Sequential screening for depression in humanitarian emergencies: a validation study of the Patient Health Questionnaire among Syrian refugees

Table 2 Validity of the screening methods relative to the PHQ-8 MDD classification (n = 129)

Screening method	Major depressive disorder^a			Systematic error^‡ p-value	Cohen’s kappa (95% CI)	Validity indices^b
Screening method	+	−	Total	Systematic error^‡ p-value	Cohen’s kappa (95% CI)	Overall agreement % (95% CI)	Positive agreement % (95% CI)	Negative agreement % (95% CI)	AUC	PPV (%)	NPV (%)
PHQ-2				0.000	0.42 (0.26–0.57)	68 (59–76)	89 (78–96)	50 (38–62)	0.69	61	83
Positive	54	34	88
Negative	7	34	41
Total	61	68	129
PHQ-2/8 sequential screening				0.008	0.70 (0.55–0.80)	95 (89–98)	89 (78–96)	100 (95–100)	0.94	100	91
Positive	54	0	54
Negative	7	68	75
Total	61	68	129

PHQ-2 2-item Patient Health Questionnaire, 95% CI 95% confidence interval, AUC area under the curve for receiver operating characteristic analysis, PPV positive-predictive value, NPV negative-predictive value
^‡McNemar’s test
^aRespondents were defined as having major depressive disorder if they had a positive result on the PHQ-8
^bPositive and negative percent agreement (PPA and NPA, respectively) were calculated as proxy measures of sensitivity and specificity, consistent with recommendations for comparisons with imperfect reference standards [37]

ISSN: 1744-859X