Investigación original / Original research

The Depressive Psychopathology Scale:

presentation and initial validation in a
sample of Peruvian psychiatric patients
Johann M. Vega-Dienstmaier,1 Santiago Stucchi-Portocarrero,2
Nancy Valdez-Huarcaya,3 Miriam Cabra-Bravo,2 and Maria I. Zapata-Vega 4

Suggested citation Vega-Dienstmaier JM, Stucchi-Portocarrero S, Valdez-Huarcaya N, Cabra-Bravo M, Zapata-Vega MI.

The Depressive Psychopathology Scale: presentation and initial validation in a sample of Peruvian psy-
chiatric patients. Rev Panam Salud Publica. 2011;30(4):317–26.

ABSTRACT Objective. Presentation and validation of the Depressive Psychopathology Scale (DPS), a
new, Spanish-language psychometric inventory, in a sample of Peruvian psychiatric patients.
Methods. From 1999 to 2001, the DPS, Zung Self-rating Depression Scale (ZSDS), and
Major Depressive Disorder (MDD) module of the Structured Clinical Interview for DSM-IV
Disorders (SCID) were administered to 226 nonpsychotic outpatients referred to the National
Institute of Mental Health in Lima, Peru, for their initial clinical evaluation. In the evalua-
tion, attending psychiatrists 1) corroborated the general diagnosis and presence or absence of
MDD and 2) rated depression severity based on clinical opinion and on Clinical Global Im-
pression—Severity scale criteria.
Results. Mean time to complete the DPS was 7.22 minutes (standard deviation, 3.99).
Cronbach’s alpha value was 0.86. For diagnosis of MDD, based on the SCID, area under re-
ceiver operating characteristic curve (AUROC) was 0.872 and the selected cutoff score (26/27)
had 81.32% sensitivity and 80% specificity; based on the attending psychiatrists’ evaluation,
AUROC was 0.832 and the selected cutoff score (25/26) had 77.67% sensitivity and 72.32%
specificity. The DPS was significantly correlated with the ZSDS (rho = 0.8, P < 0.001). Some
DPS items (“depression worse in the morning,” “appetite disturbances,” “mood reactivity,”
and “hypersomnia”) showed low loadings on the five factors extracted through principal com-
ponent analysis and/or did not significantly correlate with depression parameters.
Conclusions. The DPS can predict MDD and has convergent validity, as shown by its cor-
relation with the ZSDS. However, additional psychometric studies are recommended to sim-
plify and improve it.

Key words Psychiatric status rating scales; depression; validation studies; Peru.

Depression was estimated as the fourth signed specifically for the assessment of
1 Department of Psychiatry and Mental Health,
leading cause of disease burden in the depression. However, most have been de-
Universidad Peruana Cayetano Heredia, Lima, year 2000, representing a major public veloped in English and later translated
Peru. Send correspondence to: Johann M. health problem worldwide (1). Availabil- into other languages, including Spanish.
Vega- Dienstmaier, [email protected];
[email protected] ity of reliable and valid assessment in- Validation of Spanish-language ver-
2 Instituto Nacional de Salud Mental “Honorio struments is crucial for the screening of sions of widely used depression scales—
Delgado-Hideyo Noguchi,” Lima, Peru. mental disorders in primary care and the including the Hamilton Depression Rat-
3 Pontificia Universidad Católica del Perú, Lima, Peru.
4 Mount Sinai School of Medicine, Elmhurst Hospi- advancement of mental health research. ing Scale (HAM-D), the Montgomery-
tal Center, New York, United States of America. Numerous instruments have been de- Åsberg Depression Rating Scale (MADRS),

Rev Panam Salud Publica 30(4), 2011 317

Original research Vega-Dienstmaier et al. • Depressive Psychopathology Scale: presentation and validation in Peru

