Research report

Adaptation and validation of the depression, anxiety and stress scale(DASS) to Brazilian Portuguese

Rose Claudia Batistelli Vignola a, Adriana Marcassa Tucci b,n

a Federal University of São Paulo, Brazilb Department of Health, Education and Society, Federal University of São Paulo, Brazil

a r t i c l e i n f o

Article history:Received 24 September 2013Accepted 22 October 2013Available online 28 October 2013

Keywords:DepressionAnxietyStressScalesValidity

a b s t r a c t

Background: Depression and anxiety have been associated with a range of symptoms that often overlap.Depression, Anxiety and Stress Scale‐21 (DASS-21) is a single instrument to assess symptoms ofdepression, anxiety and stress. This study aimed to adapt and validate the DASS-21 for use in theBrazilian Portuguese language.Methods: The DASS-21 has been adapted following the translation–back translation methodology fromEnglish to Portuguese. 242 subjects completed the following assessments: the DASS-21, the BeckDepression Index (BDI), Beck Anxiety Index (BAI) and the Inventory of Stress Symptoms of Lipp (ISSL).Results: The Kaiser–Meyer–Olkin (KMO) result was .949, indicating that the adequacy of the model washigh. Cronbach's alpha was .92 for the depression, .90 for the stress, and .86 for the anxiety, indicating agood internal consistency for each subscale. The correlations between DASS scale and BDI scale, BAI scaleand ISSL inventory were strong. The factorial analysis and distribution of factors among the subscalesindicated that the structure of three distinct factors is adequate.Limitations: Older subjects over 65 years of age were not largely represented in this sample. A studyspecific to this elderly population should be conducted. Another limitation of the study was educationlevel. The impact of low education in its applicability should be considered.Conclusions: The findings support the validity of the Brazilian Portuguese version of the DASS-21 and addto the evidence of the DASS-21 quality and ability to assess emotional states separately, eliminating theuse of different instruments to assess these states.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

The diagnosis of depression or anxiety is associated with arange of symptoms and organic diseases that often overlap (Clarkand Watson, 1991; den Hollander-Gijsman et al., 2012). In addic-tion, stress has been demonstrated to be a risk factor for thedevelopment of anxiety and depression, affecting quality of lifeand hampering productivity (Kehne, 2007; Kehne and Cain, 2010).

Anxiety and depression may be described as different points onthe same “continuum”, different manifestations of the same pathol-ogy, different syndromes associated with other mood disorders,distinct phenomena which may develop from one to the other overa period of time, or as different pathologies (Clark andWatson, 1991).

Clark and Watson (1991) proposed the Tripartite Model ofAnxiety and Depression that considers the following factors:negative affect (NA), which is present in both anxiety and depres-sion; reduced positive affect (PA), which is common in depression;

and physiological hyperactivation (FH), which is common inanxiety. Issues associated with mood and affective disorders oftenencompass the aspect of subjective wellbeing (or lack thereof),such as anxiety and depression. In this sense, affectivity is relatedto wellbeing, psychological distress and suffering that may or maynot be experienced by the individual (Clark and Watson, 1991).

The conceptual basis of the Depression, Anxiety, and StressScale (DASS) was based on the tripartite model of anxiety anddepression. This model proposes that the disorder of affect (and itssubtypes) is a continuum between depression, anxiety and stress.Depression is characterized by low positive affect, hopelessness,low self-esteem and low encouragement. Anxiety is associatedwith physiological hyperstimulation and stress resulting frompersistent tension, irritability and a low threshold for frustrationor disruption (Apóstolo, 2010). Thus, depression, anxiety and stresshave common features, including negative affect, emotional dis-tress and physiological changes in the hypothalamic-pituitary-adrenal axis (HPA) (Mello et al., 2007).

Although there are many definitions of stress, the concept ofstress has conventionally been defined as an emotional experienceaccompanied by biochemical, physiological, cognitive and beha-vioral aspects. Originally, the concept of stress arose from the

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/jad

Journal of Affective Disorders

0165-0327/$ - see front matter & 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.jad.2013.10.031

n Correspondence to: Departamento de Saúde, Educação e Sociedade, Universi-dade Federal de São Paulo, Rua Silva Jardim, 136, Vila Mathias, Santos/SP,CEP 11015-020, Brazil. Tel.: þ55 13 3878 3700.

