Clarification of the cut-off score for Zung's self-rating depression scale

Abstract

Background: Zung's Self-rating Depression Scale (SDS) is an established norm-referenced screening measure used to identify the presence of depressive disorders in adults. Despite widespread usage, issues exist concerning the recommended cut-off score for a positive diagnosis. First, confusion arising from the conversion of raw scores to index scores had resulted in a considerably higher cut-off score than that recommended being used by many researchers. Second, research in China [Chin J Nervous Mental Dis. 12:267-268; 2009] and Australia [BMC Psychiatry. 17:329; 2017] had suggested that the current recommended cut-off is lower than ideal, at least in those countries.

Method: To explore these matters further, sensitivity and specificity figures for alternative cut-off points were examined in positive clinical and negative community samples respectively. The positive clinical sample (n = 57) consisted of adults receiving treatment from a medical professional for some kind of depressive disorder, whose diagnosis was positively confirmed using the Patient Health Questionnaire (PHQ). The negative community sample (n = 172) was derived from a representative sample of adults whose absence of any depressive disorder was similarly confirmed by the PHQ.

Results: Mathematical models, including Youden's Index and the Receiver Operating Characteristics Curve, suggest that the recommended cut-off (a raw score of 40) is indeed too low. More detailed comparisons, including consideration of the likely numbers of false positives and negatives given prevalence rates, confirm that, ironically, the incorrect SDS cut-off score mistakenly applied by many researchers (a raw score of 50) would appear to provide far greater accuracy.

Conclusions: Research in China [Chin J Nervous Mental Dis. 12:267-268; 2009] has resulted in an elevated SDS cut-off score of 42 being used in many Chinese studies. Research by Dunstan and Scott [BMC Psychiatry. 17:329; 2017] in an Australian context, suggested that a greater increase, to a raw score of 44 might be required. Based on this study, an even larger adjustment is required. Specifically, we recommend the use of an SDS raw score of 50 as the cut-off point for clinical significance.

Keywords: Cut-off score; Depression screening; Zung self-rating depression scale (SDS).

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

The ROC curve (blue line) for a perfectly discriminating test. Whatever cut-off point is chosen, either sensitivity or specificity = 1, hence the ROC curve in such a case consists of two straight lines. The dotted black line represents the points where sensitivity and specificity are equal

Fig. 2

The ROC curve (blue line) obtained for the combination of the Positive Clinical and Negative Community samples. For each possible SDS cut-off point, the sensitivity of the SDS amongst the Positive Clinical subsample is graphed against 1- the specificity recorded amongst the Negative Community subsample