the Beck Depression Inventory (BDI), the MATERIALS AND METHODS scale for assessing the severity of mental
Zung Self-rating Depression Scale illness (depression, in the current study)
(ZSDS), and the Center for Epidemio- Psychiatric measures and sources of according to the clinical experience of the
logic Studies Depression Scale (CES-D)— clinical information rater, who classifies the patient into one
can be found in the international litera- of seven categories: 1 (“not at all de-
ture (2–6). Depressive Psychopathology Scale pressed”); 2 (“borderline depressed”);
A recent literature review concluded (DPS).5 The DPS is a new instrument de- 3 (“mildly depressed”); 4 (“moderately
that evidence for diagnostic accuracy of veloped by the principal investigator depressed”); 5 (“markedly depressed”);
Spanish-language depression screening (PI) and lead/corresponding author of 6 (“severely depressed”); or 7 (“extremely
instruments was found only for the CES-D the current study (JMVD) and reviewed depressed”). A significant correlation has
and the Primary Care Evaluation of Men- in collaboration with other mental health been shown between the CGI-S and the
tal Disorders (PRIME-MD) (for primary researchers. The items are based on diag- HAM-D in patients with depression (14).
care patients), the Geriatric Depression nostic criteria from two standardized
Scale (GDS) (for elderly patients), and the classification systems: the ICD-10 (10) Clinical evaluation. All study subjects
Edinburgh Postnatal Depression Scale and the DSM-IV (9). DPS items were underwent a clinical assessment as de-
(EPDS) and Postnatal Depression Screen- based on DSM-IV criteria for Major De- scribed below in “Step 2.”
ing Scale (PDSS) (for postpartum patients). pressive Episode (DPS items 1, 2, 4, 5, 7,
The study and development of brief de- 8, 9, 11, 12, 14, 15, 18, 19, and 20); melan- Training of the study team
pression tests for Spanish-speaking popu- cholic features (DPS items 1, 3, 7, 8, 9, 16,
lations has thus been recommended (7). 17, and 18); and atypical features (DPS Prior to the beginning of the study, the
Language and other cultural factors items 1, 4, 6, 10, and 17); and on ICD-10 interviewers were trained by the PI. The
may not be the only elements that need to Depressive Episode symptomatology training included three phases: 1) team
be taken into account in the validation of (DPS items 1, 2, 5, 7, 12, 13, 15, 18, 19, and meetings, in which the PI explained the
depression scales in Latin America. As 20). The resulting self-rated instrument procedures for enrolling and obtaining
pointed out by Bagby et al. (8), the estab- is composed of 20 items referring to the consent of the subjects and adminis-
lished scales lack broad psychometric symptoms experienced in the last 2 tering the instruments; 2) administration
evaluations, and may be deficient in as- weeks. With the exception of items 1 and of the SCID MDD module by the PI to 10
sessing depression as it is currently con- 16 (which are scored from 0 to 2) and outpatients, in the presence of the inter-
ceptualized. For example, the HAM-D, item 20 (scored from 0 to 4), items are viewer team, followed by a discussion of
generally considered the “gold standard” scored from 0 to 3. Other than items 1 the experience; and 3) administration of
for assessing depression for more than and 20 (which measure severity of de- the SCID MDD module and other study
four decades, has come under recent pressive symptoms), ratings are based instruments by the interviewers, under
scrutiny because it includes items with on frequency of symptoms. the supervision of the PI, followed by
poor inter-rater and retest reliability, and discussion of the experience and the so-
poor content validity, and fails to capture Zung Self-rating Depression Scale licitation of feedback.
depression according to modern classifica- (ZSDS). The ZSDS (11) is a widely used
tion systems such as the Diagnostic and self-rating instrument composed of 20 Subjects and procedures
Statistical Manual of Mental Disorders, 4th items. Items are scored from 1 to 4 ac-
edition (DSM-IV) (9). The authors refer to cording to the severity of the symptom. This study was conducted from 1999 to
two potential alternatives (the Inventory The sum of the item scores is then multi- 2001 in the Department of Adult and Geri-
for Depressive Symptomatology (IDS) and plied by 10 and divided by 8. The result- atric Psychiatry at the “Honorio Del-
the MADRS, which are designed to ad- ing total score ranges from 25 to 100. The gado–Hideyo Noguchi” National Institute
dress HAM-D deficiencies) but suggest Spanish-language version of the ZSDS of Mental Health (“HD-HN” NIMH),
that a better solution may be the develop- was validated in Spain (5) and standard- which is affiliated with Cayetano Heredia
ment of new instruments based on current ized in Peru (12). Peruvian University, and located in the
knowledge of depression, taking advan- northern area of the capital city of Lima in
tage of current psychometric and statisti- Major Depressive Disorder (MDD) Peru. The study protocol was reviewed
cal advances, since neither of these two module of the Structured Clinical Inter- and approved by the “HD-HN” NIMH in-
scales was developed using contemporary view for DSM-IV Disorders (SCID). stitutional board. All participants provided
psychometric methods (8). The MDD module of the SCID is a semi- written consent to enroll in the study.
The current study was carried out to structured diagnostic interview de- The study sample included 226 ambu-
meet the need for the development of a signed to assist clinicians, researchers, latory patients between the ages of 18
new scale for the assessment of depres- and trainees in making DSM-IV diagno- and 60 years, of any gender, who had
sion, in Spanish, taking account of Peru- sis of depression (13). come to “HD-HN” NIMH for their first
vian linguistic and cultural aspects and visit and clinical evaluation and were ca-
utilizing the standardized criteria of the Clinical Global Impression–Severity pable of consenting to participate in the
DSM-IV (including atypical and melan- scale (CGI-S). The CGI-S is a seven-point study. Those who had current or previ-
cholic features) (9) as well as the ICD-10 ously known psychosis and bipolar dis-
(International Classification of Diseases, 5 Sample copies available from the corresponding order (based on information obtained
10th revision) (10). author upon request. during the clinical interview or on prior

318 Rev Panam Salud Publica 30(4), 2011

Vega-Dienstmaier et al. • Depressive Psychopathology Scale: presentation and validation in Peru Original research