E-mail addresses: atucci@unifesp.br, adritucci@uol.com.br (A.M. Tucci).

Journal of Affective Disorders 155 (2014) 104–109

observation that different types of physical or psychologicalconditions threatened homeostasis and initiated a cascade ofphysiological reactions, known as the “general adaptation syn-drome” (Selye, 1936).

The present study examined the four phases of stress that weredefined by Lipp (2005) and were expanded from a three phasesmodel developed by Selye (1936). The stress phases proposed byLipp (2005) include alertness, resistance, exhaustion and near-exhaustion. This model defines stress as “a complex reaction andoverall body components involving physical, psychological, mentaland hormonal aspects, which develops in stages or phases”.

For the DASS, stress is defined as an emotional state that variesaccording to an individual evaluation of the situations experiencedas a threat, harm or challenge. The DASS requires an assessment ofdemand and a secondary assessment of coping, with the symp-toms and emotions serving as the main concept organizer of thestressful experience as it includes stress and coping strategies(Apóstolo et al., 2006).

Depression is an emotional and mental disorder that has a verycomplex etiology, despite having well-defined symptoms. Theetiology of depression includes genetic, biological and environ-mental factors. The environment factors model focuses on “theemergence of depressive symptoms” (Stroud et al., 2008).

According to the conception of the DASS, depression is anemotional disorder, but has no specific emotion. Depression maybe associated with a range of emotions, such as anxiety, anger,guilt and shame. Depression is usually trigged by unfavorableliving conditions and may result from the experience of a greatloss or from feelings of hopeless or that life is not worth living(Apóstolo et al., 2006).

Anxiety is a vague and unpleasant feeling of fear or apprehen-sion that may be characterized by a tension or discomfort derivedfrom the anticipation of danger (Graeff, 2007). Anxiety is asso-ciated with the perception of certain environmental contexts(places, people, activities, etc.) that are compared to previousexperiences (memory) and that activate specific brain systemswith adaptive functions (Gray, 1987). In this sense, anxiety may beconsidered as an emotion that is related to a “behavior riskevaluation” (Graeff, 2007).

Lovibond and Lovibond (2004) developed the DASS as a singleinstrument to assess symptoms of depression and anxiety in aninteractive and empirically oriented process. The authors alsoidentified a third factor defined as stress. Their research is basedon the tripartite model of anxiety and depression. The DASS wasconceptualized in the form of a 42 items assessment consisting ofthree subscales with 14-items each. The items refer to symptomsexperienced by the subject in the previous week and use a Likertscale ranging from 0 (“does not apply to me at all”) to 4 (“most ofthe time applies to me”). Although the full version (DASS-42)provides data on specific symptoms of each emotional stateassessed, the authors claim that the reduced version of theinstrument (DASS-21) has the same structure as the full versionbut only requires half the time to complete the investigation ofsymptoms (Lovibond and Lovibond, 2004).

DASS-21 has already been translated to Portuguese fromPortugal (Apóstolo et al., 2006). However, Brazilian Portuguese isslightly different from that spoken in Portugal. Furthermore, thesociocultural context of Brazil is quite different from that presentin Portugal. This study aimed to adapt and validate the DASS-21 foruse in the Portuguese language spoken in Brazil.

2. Method

The present study aimed to measure the properties and psycho-metric equivalence of the DASS-21 in a Brazilian sociocultural context.

The instrument's validity was defined by its ability to demonstratethat it measures the indicators it is intended to measure, in this case,emotional state disorders.

2.1. Participants

The study included patients and caregivers in an outpatientsetting in the city of Santo André/São Paulo, Brazil. Data werecollected from November 2011 to May 2012. Inclusion criteriaincluded an age between 18 and 75 years and frequent visits to theclinic.

2.2. Procedures

The DASS-21 original English version was sent to a bilingualteacher (English – Portuguese) for translation to Portuguese(version 1, V1). V1 was given to a teacher of Portuguese languagefor semantic adjustments and idiosyncratic differences (version 2,V2). Together with the researcher and other mental health careprofessionals, V2 was compared to the final Portuguese versiontested in Portugal. Analysis of content, meaning and adaptationswere made to the Portuguese language used in Brazil to ensure anunderstanding of the content (Beaton et al., 2000). The scale wasthen translated back to the original language (English). In thissense, the goal was to minimize semantic interpretations andmisunderstandings, to ensure the best possible adaptation to theBrazilian culture.

