and reference standard	Low	Low	Low	Low
Was a consecutive or random sample of patients enrolled?	Yes: consecutive	Yes: study province → random towns → random villages → all individuals	Yes: on randomly assigned days, patients consecutively invited	Yes	Yes: consecutive	Yes: consecutive	Yes: consecutive
Was a case/control design avoided?	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Did the study avoid inappropriate exclusions?	Yes: patients with pre-existing mental illness, severe physical illness	Yes: if not in area during study period or difficulty with communication due to illness	Yes: wrong language, limited writing skills, cognitive impairment, psychosis, acute suicidality	Yes	Yes: severe psychiatric disorders or antisocial, schizotypal personality disorders	Yes	Yes
Were the index testa results interpreted without knowledge of the results of the reference standardb?	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Are the specificity and sensitivity recorded for multiple cutoff scores?	No: only one (9)	Yes	Yes	Yes	Yes	Yes	Yes
Was the index test administered in a standardized fashion?	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Was an appropriate version of the index test used?	Information not found	Yes: cited Yeung et al. 2008	Information not found	Information not found	Information not found	Information not found	Yes: study team translated and back translated
Is the reference standard likely to correctly classify the target condition?	Yes: SCID	Yes: SCID	Yes: MINI	Yes: SCID	Yes: MINI	Yes: SCID	Yes: SCAN
Were the reference standard results interpreted without knowledge of the results of the index test?	Information not found	No: SCID conducted immediately after PHQ-9 by same interviewer	Information not found	Yes	Yes	Yes	Yes
Was the reference standard administered in a standardized fashion?	Yes: two psychiatrists	Yes: standardized interview training	Yes	Yes: standardized training	Yes: one interviewer	Yes: two psychiatrists	Yes: trained staff with 0.88 interrater reliability
Was there an appropriate interval between index tests and reference standard?	Information not found	Yes: immediate	Information not found	Yes: <2 weeks	Yes: <24h	Information not found	Information not found
Did all patients receive a reference standard?	Yes	Yes	Yes	Yes	Yes	No: subsample analysis	Yes
Did patients receive the same reference standard?	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Were all patients included in the analysis without a significant percentage of dropouts?	Yes: 8% dropout rate (275/300)	Yes: 0.7% dropout rate (839/845)	Yes: 0% dropout rate (491/491)	Significant: 13.8% dropout rate (959/1112)	Yes: 0% dropout rate (213/213)	Significant: 20% dropout rate (77/96) for subsample chosen to undergo reference standard	Significant: 21.6% dropout rate, however, sample was large (1532/1954)

Question	Jin et al. (2011)39,c		Xu et al. (2007)38,c	Yang et al. (2015)37,c	Zhi et al. (2013)36,c	Chen (2015)35,c	Wang et al. (2015)33,c
Overall risk of bias	Indeterminate: unknown blinding protocol, interval between test and reference standard		Low	Low	Low	Low	Low
Was a consecutive or random sample of patients enrolled?	Yes: random		Yes: random	Yes: consecutive	Yes: consecutive	Yes	Yes: consecutive
Was a case/control design avoided?	Yes		Yes	Yes	Yes	Yes	Yes
Did the study avoid inappropriate exclusions?	Yes: dementia or physically incapable of completing questionnaire		Yes	Yes: incapable of understanding and responding to questionnaire, severe mental health symptoms or physical disability, substance use disorders, currently receiving medication or therapy treatment for mental illness	Yes: physically incapable of completing questionnaire, severe physical or mental illness	Yes: those incapable of completing study or consent due to language or hearing	Yes
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes		Yes	Yes	Yes	Information not found	Yes
Are the specificity and sensitivity recorded for multiple cutoff scores?	No: only one (10)		No: only one (15)	No: only one (10)	Yes	Yes	Yes
Was the index test administered in a standardized fashion?	Yes		Yes	Yes	Yes	Yes	Yes
Was an appropriate version of the index test used?	Information not found		Information not found	Information not found	Information not found	Information not found	Yes: cited Liu 2011's translation
Is the reference standard likely to correctly classify the target condition?	Yes: SCID		Yes: SCID	Yes: SCID	Yes: MINI	Yes: MINI	Yes: CIDI
Were the reference standard results interpreted without knowledge of the results of the index test?	Information not found		Information not found	Yes	Information not found	Information not found	Yes
Was the reference standard administered in a standardized fashion?	Yes: standardized training		Yes: two psychiatrists	Yes: standardized training	Yes: standardized training	Yes	Yes: one interviewer
Was there an appropriate interval between index tests and reference standard?	Information not found		Yes: 1 week	Information not found	Yes: immediate	Yes: immediate	Yes: immediate
Did all patients receive a reference standard?	No: subsample analysis		No: subsample analysis	No: subsample analysis	Yes	Yes	Yes
Did patients receive the same reference standards?	Yes		Yes	Yes	Yes	Yes	Yes
Were all patients included in the analysis without a significant percentage of dropouts?	Yes: 0% dropout rate (117/117)		Yes: 0% dropout rate (195/195)	Yes: 0% dropout rate (97/97)	0.6% dropout rate (1997/2009)	Yes: 0% dropout rate (319/319)	Yes: 0% dropout rate (201/201)

^a Index test=PHQ.

^b Reference standard=old standard clinical interview used by the particular study.

^c Study published in Chinese.