documented information) or were inca- Statistical analysis RESULTS

pable of providing consent due to ex-
treme psychiatric pathology or cognitive The evaluation parameters used in the Characteristics of the study subjects
deficits, and illiterates, were excluded. statistical analysis of the DPS included
the following: A total of 226 subjects were included
Subjects were evaluated using a two- in the study—94 male (41.6%) and 132
step assessment: • Feasibility (acceptability, quality of female (58.4%). The mean age was 28.49
data, ease of use, and time required years, with a standard deviation (SD) of
• Step 1. A team of interviewers (three for completion) 9.25, and the mean educational level was
psychiatry residents, one psychiatric 12.9 years (SD = 2.36). The marital status
nurse, and two psychologists) ex- • Internal structure, including: breakdown was 59.1% single, 30% mar-
plained the nature of the research to 䡩 Internal consistency (Cronbach’s ried or living together, 10% divorced or
prospective subjects and requested alpha coefficient) separated, and 0.9% widowed.
their written consent to participate in 䡩 Factor structure (extracted through Out of the 226 subjects, 218 com-
the study. Subjects who provided principal component analysis with pleted the DPS, 197 the ZSDS, and 224
written consent were asked to com- eigenvalues > 1 and rotated through the SCID MDD module. The diagnosis
plete/respond to the following instru- varimax with Kaiser’s normalization) resulting from the initial clinical evalu-
ments: 1) the DPS, and an evaluation • Convergent validity, based on Spear- ation was fully documented in 223
form for rating its ease of use; 2) the man’s correlations between: cases.
ZSDS; and 3) the SCID MDD module. According to the attending psychia-
䡩 DPS items and corresponding (simi-
Because the DPS and the ZSDS are trists, MDD was diagnosed in 109 pa-
lar) items in the ZSDS
self-rated, interviewer responsibilities tients. In this group of depressed patients,
䡩 DPS individual item and overall
for those instruments were limited to other DSM-IV Axis I (principal clinical
1) registering the time needed to com- scores and three different parame-
disorder) comorbid conditions included
plete the DPS and 2) verifying that all ters: 1) the overall ZSDS score, 2)
social phobia (16.5%), generalized anxiety
answerable items from both scales the attending psychiatrist’s rating of disorder (9.2%), dysthymic disorder
had been completed (unanswerable depression severity (one of five cat- (8.3%), panic disorder with agoraphobia
items—those for which, after addi- egories), and 3) the CGI-S score for (8.3%), obsessive-compulsive disorder
tional explanation, the subjects were depression severity (to further eval- (5.5%), panic disorder without agorapho-
unable to determine an appropriate uate convergence in terms of disor- bia (2.8%), bulimia nervosa (1.8%), anxi-
response—were left blank). der severity) ety disorder not otherwise specified
• Step 2. As part of the initial clinical • Discriminant validity, using two (0.9%), anorexia nervosa (0.9%), hypo-
evaluation, subjects were assigned to methods: chondriasis (1.8%), and other mental dis-
psychologists or residents working in 䡩 Comparison (through analysis of orders (11%).
the Diagnostic and Evaluation Unit of variance [ANOVA]) of mean overall Among the other 114 individuals with-
the “HD-HN” NIMH for intake as- DPS scores for 1) patients with and out MDD, the diagnosis included social
sessment. Later, each case was pre- without MDD (according to the SCID phobia (29.8%), dysthymic disorder (14%),
sented to one of eight attending psy- and the attending psychiatrist’s diag- generalized anxiety disorder (17.5%),
chiatrists. The attending psychiatrist nosis) and 2) subgroups of patients panic disorder with agoraphobia (14%)
conducted a face-to-face interview to with different severity of depressive and without agoraphobia (14%), obses-
1) corroborate the main diagnosis, 2) symptoms (based on both the CGI-S sive-compulsive disorder (7.9%), depres-
determine the presence or absence of score and the attending psychiatrist’s sive disorder not otherwise specified
MDD, and 3) rate the subject’s depres- rating) (7.9%), adjustment disorder (6.1%), anxi-
sion severity as one of five categories ety disorder not otherwise specified
䡩 Calculation, for the diagnosis of
(“no depressive symptoms”; “depres- (3.5%), bulimia nervosa (2.6%), hypo-
MDD (based on SCID criteria and on
sive disorder but no major depres- chondriasis (1.8%), anorexia nervosa
the attending psychiatrist’s evalua-
sion”; “major depression, mild”; (0.9%), dissociative disorder (0.9%), and
tion), of the area under receiver oper-
“major depression, moderate”; and other mental disorders (27.2%).
“major depression, severe”), based on ating characteristic curve (AUROC)
his/her clinical opinion, and as one of and sensitivity, specificity, number
Feasibility (acceptability, quality of
seven categories (described above) needed to diagnose (NND, 1/(sensi- data, ease of use, and time required
based on the CGI-S. tivity – (1 – specificity)) (15)), and pos- for completion)
itive and negative likelihood ratios
Neither the staff responsible for the in- (LR+ and LR–) for selected DPS cut- Four subjects (who met the inclusion cri-
take assessment nor the attending psy- off scores, and determination of the teria) refused to participate in the study.
chiatrists had access to the results ob- optimum cutoff score. The number of subjects who completed the
tained during Step 1 of the study (the DPS was significantly greater than the
administration of the DPS, the ZSDS, Statistical analyses were performed number of subjects who completed the
and the SCID MDD module) prior to using SPSS version 12 (IBM, Armonk, ZSDS (96.46% versus 87.17%; χ2 = 12.98, de-
their clinical evaluation. NY, USA). grees of freedom [df] = 1, P < 0.001). For the

Rev Panam Salud Publica 30(4), 2011 319

Original research Vega-Dienstmaier et al. • Depressive Psychopathology Scale: presentation and validation in Peru

DPS, two patients (0.9%) each failed to re- derstand (94.5%) and that the responses thinking”); 4) “insomnia” (“sleep distur-
spond to the “restlessness” and “early were easy to rate (90.3%). Most subjects bances” and “early morning awaken-
morning awakening” items, and one pa- (92.4%) reported that they would be will- ing”); and 5) “absence of positive affects”
tient (per item) (0.4%) failed to respond to ing to respond to the questionnaire again (“lack of optimism” and “anhedonia”)
the “psychomotor retardation,” “irritabil- during their next visit to “HD-HN” (the “anhedonia” item also showed a rel-
ity,” “difficulty with concentration/trouble NIMH. Negative feedback was minimal atively strong loading (≥ 0.40) on the
thinking,” “lack of optimism,” “indecisive- (7.1% said the questionnaire had too “uneasiness” factor) (Table 1). Three
ness,” “sadness,” “suicidal thinking,” many items, 3.3% said the items them- items (“appetite disturbances,” “depres-
“leaden paralysis,” “rejection sensitivity,” selves were too long, and 12.2% said too sion worse in the morning,” and “mood
“depression worse in the morning,” and much time was required to complete all reactivity”) showed low loadings (< 0.40)
“mood reactivity” items. For the remaining of the items). on all five factors. Spearman’s correla-
seven items, the data was complete (i.e., The mean time required by the sub- tions between “depression” and other
there were no missing responses). In com- jects to complete the DPS was 7.22 min- factors were as follows: 0.472 for “aner-
parison, for the ZSDS, 9 patients (4%) failed utes (SD = 3.99), with a median of 6 min- gia,” 0.541 for “uneasiness,” 0.294 for “in-
to complete the “fatigue” item; 8 (3.5%) utes, and 95% of subjects completed it in somnia,” and 0.464 for “absence of posi-
skipped the “insomnia” and “suicidal less than 15 minutes. tive affects” (all P < 0.01).
thinking” items; 7 (3.1%) skipped “depres-
sion worse in the morning,” “crying Factor structure Internal consistency
spells,” “weight loss,” and “palpitations”;
6 (2.7%) skipped “gastrointestinal symp- Based on principal component analy- The internal consistency analysis of
toms,” “hopelessness,” “indecisiveness,” sis, the DPS comprises five main factors: the 20 DPS items yielded an overall
and “feelings of emptiness”; 5 (2.2%) 1) “depression” (“sadness,” “feelings of Cronbach’s alpha value of 0.86. In com-
skipped “low appetite,” “loss of libido,” guilt,” “irritability,” “rejection sensitiv- parison, the 20 ZSDS items yielded an
“nervousness,” “irritability,” “low self- ity,” “low self-esteem,” and “suicidal overall alpha value of 0.83 for the same
esteem,” and “difficulty doing usual thinking”); 2) “anergia” (“hypersomnia,” sample. The alpha values for the five fac-
things”; and 4 (1.8%) skipped “sadness,” “fatigue,” “leaden paralysis,” and “psy- tors were as follows: “depression,” 0.76;
“trouble thinking,” and “anhedonia.” chomotor retardation”); 3) “uneasiness” “anergia,” 0.702; “uneasiness,” 0.63; “in-
The vast majority of subjects reported (“restlessness,” “indecisiveness,” and somnia,” 0.681; and “absence of positive
that the DPS items seemed easy to un- “difficulty with concentration/trouble affects,” 0.538.