A pre-test was conducted with a sample of six subjects duringan outpatient medical visit, which revealed no difficulties inunderstanding the content of the statements. The scale was laterapplied to a pilot group (N¼48) with similar characteristics to thefinal sample. The assessment was answered pertinently, therebydemonstrating the subjects had a comprehension and under-standing of the propositions.

A back or reverse translation of the scale from BrazilianPortuguese to the original form written in Australian English wasperformed. This back translation was performed by a nativeteacher of the English language without prior knowledge of theoriginal scale.

2.3. Subjects

Participants were invited to participate in the study during theperiod in which they visited an outpatient clinic for consultationsand examinations. The subjects completed the following assess-ments: the DASS 21, the Beck Depression Inventory (BDI), the BeckAnxiety Inventory (BAI) and the Inventory of Stress Symptoms ofLipp (ISSL). The completion of all four instruments lasted 30 minon average.

2.4. Instruments

To validate the DASS, other previously validated instruments wereincluded in the assessment: the BDI for depression (Beck et al., 1961;Gomes-Oliveira et al., 2012; Gorenstein and Andrade, 1996;Gorenstein et al., 1999;), the BAI for anxiety (Beck et al., 1988;Gorenstein and Andrade, 1996) and ISSL for stress (Lipp, 2005). Beckinstruments were used in the construction of the full version DASSthat consists of 42 items. There is no instrument in the literature thatis comparable to the DASS stress subscale. This subscale has affinitieswith measures of negative affect and was originally compared toscales that measure risk for coronary heart disease (Lovibond andLovibond, 2004). The emphasis in these previous studies was toevaluate certain traits or symptoms, not the emotional state itself(Lovibond and Lovibond, 2004). The stress subscale was compared toa developed and validated Brazilian instrument known as the

R.C.B. Vignola, A.M. Tucci / Journal of Affective Disorders 155 (2014) 104–109 105

Inventory of Stress Symptoms of Lipp (Lipp, 2005), which is widelyused in research in Brazil to evaluate stress.

The DASS-21 is a self-report assessment that contains threesubscales scored on a Likert four point scale (0, 1, 2 and 3), rangingfrom 0 (“Strongly Disagree”) to 3 (“Totally Agree”). Each subscaleof the DASS consists of seven items that evaluate the emotionalstates of depression, anxiety and stress.

Table 1 shows the symptoms evaluated by the DASS assessmentby each subscale: depression, anxiety and stress.

The individual questions rate how the individual was feeling inthe last week. There are four possible answers for severity orfrequency, ranging on a scale from 0 to 3. The final result isobtained by summing the scores of the items on each subscale.

Table 2 shows the cutoff points given by the authors (Lovibondand Lovibond, 2004) to evaluate the level of severity through thesum of scores. To calculate the final score of the DASS-21, totalscores for each subscale were multiplied by two.

The BDI was created by Beck et al., (1961) and consists of a self-report questionnaire with 21 multiple-choice items. It is an instru-ment used to measure the severity of depressive episodes. Thisinstrument is widely used by health professionals and researchers ina variety of clinical and research settings.

The BAI was created by Beck et al., (1988) and consists of a self-report questionnaire with 21 questions used to measure theseverity of an individual's anxiety. There are four possible answersin terms of severity or frequency, including “Absolutely not”,“Lightly”, “Moderately” and “Severely”. The final result is obtainedby summing the individual scores, ranging from 0 (absence ofanxiety) to 3 (severe anxiety).

The ISSL was developed by a Brazilian researcher, Marilda Lipp(2005). This assessment is a questionnaire that evaluates stress infour levels – alert, resistance, almost exhaustion and exhaustion.The instrument assesses the presence of symptoms and how theindividual has felt in the last 24 h, in the last month and in the last3 months. For each period of time, a set of 15 physical andemotional symptoms is evaluated for the first 2 periods and 23symptoms are evaluated for the last period.

2.5. Ethical considerations

This study was approved by the Ethics in Research Committeeof the Federal University of São Paulo, Brazil (1749/11) and writteninformed consent was obtained. Participants were guaranteed

confidentiality regarding identifying information, the possibilityof access to the final data from the study and option to endparticipation at any time during the study.

