Background
Methods for setting cut-off scores
Zung’s recommended cut-off score
The current study
Method
Participants
Procedure
Measures
-
Zung Self-rating Depression Scale (SDS)
-
The Zung SDS consists of 20 self-report items that were identified in factor analytic studies of the syndrome of depression [4]. Items tap psychological and physiological symptoms; 10 express negative experience such as “I feel down-hearted and blue” and 10 express positive experience and are reverse scored such as “I eat as much as I used to”. Respondents rate each item according to how it applied to them within the past week using a 4-point scale ranging from 1 (none, or a little of the time) to 4 (most, or all of the time). Total raw scores range from 20 to 80. The SDS has fair internal consistency, with a split-half reliability of .73. An alpha coefficient of .68 was reported by Deforge and Sobal [17] while other authors have reported .79 [18] and .81 [19]. Reported correlations with other depression scales include .41 with the Hamilton Rating Scale [20], .54 with the Depression Adjective Checklist, and .68 with the Beck Depression Inventory [19]. Cronbach’s alpha for the scale on the current study was .87.
-
-
Patient Health Questionnaire (PHQ)
-
The PHQ is designed as a brief, user-friendly self-report measure with items corresponding to a range of DSM-IV diagnostic criteria. It should be noted that all the DSM-IV criteria on which the PHQ is based remain unchanged in the current edition of the manual, DSM-5 [1, 21]. This study utilised the two-page version, covering Major Depressive Disorder and Other Depressive Disorder (9 items), and Panic Disorder and Other Anxiety Disorder (22 items). Compared with diagnoses made by mental health professionals, the sensitivity of the PHQ in relation to depressive disorders is 61% and the specificity 94% [2].
-
Results
Clinical sample (n = 148) | Community sample (n = 210) | |
---|---|---|
Major Depressive Disorder | 46 | 28 |
Other Depressive Disorder | 11 | 10 |
Total Depressive Disorder Diagnoses | 57 | 38 |
SDS mean score (SD) | Clinical sample | Community sample |
---|---|---|
PHQ Positive Depressive Diagnosis | 54.98 (7.25) | 53.92 (6.91) |
No PHQ Diagnosis | 42.77 (9.87) | 38.10 (9.45) |
PHQ Diagnosis/Original sample | ||
---|---|---|
True Positives | False Positives | |
Positive Clinical (n = 57) | 53–54 | 3–4 |
Positive Community (n = 38) | 28 | 10 |
True Negatives | False Negatives | |
Negative Clinical (n = 91) | 57 | 34 |
Negative Community (n = 172) | 154 | 18 |
Cut-off point | Sensitivity |
---|---|
38 | 100.0 |
39 | 100.0 |
40 | 98.2 |
41 | 98.2 |
42 | 96.5 |
43 | 96.5 |
44 | 93.0 |
45 | 93.0 |
46 | 91.2 |
47 | 91.2 |
48 | 87.7 |
49 | 82.5 |
50 | 78.9 |
51 | 74.7 |
52 | 64.9 |
Cut-off point | Community sample (n = 172) |
---|---|
38 | 50.6 |
39 | 53.5 |
40 | 57.0 |
41 | 58.7 |
42 | 59.3 |
43 | 62.8 |
44 | 67.4 |
45 | 70.9 |
46 | 72.1 |
47 | 74.4 |
48 | 76.7 |
49 | 82.0 |
50 | 83.7 |
51 | 93.6 |
52 | 95.3 |
Discussion
40 | 50 | |
---|---|---|
Mean ± 2SD method: cut-off expressed in these terms for: | ||
Negative Community Sample | Mean + 0.20 SD | Mean + 1.26 SD |
Positive Clinical Sample | Mean – 2.07 SD | Mean – 0.69 SD |
Youden Index | .552 | .626 |
ROC curve distance from optimum point | .430 | .267 |
Sensitivity | 98.2% | 78.9% |
Specificity | 57.0% | 83.7% |
Cut-off | False Negatives | False Positives | Total Misdiagnoses |
---|---|---|---|
40 | 0 | 40 | 40 |
50 | 1 | 15 | 16 |