TABLE 1. Depressive Psychopathology Scale (DPS): five factorsa and their loading scores for 20 items in study of
Peruvian psychiatric outpatients (n = 218), Lima, Peru, 1999–2001

Factor loadings
Absence of
DPS item Depression Anergia Uneasiness Insomnia positive affects

1. Appetite disturbances 0.28 0.26 –b – –

2. Sleep disturbances – – 0.29 0.79 –
3. Early morning awakening – – – 0.81 –
4. Hypersomnia – 0.59 – – –
5. Fatigue 0.26 0.70 – 0.22 –
6. Leaden paralysis – 0.82 – – –
7. Anhedonia 0.27 0.21 0.44 – 0.46
8. Restlessness 0.26 – 0.47 0.24 –
9. Psychomotor retardation – 0.61 0.33 – –
10. Rejection sensitivity 0.53 – 0.35 – –
11. Irritability 0.73 – – – –
12. Difficulty with concentration/
trouble thinking – – 0.81 – –
13. Lack of optimism – – – – 0.85
14. Indecisiveness 0.25 – 0.71 – –
15. Sadness 0.58 – 0.36 0.26 –
16. Depression worse in the
morning – – – – –
17. Mood reactivity 0.23 0.29 0.21 – 0.22
18. Feelings of guilt 0.67 – – – –
19. Low self-esteem 0.59 0.31 0.29 – –
20. Suicidal thinking 0.58 – – – –
% of variance 14.12 12.13 11.50 9.08 6.15
a Extracted through principal component analysis and rotated using varimax with Kaiserʼs normalization. Kaiser-Meyer-Olkin (KMO) measure of sampling
adequacy = 0.857. P < 0.001 based on Bartlettʼs test of sphericity.
b Loadings ≤ 0.2 were eliminated from the table.

320 Rev Panam Salud Publica 30(4), 2011

Vega-Dienstmaier et al. • Depressive Psychopathology Scale: presentation and validation in Peru Original research

TABLE 2. Convergent validity of depression screening scales: correlations between items from the Depressive Psy-
chopathology Scale (DPS) and the Zung Self-rating Depression Scale (ZSDS) in study of Peruvian psychiatric outpa-
tients, Lima, Peru, 1999–2001

Spearmanʼs
DPS item Corresponding ZSDS item correlation valuea n

1. Appetite disturbances 5. Low appetite 0.297 221

7. Weight loss 0.218 219
2. Sleep disturbances 4. Insomnia 0.677 218
3. Early morning awakening 4. Insomnia 0.491 216
4. Hypersomnia NAb
5. Fatigue 10. Fatigue 0.566 217
12. Difficulty doing usual things 0.459 221
6. Leaden paralysis NA
7. Anhedonia 6. Loss of libido 0.087c 221
20. Anhedonia 0.323 222
8. Restlessness 13. Restlessness 0.427 219
9. Psychomotor retardation NA
10. Rejection sensitivity NA
11. Irritability 15. Irritability 0.576 220
12. Difficulty with 11. Trouble thinking 0.435 221
concentration/
trouble thinking
13. Lack of optimism 14. Hopelessness 0.439 219
14. Indecisiveness 16. Indecisiveness 0.379 219
15. Sadness 1. Sadness 0.602 221
3. Crying spells 0.497 218
16. Depression worse in 2. Depression worse in the morning 0.123c 218
the morning
17. Mood reactivity NA
18. Feelings of guilt NA
19. Low self-esteem 17. Low self-esteem 0.533 221
18. Feelings of emptiness 0.381
20. Suicidal thinking 19. Suicidal thinking 0.454 217
8. Constipation
9. Palpitations
a P < 0.01 (unless otherwise indicated).
b NA: not applicable.
c P > 0.05.

Convergent validity item and the other parameters of depres- found across subgroups for depression
sion are shown in Table 3. severity (P < 0.001).
Six DPS items and two ZSDS items ROC analysis and psychometric mea-
had no corresponding items in the oppo- Discriminant validity sures (sensitivity, specificity, NND, LR+,
site scale. Correlations between corre- and LR–) for selected DPS cutoff scores
sponding items across both scales are Table 4 shows mean DPS scores classi- for MDD diagnosis (according to the
shown in Table 2. Statistically significant fied by groups based on the presence/ ab- SCID and the attending psychiatrist’s
correlations (P < 0.01) were found for all sence of depression (according to the evaluation) are shown in Tables 5 and 6,
pairs of corresponding items except two: SCID and the attending psychiatrist’s respectively.
“depression worse in the morning” evaluation) and the severity of depression When the SCID MDD module was
(both scales), and “anhedonia” (DPS) (based on the CGI-S score and the attend- used as the gold standard, the AUROC
and “loss of libido” (ZSDS). ing psychiatrist’s rating). Mean scores for was 0.872 (standard error [SE] = 0.024);
The overall DPS score was signifi- the ZSDS for the above-mentioned groups the lowest NND (1.54) corresponded to a
cantly correlated with the following pa- are also shown for comparison. As shown, cutoff score of 28/29, with a sensitivity
rameters: the overall ZSDS score (rho = mean DPS scores are significantly higher of 76.92% and specificity of 88%; and
0.804, P < 0.001); the CGI-S score for de- among those classified as depressed ver- equilibrium between sensitivity (81.32%)
pression severity (rho = 0.621, P < 0.001); sus those classified as nondepressed (ac- and specificity (80%) was found for a
and the attending psychiatrist’s rating of cording to the SCID and the attending cutoff score of 26/27. When using the at-
depression severity (rho = 0.589, P < psychiatrist’s evaluation) (P < 0.001), and tending psychiatrist’s diagnosis as the
0.001). Correlations between each DPS a significant statistical difference was gold standard, the AUROC was 0.832

Rev Panam Salud Publica 30(4), 2011 321

Original research Vega-Dienstmaier et al. • Depressive Psychopathology Scale: presentation and validation in Peru

TABLE 3. Convergent validity: correlations (Spearman’s rho)a between DPSb individual item and overall scores and three
other measures of depression (overall ZSDSc score, CGI-Sd score, and attending psychiatrist’s rating of depression
severity) in study of Peruvian psychiatric outpatients, Lima, Peru, 1999–2001