2.6. Statistical analysis

Data were processed and analyzed using the Statistical Packagefor the Social Sciences – SPSS version 18.0 for Windows (SPSS Inc.Released, 2009). The Pearson correlation test (r) and the Kaiser–Meyer–Olkin (KMO) Bartlett's test were used to evaluate thepsychometric properties (reliability and validity) with a confi-dence level of 95% and a po .05. The Cronbach's alpha coefficientwas used to evaluate the internal consistency of the instrument.

Exploratory factor analysis of principal components withorthogonal varimax rotation (Streiner and Norman, 2003) assessedthe construct validity. To complement the factor analysis, thecorrelation between the items and the total scores of each subscalewere analyzed. This analysis was used to indicate the structuralvalidity and to demonstrate that each construct measured speci-fically what it was intended to measure. A good validity for theitem was demonstrated by a substantially higher correlation ofthe item with the subscale to which it belonged compared to thecorrelation with the other subscales for which it did not belong.

3. Results

A total of 242 subjects, aged 18–75 years (mean¼39.9 years;standard deviation¼12.4 years), participated in the study. Themajority of subjects were women (62%).

The KMO test result was .949 (x²¼3542.253, po .01), indicatingthat the adequacy of the model was high.

Cronbach's alpha was .92 for the depression subscale, .90 forthe stress subscale and .86 for the anxiety subscale, indicating agood internal consistency for each subscale. These results alsoprovide evidence that each subscale measured what it proposed tomeasure and that subjects who responded to the questions under-stood them in the same way. The variable distribution according toits factorial loading in each of the three factors (subscales) isshown in Table 3.

The findings also indicate a strong correlation between theDASS-21 and the others scales, BDI, BAI and ISSL. The correlationbetween the depression subscale and the BDI was .86. Thecorrelation between the anxiety subscale and the BAI was .80.The correlation between the stress subscale and the ISSL was .74.The results provide evidence for the convergence between theo-retically equivalent concepts.

4. Discussion

The relationship between anxiety and depression has beencomplex despite several decades of research aimed at under-standing this interaction. They might be different points on the

Table 1DASS subscales and assessed symptoms.

Subscales Depression Anxiety Stress

Symptoms Inertia; Excitation of the autonomous nervous system; Difficulty to relax;Anhedonia; Musculoskeletal effects; Nervous excitation;Dysphoria; Situational anxiety; Easy perturbation/agitation;Lack of interest/involvement; Subjective anxiety Irritability/exaggerated reaction Impatience.Self-depreciation; Experiences.Devaluation of life;Discouragement.

Table 2DASS severity ratings (Lovibond and Lovibond, 2004).

Z Escore Percentile DEPRESSION ANXIETY STRESS

Normal o0,5 0–78 0–9 0–7 0–14Mild 0,5–1,0 78–87 10–13 8–9 15–18Moderate 1,0–2,0 87–95 14–20 10–14 19–25Severe 2,0–3,0 95–98 21–27 15–19 26–33Extremely severe 4 3,0 98–100 28 þ 20þ 34þ

R.C.B. Vignola, A.M. Tucci / Journal of Affective Disorders 155 (2014) 104–109106

same “continuum” mood/affective disorders or distinct disorders.On the other hand, stress is closely related to both depression andanxiety and may lead to the onset of either or both disorders(Kehne, 2007; Kehne and Cain, 2010). The DASS-21 scale wasbased on the tripartite model of anxiety and depression, whichindicates a continuum between depression, anxiety and stress(Clark and Watson, 1991).

This study aimed to validate the DASS-21 to Brazilian Portu-guese. The strong correlations (.86, .80 and .74) between instru-ments provided evidence for the validity of measuring depression,anxiety and stress as emotional states.

Furthermore, the factor analysis indicated that the most sui-table structure included three factors according to the originalscale (Lovibond and Lovibond, 2004). The factor loadings, whichmeasure the correlation of the item with an individual factor,distributed the items in the subscales in a very similar waycompared to the original scale: .81 for depression, .83 for anxietyand .81 for stress (Lovibond and Lovibond, 2004).