Psychiatristʼs rating
DPS item ZSDS CGI-S of depression severity

1. Appetite disturbances 0.372 0.284 0.289

(n = 197) (n = 214) (n = 216)
2. Sleep disturbances 0.545 0.343 0.399
(n = 197) (n = 214) (n = 216)
3. Early morning awakening 0.404 0.328 0.347
(n = 196) (n = 212) (n = 214)
4. Hypersomnia 0.126e 0.018e –0.008e
(n = 197) (n = 214) (n = 216)
5. Fatigue 0.584 0.505 0.490
(n = 197) (n = 214) (n = 216)
6. Leaden paralysis 0.423 0.430 0.392
(n = 197) (n = 213) (n = 215)
7. Anhedonia 0.473 0.377 0.413
(n = 197) (n = 214) (n = 216)
8. Restlessness 0.346 0.221f 0.200f
(n = 195) (n = 212) (n = 214)
9. Psychomotor retardation 0.470 0.416 0.361
(n = 196) (n = 213) (n = 215)
10. Rejection sensitivity 0.364 0.306 0.253
(n = 196) (n = 213) (n = 215)
11. Irritability 0.403 0.332 0.281
(n = 196) (n = 213) (n = 215)
12. Difficulty with concentration/trouble thinking 0.366 0.246 0.210f
(n = 196) (n = 213) (n = 215)
13. Lack of optimism 0.453 0.297 0.277
(n = 196) (n = 213) (n = 215)
14. Indecisiveness 0.511 0.331 0.311
(n = 196) (n = 213) (n = 215)
15. Sadness 0.534 0.463 0.411
(n = 196) (n = 213) (n = 215)
16. Depression worse in the morning 0.057e 0.005e 0.022e
(n = 197) (n = 213) (n = 215)
17. Mood reactivity 0.537 0.364 0.424
(n = 196) (n = 213) (n = 215)
18. Feelings of guilt 0.371 0.286 0.288
(n = 197) (n = 214) (n = 216)
19. Low self-esteem 0.618 0.501 0.426
(n = 197) (n = 214) (n = 216)
20. Suicidal thinking 0.452 0.306 0.319
(n = 197) (n = 213) (n = 215)
Overall DPS score 0.804 0.621 0.589
(n = 193) (n = 206) (n = 208)
a P < 0.001 (unless otherwise indicated).
b DPS: Depressive Psychopathology Scale.
c ZSDS: Zung Self-rating Depression Scale.
d CGI-S: Clinical Global Impression–Severity scale.
e P > 0.05.
f P < 0.005.

(SE = 0.028) and the lowest NND (1.82) known as the DPS adequately detects pared with curves found for other scales
corresponded to a cutoff score of 30/31. MDD, has a good convergent validity, (6, 16, 17). A cutoff score of 26/27, with a
Equilibrium between sensitivity (77.67%) and is easy to use. sensitivity and specificity of at least 80%
and specificity (72.32%) was found for a ROC analysis supports the discrimi- for MDD (according to the SCID), is
cutoff score of 25/26. nant validity of the DPS, demonstrating proposed.
its ability to differentiate individuals The convergent validity of the DPS was
DISCUSSION with MDD from those without it (ac- demonstrated based principally on its
cording to the SCID and the attending strong correlation (0.8) with the ZSDS—a
According to the results of the current psychiatrist’s evaluation). In addition, widespread psychometric measure for de-
study, the newly developed instrument the DPS AUROC (> 0.8) is good com- pression. Correlations between the DPS

322 Rev Panam Salud Publica 30(4), 2011

Vega-Dienstmaier et al. • Depressive Psychopathology Scale: presentation and validation in Peru Original research

TABLE 4. Discriminant validity: analyses of mean overall DPSa and ZSDSb scores for diagnosis (presence or absence of depression) and
severity criteria, based on the SCID,c the attending psychiatrist’s evaluation, and the CGI-S,d in study of Peruvian psychiatric outpatients,
Lima, Peru, 1999–2001

DPS ZSDS
Diagnosis/severity criteria Category n score (SDe) n score (SD)

Major Depressive Disorder (MDD)

SCID Present 91 32.91 (7.40) 81 69.43 (8.95)
Absent 125 20.43 (8.17) 116 55.58 (10.14)
F = 132.88, P < 0.001 F = 97.81, P < 0.001
Attending psychiatristʼs Present 103 31.57 (8.74) 91 67.79 (10.19)
evaluation Absent 112 20.36 (7.94) 103 55.49 (10.24)
F = 97.2, P < 0.001 F = 70.13, P < 0.001
Severity of depression
Attending psychiatristʼs No depression 30 16.70 (8.30) 26 50.77 (9.77)
evaluation Other depressive disorder 72 22.11 (6.89) 67 56.74 (8.93)
MDD, mild 14 25.64 (8.44) 16 64.69 (6.65)
MDD, moderate 87 31.43 (9.00) 75 68.15 (10.64)
MDD, severe 5 39.00 (7.91) 5 76.00 (5.55)
F = 26.63, P < 0.001 F = 24.74, P < 0.001
CGI-S 1 = Not at all depressed 28 16.68 (8.44) 24 51.51 (9.60)
2 = Borderline depressed 27 18.85 (6.85) 26 53.46 (9.38)
3 = Mildly depressed 40 22.48 (6.41) 39 58.43 (8.43)
4 = Moderately depressed 70 28.97 (8.15) 64 65.02 (9.32)
5 = Markedly depressed 32 34.00 (9.26) 25 71.25 (12.69)
6 = Severely depressed 9 35.00 (8.69) 9 71.94 (7.63)
7 = Extremely depressed NAf NA NA NA
F = 24.44, P < 0.001 F = 18.48, P < 0.001
a DPS: Depressive Psychopathology Scale.
b ZSDS: Zung Self-rating Depression Scale.
c SCID: Structured Clinical Interview for DSM-IV Disorders, Major Depression Disorder (MDD) module.
d CGI-S: Clinical Global Impression–Severity scale.
e SD: standard deviation.
f NA: not applicable.

and both the CGI-S score (0.621) and the “uneasiness,” “insomnia,” and “absence fulness”) (17). Review of other factors re-
attending psychiatrist’s rating of depres- of positive affects.” ported for the above-mentioned scales
sion severity (0.589) were also strong (and Review of the literature revealed that, showed similarities with other DPS fac-
statistically significant). in some cases, other established depres- tors. For example, the DPS “uneasiness”
Furthermore, the vast majority of sub- sion instruments have up to five dimen- factor is similar to the “anxiety” factor in
jects had no difficulty completing the sions (e.g., the HAM-D (18)), similar to the SCAN (23), HAM-D (18), and ZSDS
DPS. Acceptability of the instrument the multifactor structure found in the (22), and the DPS “absence of positive
was also suggested by the fact that only DPS. In addition, symptoms identical or affects” factor is similar to both the
a minimal number of subjects declined similar to those of the core “depression” EURO-D “factor 4” (17) and the CES-D
to participate in the study, more than factor in the DPS (“sadness,” “feelings of “positive affect” factor (16, 19–21). Other
96% of subjects completed all items, and guilt,” “low self-esteem,” and “suicidal corresponding factors include the DPS
a significantly higher number of subjects thinking”) can be found in other instru- “anergia” factor and the “problems initi-
responded to the DPS versus the widely ments such as the CES-D (“sadness,” “de- ating behaviors” factor of the CES-D (16),
used ZSDS. In addition, the time re- pressive mood,” “could not get going,” and the DPS “insomnia” factor and the
quired to complete the instrument (7 “life had been a failure,” and “crying” (16, “sleep disturbance” factor of the HAM-D
minutes on average) seems rather brief 19–21); the ZSDS (“depressed mood,” (18).
when compared with other widely used “crying,” “personal devaluation,” and Five DPS items were problematic in
depression scales such as the CES-D, “suicidal rumination”) (22); the HAM-D terms of correlation with the depression
which requires 5 minutes; the BDI, (“depressed mood,” “feelings of guilt,” parameters and/or factor structure: 1)
which takes 5–10 minutes; and the “suicidality,” and “worthlessness”) (18); “depression worse in the morning”
MADRS, the HAM-D, and the IDS, the Schedules for Clinical Assessment in (melancholic symptom), which had weak
which take 15 minutes, 15–20 minutes, Neuropsychiatry (SCAN) (“depressed loadings on all factors and failed to sig-
and 15–30 minutes, respectively (11). mood,” “loss of self-esteem,” and “suici- nificantly correlate with the correspond-
Extracting factors through principal dality”) (23); and the European screening ing item in the ZSDS and other parame-
component analysis showed that most instrument for depression in the elderly, ters of depression, such as the overall
items could be reasonably grouped into known as the EURO-D (“depressed ZSDS score, the CGI-S score, and the at-
five domains: “depression,” “anergia,” mood,” “suicidality,” “guilt,” and “tear- tending psychiatrist’s rating of depres-

Rev Panam Salud Publica 30(4), 2011 323

Original research Vega-Dienstmaier et al. • Depressive Psychopathology Scale: presentation and validation in Peru

TABLE 5. Discriminant validity: ROCa analysis and psychometric measures (sensitivity, specificity, NND,b and
LR+/LR–c) for selected DPSd cutoff scores for diagnosis of MDDe (based on SCIDf criteria) in study of Peruvian psy-
chiatric outpatients,g Lima, Peru, 1999–2001

Cutoff
score Sensitivity 1 – specificity Specificity NND LR+ LR–

2/3 1.0000 0.9760 0.0240 41.6667 1.0246 0.0000

4/5 1.0000 0.9680 0.0320 31.2500 1.0331 0.0000

10 1.0000 0.8960 0.1040 9.6154 1.1161 0.0000

11/12 1.0000 0.8720 0.1280 7.8125 1.1468 0.0000

15/16 1.0000 0.7200 0.2800 3.5714 1.3889 0.0000

16/17 0.9890 0.6800 0.3200 3.2361 1.4544 0.0343

19/20 0.9560 0.5520 0.4480 2.4750 1.7320 0.0981

20/21 0.9451 0.4960 0.5040 2.2269 1.9054 0.1090

24/25 0.8681 0.3120 0.6880 1.7981 2.7825 0.1917

25/26 0.8352 0.2800 0.7200 1.8013 2.9827 0.2289
26/27 0.8132 0.2000 0.8000 1.6308 4.0659 0.2335
27/28 0.7912 0.1440 0.8560 1.5451 5.4945 0.2439
28/29 0.7692 0.1200 0.8800 1.5403 6.4103 0.2622
29/30 0.7363 0.1120 0.8880 1.6019 6.5738 0.2970
30/31 0.6813 0.0960 0.9040 1.7085 7.0971 0.3525

34/35 0.3626 0.0480 0.9520 3.1783 7.5549 0.6695

35/36 0.3407 0.0320 0.9680 3.2398 10.6456 0.6811

39/40 0.1868 0.0080 0.9920 5.5924 23.3516 0.8197

40/41 0.1648 0.0080 0.9920 6.3761 20.6044 0.8419

44/45 0.0659 0.0080 0.9920 17.2610 8.2418 0.9416

45/46 0.0549 0.0000 1.0000 18.2000 –h 0.9451
a ROC: receiver-operating characteristic.
b NND: number needed to diagnose.
c LR+/LR–: positive/negative likelihood ratios.
d DPS: Depressive Psychopathology Scale.
e MDD: Major Depressive Disorder.
f SCID: Structured Clinical Interview for DSM-IV Disorders, Major Depression Disorder (MDD) module.
g 91 with MDD and 125 without MDD (with 10 excluded because of missing values).
h Non-calculable.