The results obtained in the current study support a positiveevaluation of the quality of the DASS-21 compared to othervalidation studies of the same scale. In an earlier Spanish study(Daza et al., 2002), the alpha values were .96 for the total scale, .93for depression, .86 for anxiety and .91 for stress.

An evaluation study of the psychometric properties of the DASS-21 applied to 222 older patients in Houston (EUA) had an alpha equalto .94 for the total scale and .87, .89 and .69 for the subscales ofdepression, anxiety and stress, respectively (Gloster et al., 2008).Other studies supported favorable psychometric properties of theDASS when applied to adults with anxiety and/or mood disorders(Antony et al., 1998; Brown et al., 1997; Clara et al., 2001; Crawfordand Henry, 2005); Covic et al., 2012). The studies showed goodinternal consistency with the DASS in its full version (DASS-42) and

its short version (DASS-21), which had values ranging from .91 to .97for depression, .81 to .92 for anxiety and .88 to .95 for stress.

Considering the three factors proposed by the original scale, asmentioned above, the structure of the current construct had a verysimilar distribution of the subscales compared to the original model.However, item 18 which stated “I felt I was emotional/too sensitive”was characterized as a depressive symptom in this study but wasinitially proposed as a stress symptom. Factor loading was used toindicate how each item (variable) contributed to the formation of thefactor. Thus, this item had a factor loading of .572 in factor 1, whichgrouped the variables that measure depression, and a factor loadingof .497 in factor 2, which grouped the variables that measure stress(the original location of the item). This difference may have occurreddue to cultural differences in Brazil, given that it is possible theemotional sensitivity was related to the excessive sadness andunpleasant events characteristic of depression.

Furthermore, the phrase on the original scale written as “I felt Iwas rather touchy” was translated to “I felt I was very irritable” inthe Portugal version (Apóstolo et al., 2006) and “I felt that I was abit emotional/too sensitive” in this Brazilian version. When thesame phrase was written in the back-translated version, thetranslation “I felt like I was being a little too emotional/sensitive”was quite different from the original meaning. Thus, the originalEnglish expression did not have an equivalent term in thesemantics of Portuguese language in either Brazil or Portugal.

In the validation study of the DASS-21 for the Spanish language(Daza et al., 2002), the three-factor structure explained the datamore clearly. The depression subscale was highly correlated withthe BDI (.86). The anxiety subscale was highly correlated with theBDI (.82). Comparing the correlation results between the stresssubscale and the Beck Anxiety and Depression Inventories, thecorrelation index was highest between stress and depression (.74).

Table 3Main component matrix with varimax rotation forced to three factors (depression, stress and anxiety) and correlation matrix of DASS-21 items (n¼242).

ITEM Factor loadings

Depression Stress Anxiety

D16 I didn't feel enthusiastic about anythinga .747 .417 .264D17 I felt like i was worthless as a persona .743 .347 .222D10 I felt I had no desire for anythinga .728 .308 .258D13 I felt depressed and had no motivationa .707 .397 .307D21 I felt there was no meaning to lifea .657 .219 .471D3 I didn't experience any positive feelingsa .622 .387 .193D5 It was hard for me to have the iniciative to do thingsa .494 .386 .259D18 I felt like I was being a little too emotional/sensitiveb .572 .497 .307D6 I intended to exaggerate when I reacted to situationsb .217 .749 .163D8 I felt I was always nervousb .299 .675 .393D12 I found it difficult to relaxb .397 .671 .236D11 I felt restless .363 .665 .255D1 I found it difficult to calm myselfb .299 .657 .332D14 I was intolerant of the things that kept me from continuing to do what I had been doingb .403 .653 .205D20 I felt afraid for no reasonc .364 .205 .730D9 I got worried about situations in which I could have panicked and looked ridiculousc .248 .335 .688D19 I knew my heartbeat had changed even though I hadn't done anything physically rigorous (for exemple. increased heart rate.irregular heartbeat)c

.493 .094 .605

D2 My mouth felt dryc .064 .399 .586D15 I felt like I was going to panicc .371 .381 .585D7 I felt shaky (for example. in my hands)c .123 .351 .585D4 I had difficulty brething at times (such as wheezing and breathlessness without having made any physical effort)c .453 .028 .560

α (total: .95) .92 .90 .85% Explained variance (total: 63.679) 23.850 21.416 18.414

n Kaiser–Meyer–Olkin measure of sampling adequacy: 949.Bartlett's test of sphericity (Approx. χ2): p¼ .000 3542.253.

a Depression items.b Stress items.c Anxiety items.

R.C.B. Vignola, A.M. Tucci / Journal of Affective Disorders 155 (2014) 104–109 107

The results of this study indicate that the structure of threeseparate factors is closer to that originally proposed by the authorsof the DASS-21 as a means to evaluate three different butcorrelated emotional states.

From a phenomenological standpoint, anxiety and depressionare clearly distinct phenomena. However, it has been very difficultto compare and understand both constructs empirically and tounderstand the relationship between the two emotional states.This study provides further support that depression, anxiety andstress are distinct phenomena. Based on the data obtained, it canbe argued that the constructs under consideration may be differ-ent points along the same continuum and may share somecommon symptoms as proposed by Clark and Watson (1991) intheir Tripartite Model of Anxiety and Depression.

According to Apóstolo et al. (2011), major depression occursthrough a process very similar to a continuous physiological processand to chronic stress. The authors argued that the experience ofcontinuous negative effects may cause general distress in theindividual that may result in recurrent disturbances of brain struc-tures underling some depression disorders (Apóstolo et al., 2011).In this context, it is possible to assume that there will inevitablybe a substantial comorbidity between the two disorders and theirsymptoms.

Anxiety can be defined as an emotional state associated withthe future, which consists of psychological and physiologicalcomponents that are part of the normal spectrum of humanexperience (Bernik and Lopes, 2011). According to these sameauthors, stress may progress to pathological anxiety and/ordepression. In this sense, anxiety or depression may be twosymptoms of the same disease, thereby increasing the difficultyof diagnosing the symptoms.

However, although stress subscale has some affinities with theanxiety subscale, it is also related to the depression subscale.Stress factors may be perceived as negative events, such as sociallosses and psychological suffering. However, the meaning of theseevents must be considered for each individual.

It is worth mentioning that not all individuals with anxietyhave depressive symptoms or vice versa. Some individuals withanxiety disorders are free of depressive symptoms or the symp-toms are insignificant. Pure anxiety without depressive symptomsis more common than pure depression without anxiety symptoms.The comorbidity between anxiety and depression was 57% in avalidation study of the Portuguese version of the DASS-21 com-pleted in Portugal (Apóstolo, 2010). The high co-occurrence ofsymptoms of anxiety and depression suggests that the identifica-tion of these symptoms may clarify the etiology of these twodisorders.

In addition to the affective-emotional states and their respec-tive contexts, studies of depression, anxiety and stress shouldconsider the cognitive issues related to thoughts and individualperceptions of life events.

This study has some limitations that should be considered.Older subjects over 65 years of age were not largely represented inthis sample. Given the increased prevalence of individuals ofadvanced age who suffer from depression and anxiety, a studyspecific to this elderly population should be conducted. Anotherlimitation of the study is education level. In this sample, allrespondents were literate and most individuals had received ahigh school degree. Possible difficulties in understanding the scale,which assesses subjective emotional states, and the impact of loweducation in its applicability should be considered.

Despite these limitations, the findings of our study support thereliability and validity of the DASS-21 to be used in the Portugueselanguage spoken in Brazil. The validity of the Brazilian Portugueseversion of the DASS-21 is supported by evidence of its quality andability to assess emotional states separately, ensuring the

legitimacy of each subscale as an independent measure. Moreover,the current findings support the use of the DASS-21 for the initialdiagnosis and evaluation of therapeutic measures. However, cul-tural and social issues should be considered in understandingstress as an emotional state, which may predict or aggravatedepression and anxiety disorders.

This study provides further support for methodology aimed atidentifying symptoms of depression, anxiety and stress using avalidated instrument. The DASS-21 enables a more accuratediagnosis of these disorders by different professionals outsidethe field of psychiatry, eliminating the use of different instrumentsto assess these states. This also facilitates seeking earlier treatmentand referring an individual for specialized treatment when con-gruent with the individual's needs.

Role of funding sourceFirst author got a scholarship for her master degree from Coordenação de

Aperfeiçoamento de Pessoal de Nível Superior (CAPES). CAPES had no further rolein study design; in the collection, analysis and interpretation of data; in the writingof the report; or in the decision to submit the paper for publication.