sion severity; 2) “hypersomnia” (atypical or modified in a revised version of the is highly acceptable, efficient (requiring
feature), which did not correlate signifi- instrument. only about 7 minutes to complete), and
cantly with any of the depression para- One potential limitation of the current easy to use. However, before the scale can
meters; 3) “mood reactivity” and 4) “ap- study is the fact that nondepressed sub- be recommended for clinical use, further
petite disturbances,” which showed low jects participating in it had other condi- studies are needed to improve and sim-
factor loadings based on principal com- tions, such as anxiety disorders and other plify it (e.g., eliminating items with
ponent analysis; and 5) “anhedonia,” psychopathology, that could have ele- deficits related to either the factor struc-
which had moderate loadings on two vated the cutoff scores for discriminating ture or the correlation with depression
distinct factors and a nonsignificant cor- between depressed and nondepressed parameters). Psychometric assessment of
relation with the ZSDS’ corresponding patients and thus resulted in an underes- the DPS using statistical methods based
item “loss of libido.” The low loadings of timation of sensitivity and specificity in on Item Response Theory is another rec-
the “appetite disturbances” item may be relation to other studies comparing de- ommended area of research.
attributable to its measurement of both pressed subjects with non-psychiatrically
increased and decreased appetite and the ill individuals. In addition, because psy- Acknowledgments. The authors ex-
fact that its rating scale differs from that chotic patients were excluded from the press their gratitude for the contribu-
of most other items. study, it was not possible to evaluate the tions of the following professionals: An-
Further studies using this new scale ability of the DPS to discriminate be- drea Rubini, Javier Saavedra, Bernardo
could help determine if the five items tween depression and psychotic disor- Guimas, Héctor Vidal, Freddy Vásquez,
listed above contribute to the measure- ders such as schizophrenia. José Carlos San Martín, Abel Sagástegui,
ment of depression in this population, This study suggests the DPS has strong Guido Mazzotti, Ricardo Chirinos, and
and if any of them should be eliminated convergent and discriminant validity and María del Carmen Díaz.

324 Rev Panam Salud Publica 30(4), 2011

Vega-Dienstmaier et al. • Depressive Psychopathology Scale: presentation and validation in Peru Original research

TABLE 6. Discriminant validity: ROCa analysis and psychometric measures (sensitivity, specificity, NND,b and
LR+/LR–c) for selected DPSd cutoff scores for diagnosis of MDDe (based on the attending psychiatrist’s evaluation)
in study of Peruvian psychiatric outpatients,f Lima, Peru, 1999–2001

Cutoff
score Sensitivity 1 – specificity Specificity NND LR+ LR–

1 1.0000 1.0000 0.0000 –g 1.0000 –

2/3 1.0000 0.9732 0.0268 37.3333 1.0275 0.0000
4/5 1.0000 0.9643 0.0357 28.0000 1.0370 0.0000

10 0.9806 0.9018 0.0982 12.6909 1.0874 0.1977

11/12 0.9806 0.8750 0.1250 9.4713 1.1207 0.1553

14/15 0.9806 0.7768 0.2232 4.9068 1.2624 0.0870

15/16 0.9612 0.7232 0.2768 4.2026 1.3290 0.1403

19/20 0.8932 0.5625 0.4375 3.0239 1.5879 0.2441

20/21 0.8738 0.5089 0.4911 2.7408 1.7169 0.2570

23/24 0.8155 0.3661 0.6339 2.2249 2.2278 0.2910

24/25 0.8058 0.3036 0.6964 1.9910 2.6545 0.2788
25/26 0.7767 0.2768 0.7232 2.0003 2.8061 0.3088
26/27 0.7184 0.2232 0.7768 2.0193 3.2186 0.3625
27/28 0.6990 0.1607 0.8393 1.8576 4.3495 0.3586
28/29 0.6796 0.1339 0.8661 1.8326 5.0744 0.3699
29/30 0.6505 0.1250 0.8750 1.9030 5.2039 0.3994
30/31 0.6311 0.0804 0.9196 1.8158 7.8533 0.4012
31/32 0.5340 0.0536 0.9464 2.0816 9.9676 0.4924
32/33 0.4757 0.0446 0.9554 2.3197 10.6563 0.5488

35/36 0.3204 0.0268 0.9732 3.4060 11.9612 0.6983

36/37 0.2816 0.0268 0.9732 3.9251 10.5113 0.7382

39/40 0.1650 0.0089 0.9911 6.4053 18.4854 0.8425

40/41 0.1456 0.0089 0.9911 7.3152 16.3107 0.8621
41/42 0.1262 0.0000 1.0000 7.9231 – 0.8738
a ROC: receiver-operating characteristic.
b NND: number needed to diagnose.
c LR+/LR–: positive/negative likelihood ratios.
d DPS: Depressive Psychopathology Scale.
e MDD: Major Depressive Disorder.
f 103 with MDD and 112 without MDD (with 11 excluded because of missing values).
g Non-calculable.

REFERENCES

1. Ustün TB, Ayuso-Mateos JL, Chatterji S, 4. Wiebe JS, Penley JA. A psychometric compar- 8. Bagby RM, Ryder AG, Schuller DR, Marshall
Mathers C, Murray CJ. Global burden of ison of the Beck Depression Inventory-II in MB. The Hamilton Depression Rating Scale:
depressive disorders in the year 2000. Br J English and Spanish. Psychol Assess. 2005; has the gold standard become a lead weight?
Psychiatry. 2004;184:386–92. 17(4):481–5. Am J Psychiatry. 2004;161(12):2163–77.
2. Bobes J, Bulbena A, Luque A, Dal-Ré R, 5. Aragonès Benaiges E, Masdèu Montalà RM, 9. American Psychiatric Association. Diagnostic
Ballesteros J, Ibarra N; Grupo de Validación Cando Guasch G, Coll Borràs G. Validez di- and Statistical Manual of Mental Disorders.
en Español de Escalas Psicométricas. Eva- agnóstica de la Self-Rating Depression Scale 4th ed. Washington: American Psychiatric
luación psicométrica comparativa de las de Zung en pacientes de atención primaria. Association; 1994.
versiones en español de 6, 17 y 21 ítems de la Actas Esp Psiquiatr. 2001;29(5):310–6. 10. World Health Organization. The ICD-10 clas-
Escala de valoración de Hamilton para la 6. Camacho PA, Rueda-Jaimes GE, Latorre JF, sification of mental and behavioral disorders:
evaluación de la depresión. Med Clin (Barc). Navarro-Mancilla AA, Escobar M, Franco JA. clinical descriptions and diagnostic guide-
2003;120(18):693–700. Validez y confiabilidad de la escala del Cen- lines. Geneva: WHO; 1992.
3. Lobo A, Chamorro L, Luque A, Dal-Ré R, ter for Epidemiologic Studies–Depression 11. Yonkers KA, Samson J. Mood disorders mea-
Badia X, Baró E; Grupo de Validación en en estudiantes adolescentes de Colombia. sures. In: Rush AJ Jr, Pincus HA, First MB,
Español de Escalas Psicométricas. Validación Biomedica. 2009;29(2):260–9. Blacker D, Endicott J, Keith SJ, et al., editors.
de las versiones en español de la Montgomery- 7. Reuland DS, Cherrington A, Watkins GS, Handbook of psychiatric measures. Washing-
Asberg Depression Rating Scale y la Hamil- Bradford DW, Blanco RA, Gaynes BN. Diag- ton: American Psychiatric Association; 2000.
ton Anxiety Rating Scale para la evaluación nostic accuracy of Spanish language depres- 12. Novara J, Sotillo C, Warthon D. Estan-
de la depresión y de la ansiedad. Med Clin sion-screening instruments. Ann Fam Med. darización y correlación de las escalas de
(Barc). 2002;118(13):493–9. 2009;7(5):455–62. Beck, Hamilton y Zung para depresión en

Rev Panam Salud Publica 30(4), 2011 325

Original research Vega-Dienstmaier et al. • Depressive Psychopathology Scale: presentation and validation in Peru

población de Lima Metropolitana. Lima: Ins- 17. Jirapramukpitak T, Darawuttimaprakorn N, 21. Makambi KH, Williams CD, Taylor TR,
tituto Nacional de Salud Mental “Honorio Punpuing S, Abas M. Validation and factor Rosenberg L, Adams-Campbell LL. An as-
Delgado-Hideyo Noguchi”; 1985. structure of the Thai version of the EURO-D sessment of the CES-D scale factor structure
13. First MB, Spitzer RL, Gibbon M, Williams scale for depression among older psychiatric in black women: the Black Women’s Health
JBW. Structured Clinical Interview for DSM- patients. Aging Ment Health. 2009;13(6): Study. Psychiatry Res. 2009;168(2):163–70.
IV Axis I Disorders, Clinician Version (SCID- 899–904. 22. Romera I, Delgado-Cohen H, Perez T,
CV). Washington: American Psychiatric Press; 18. Milak MS, Parsey RV, Keilp J, Oquendo MA, Caballero L, Gilaberte I. Factor analysis of the
1996. Malone KM, Mann JJ. Neuroanatomic corre- Zung self-rating depression scale in a large
14. Guy W. Clinical Global Impressions (CGI) lates of psychopathologic components of ma- sample of patients with major depressive dis-
Scale. In: Rush AJ Jr, Pincus HA, First MB, jor depressive disorder. Arch Gen Psychiatry. order in primary care. BMC Psychiatry.
Blacker D, Endicott J, Keith SJ, et al., editors. 2005;62(4):397–408. 2008;8(1):4.
Handbook of psychiatric measures. Washing- 19. Villalobos Galvis FH. Estructura factorial y 23. Korszun A, Moskvina V, Brewster S, Craddock
ton: American Psychiatric Association; 2000. propiedades psicométricas de una escala de N, Ferrero F, Gill M, et al. Familiality of
15. How good is that test? II [Internet]. Bandolier. depresión en universitarios de Colombia. Rev symptom dimensions in depression. Arch
1996;3(5):1–3. Available from: http://www. Panam Salud Publica. 2010;27(2):110–6. Gen Psychiatry. 2004;61(5):468–74.
medicine. ox.ac.uk/bandolier/band27/b27-2. 20. Canady RB, Stommel M, Holzman C. Mea-
html Accessed 28 July 2010. surement properties of the centers for epi-
16. Batistoni SS, Neri AL, Cupertino AP. Validade demiological studies depression scale (CES-
da escala de depressão do Center for Epidemi- D) in a sample of African American and
ological Studies entre idosos brasileiros. Rev non-Hispanic White pregnant women. J Nurs Manuscript received on 21 August 2010. Revised version
Saude Publica. 2007;41(4):598–605. Meas. 2009;17(2):91–104. accepted for publication on 20 April 2011.

RESUMEN Objetivo. Presentación y validación de la Escala de Psicopatología Depresiva (EPD),

un nuevo inventario psicométrico en lengua española, en una muestra de pacientes
psiquiátricos peruanos.
La Escala de Psicopatología Métodos. De 1999 al 2001, se administraron la EPD, la Escala de Autoevaluación de
Depresiva: presentación y la Depresión de Zung (ZSDS) y el módulo correspondiente al Trastorno Depresivo
validación inicial en una Mayor (TDM) de la Entrevista Clínica Estructurada para trastornos del DSM-IV
(SCID) a 226 pacientes ambulatorios no psicóticos derivados al Instituto Nacional de
muestra de pacientes Salud Mental de Lima para su evaluación clínica inicial. En la evaluación, los psi-
psiquiátricos peruanos quiatras responsables 1) corroboraron el diagnóstico general y la presencia o ausencia
de TDM, y 2) clasificaron la gravedad de la depresión con base en la opinión clínica y
los criterios de gravedad de la Escala de Impresión Clínica Global.
Resultados. El tiempo medio para completar la EPD fue de 7,22 minutos (desviación
estándar, 3,99). El valor del índice alfa de Cronbach fue de 0,86. Para el diagnóstico
del TDM, basado en el SCID, el área bajo la curva ROC (de eficacia diagnóstica) fue
de 0,872, y la puntuación discriminatoria seleccionada (26/27) mostró una sensibili-
dad de 81,32% y una especificidad de 80%; con base en la evaluación de los psiquia-
tras responsables, el área bajo la curva ROC fue de 0,832, y la puntuación discrimina-
toria seleccionada (25/26) mostró una sensibilidad de 77,67% y una especificidad de
72,32%. La EPD se correlacionó significativamente con la ZSDS (rho = 0,8, P < 0,001).
Algunos elementos de la EPD (“depresión peor por la mañana”, “alteraciones del ape-
tito”, “reactividad del estado de ánimo” e “hipersomnia”) mostraron cargas bajas de
los cinco factores extraídos mediante el análisis de los componentes principales o no
se correlacionaron significativamente con los parámetros de depresión.
Conclusiones. La EPD puede predecir el TDM y tiene validez convergente, tal como
lo demuestra su correlación con la ZSDS. Sin embargo, se recomienda realizar estu-
dios psicométricos adicionales con objeto de simplificarla y mejorarla.

Palabras clave Escalas de valoración psiquiátrica; depresión; estudios de validación; Perú.

326 Rev Panam Salud Publica 30(4), 2011