Conflict of interestAuthors declare that there is no conflict of interest.

AcknowledgmentsThe authors thank to Souza, A.A.L. for helping in analysis and to the

Cardiologists Partners-Owners of Clinica Eucor, who supported this research andallowed data collection.

References

Antony, M., Bieling, P.J., Cox, B.J., Enns, M.W., Swinson, R.P., 1998. Psychometricproperties of the 42-item and 21-item versions of the Depression AnxietyStress Scales in clinical groups and community a sample. Psychol. Assessment10, 176–181.

Apóstolo, J.L.A., Mendes, A.C., Azeredo, Z.A., 2006. Adaptation to Portuguese of theDepression, Anxiety and Stress Scales (DASS). Rev. Lat. Am. Enfermagem 14 (6),863–871.

Apóstolo, J.L.A., 2010. Depressão, Ansiedade e Stress. In: Apóstolo, J.L.A. (Ed.),O conforto pelas imagens mentais na depressão ansiedade e estresse (e-book).Imprensa da Universidade de Coimbra, Coimbra. Available from: ⟨http://books.google.com.br/books?isbn=9892600363⟩ (accessed 12.12.12).

Apóstolo, J.L.A., Figueiredo, M.H., Mendes, A.C., Rodrigues, M.A., 2011. Depression,anxiety and stress in primary health care users. Rev. Lat. Am. Enfermagem[online] 19 (2), 348–353. (Available from)⟨http://www.scielo.br/pdf/rlae/v19n2/17.pdf⟩.

Beaton, D., Bombadier, C., Guillemin, F., Ferraz, M., 2000. Guidelines for the processcross-cultural adaptation of self-report measures. J. Spine [eletronic version] 24,3186–3191. (Lippincott Williams & Wilkins, Inc. Available from)⟨www.emgo-extra.nl/kc/Beaton.pdf⟩.

Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., Erbauch, G., 1961. An inventory formeasuring depression. [eletronic version]. Arch. Gen. Psychiatry 4 (6), 561–571.

Beck, A.T., Epstein, N., Brown, G., Steer, R.A., 1988. The Beck anxiety inventory.J. Consult. Clín. Psychol. 56, 893–897.

Bernik, V., Lopes, K.V., 2011. Estresse, Depressão e Ansiedade Revista Bras. Medicina[online] 68, (3. especial, Neuropsiquiatria), pp. 4–6. Available from: ⟨www.moreirajr.com.br/revistas.asp?fase=r003&id_materia=4780⟩ (accessd 06.01.13).

Brown, T.A., Chorpita, B.F., Korotitsch, W., Barlow, D.H., 1997. Psychometric proper-ties of the Depression Anxiety Stress Scales (DASS) in clinical samples. Behav.Res. Ther. 35, 79–89. (Available from)⟨www.ncbi.nlm.nih.gov/pubmed/9009048⟩.

Clara, I.P., Cox, B.J., Enns, M.W., 2001. confirmatory factor analysis of the DepressionAnxiety Stress Scales in depressed and anxious patients. J. Psychopathol. Behav.Assessment 23, 61–67.

Clark, L.A., Watson, D., 1991. Tripartite model of anxiety and depression: psycho-metric evidence and taxonomic implications. J. Abnorm. Psychol. 100 (3),316–336. (Available from)⟨www.ncbi.nlm.nih.gov/pubmed/1918611⟩.

Covic, T., Cumming, S.R., Pallant, J.F., Manolios, N., Emery, P., Conaghan, P.G., Tennant,A., 2012. Depression and anxiety in patients with rheumatoid arthristis:prevalence rates based on Depression, Anxiety and Stress Scale (DASS), theHospital Anxiety and Depression Scale (HADS). BMC Psychiatry 12, 6.

Crawford, J.R., Henry, J.D., 2005. The short-form version of the depression anxietyand stress scales (DASS-21): construct validity and normative data in a largenon-clinical sample. Br. J. Clin. Psychol. 44, 227–239. (Available from)⟨www.ncbi.nlm.nih.gov/pubmed/16004657⟩.

R.C.B. Vignola, A.M. Tucci / Journal of Affective Disorders 155 (2014) 104–109108

Daza, P., Novy, D.M., Stanley, M.A., Averyl, L.P., 2002. The depression anxiety stressscale-21: Spanish translation and validation with a Hispanic sample. J. Psycho-pathol. Behav. Assessesment 24 (3), 195–205. (Available from)⟨http://link.springer.com/article/10.1023%2FA%3A1016014818163#⟩.

den Hollander-Gijsman, M.E., Wardenaar, K.J., de Beurs, E., van der Wee, N.J.A.,Mooijaart, A., van Buuren, S., Zitman, F.G., 2012. Distinguishing symptomdimensions of depression and anxiety: an integrative approach. J. AffectiveDisord. 136 (3), 693–701.

Gloster, A.T., Rhoades, H.M., Novy, D., Klostshe, J., Senior, A., Kunic, M., Wilson, N.,Stanley, M.A., 2008. Psychometric properties of the depression, anxiety andstress scale – 21 in older primary care patients. J. Affective Disord. 110 (3),248–259. (Available from)⟨www.ncbi.nlm.nih.gov/pubmed/18304648⟩.

Gomes-Oliveira, M.H., Gorenstein, C., Lotufo Neto, F., Andrade, L.H., Wang, Y.P.,et al., 2012. Validation of the Portuguese version of the Beck DepressionInventory‐II in a community sample. Rev. Bras. Psiquiatr. [online] 34 (4),389–394.

Gorenstein, C., Andrade, L., 1996. Validation of a Portuguese version of the BeckDepression Inventory and the State‐Trait Anxiety Inventory in Brazilian sub-jects. Braz. J. Med. Biol. Res. 29, 453–457.

Gorenstein, C., Andrade, L., Vieira Filho, A.H.G., Tung, T.C., Artes, R., 1999. Psycho-metric properties of the Portuguese version of the Beck Depression Inventoryon Brazilian college students. J. Clin. Psychol. 55, 553–562.

Graeff, F.G., 2007. Anxiety, panic and the hypothalamic-pituitary-adrenal axis. Rev.Bras. Psiquiatr. [online] 29 (supl. 1), s3–s6. (Available from)⟨www.scielo.br/scielo.php?script=sci_arttext&pid=S1516-44462007000500002⟩.

Gray, J.A., 1987. The Psychology of Fears and Stress. Cambridge University Press,Cambridge.

Kehne, J.H., 2007. The CRF1 receptor, a novel target for the treatment of depression,anxiety, and stress-related disorders. CNS Neurol. Disord. Drug Targets 6,163–182.

Kehne, J.H., Cain, C.K., 2010. Therapeutic utility of non-peptidic CRF1 receptorantagonists in anxiety, depression, and stress-related disorders: evidence fromanimal models. Pharmacol. Ther. 128 (3), 460–487.

Lipp, M.E.N., 2005. Manual de Inventário de Sintomas de Stress para Adultos deLipp (ISSL), 3ª. ed. Casa do Psicólogo, São Paulo.

Lovibond, S.H., Lovibond, P.F., 2004. Manual for the Depression Anxiety StressScales, fourth ed. Psychology Foundation, Sydney.

Mello, A.F., Juruema, M.F., Pariante, C.M., Tyrca, A.R., Price, L.H, Carpenter, L.L., Porto,J.A.D., 2007. Depression and stress: is there an endophenotype? Rev. Bras.Psiquiatr. 29 (sup 1), s13–s18. (Available from)⟨http://www.scielo.br/pdf/rbp/v29s1/en_a04v20s1.pdf⟩.

Selye, H., 1936. A syndrome produced by diverse nocuous agents. Nature 138 (1),32.

SPSS Inc. Released, 2009. PASW Statistics for Windows, Version 18.0.. SPSS Inc.,Chicago.

Streiner, D.L., Norman, G.R., 2003. Health measurement scale, A Practical Guide totheir Development and Use, third ed. Oxford University Press, New York.

Stroud, C.B., Davila, J., Moyer, A., 2008. The relationship between stress anddepression in first onsets versus recurrences: a meta-analytic review. J.Abnorm. Psychol. 117, 206–212. (Available from)⟨www.ncbi.nlm.nih.gov/18266498⟩.

R.C.B. Vignola, A.M. Tucci / Journal of Affective Disorders 155 (2014) 104–109 109