ORIGINAL

ISSN 0120-0534

Volumen 58

* Corresponding author.

E-mail: mcsanmartin@uandes.cl https://doi.org/10.14349/rlp.2026.v58.4 0120-0534/© 2026 Fundación Universitaria Konrad Lorenz. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/). Revista Latinoamericana de Psicología (2026) 58, e5804 77 https://doi.org/10.14349/rlp.2026.v58.4 https://revistalatinoamericanadepsicologia.konradlorenz.edu.co/ Psychometric properties of the Overall Anxiety and Depression Severity and Impairment Scales in a Chilean sample Marcela Soto a , Luis Pardo-Garrido a,b , Paulina Pino-Ruy-Pérez b , Paula San Martín b

,

Rodrigo Beltrán c , Jorge Osma e , Vanetza Quezada-Scholz b , María Consuelo San Martín d, * a School of Psychology, Pontificia Universidad Católica de Chile, Santiago, Chile bDepartament of Psychology, Universidad de Chile, Santiago, Chile c Faculty of Psychology, Universidad Nacional de Educación a Distancia, Madrid, Spain d School of Psychology, Universidad de los Andes, Santiago, Chile e Departament of Psychology and Sociology, Universidad de Zaragoza, Teruel, Spain Received 22 October 2025; accepted 12 Mayo 2026 Abstract | Introduction: Anxiety and depressive disorders are common worldwide; their frequent co-occurrence is ex­ plained by shared emotion dysregulation. The Overall Anxiety Severity and Impairment Scale (OASIS) and the Overall Depression Severity and Impairment Scale (ODSIS) enable efficient assessment of emotional disorders by focusing on func­ tional impairment rather than on summing symptoms within diagnostic categories. Objective: To validate the use of the ODSIS and OASIS in the Chilean population. Method: A unidimensional structure was evaluated via confirmatory factor analysis in 740 adults. Internal consistency, temporal stability, and convergent validity were estimated; cutoff points and sex-based measurement invariance were also examined. Results: Acceptable fit indices and very high reliability levels were found for both scales. In addition, cutoff scores of ≥ 6 for ODSIS and ≥ 7 for OASIS were identified, indicating significant depression and anxiety. Finally, multigroup scalar invariance was established for comparisons between men and women. Conclusions: Both scales are valid for use in the Chilean population, with good fit and cutoff points that allow categoriza­ tion according to severity. Due to their brevity and simplicity, their use is recommended in various clinical care settings. Keywords: Emotional disorders, depression, anxiety, psychometric properties, validation © 2026 Fundación Universitaria Konrad Lorenz. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/ by-nc-nd/4.0/).

Propiedades psicométricas de las Escalas de Severidad e Interferencia de Ansiedad y Depresión en una muestra chilena Resumen | Introducción: Los trastornos de ansiedad y depresión son frecuentes a nivel mundial; su coocurrencia se ex­ plica por una desregulación emocional compartida. La Overall Anxiety Severity and Impairment Scale (OASIS) y la Overall Depression Severity and Impairment Scale (ODSIS) permiten evaluar con eficiencia los trastornos emocionales al centrar­ se en la interferencia funcional, más que en la suma de síntomas por categorías diagnósticas. Objetivo: Validar el uso de ODSIS y OASIS en población chilena. Método: Se evaluó la estructura unidimensional mediante análisis factorial confir­ matorio en 740 adultos. Se estimaron consistencia interna, estabilidad temporal y validez convergente; además, puntos de corte e invariancia por sexo. Resultados: Se encontraron índices de ajuste aceptables y nivel de confiabilidad muy alto para

2/9 M. Soto et al.

The World Health Organisation (WHO) defines common mental disorders (CMD) as depressive and anxiety dis­ orders, given their high prevalence. These conditions affect mood and emotions, exceeding ordinary expe­ riences of sadness, stress, or fear (WHO, 2017). In 2015, global cases reached 264 million for anxiety and 322 million for depression, corresponding to 3.6% and 4.4% of the population, respectively, with higher prevalence among women (WHO, 2023).

In Chile, depressive disorder affects 6.2% of the pop­ ulation, with a higher prevalence in women (10.1%) than men (2.1%; MINSAL, 2016-2017). Unipolar depression ranks as the second leading cause of disability-adjust­ ed life years (4.5%; Vicente et al., 2016). Anxiety is con­ sidered the main psychological distress in the country (Bravo et al., 2024), with a prevalence of 6.5%; lifetime rates reach 17.5% in women and 9.5% in men (INJUV,

2024).

CMDs share more than prevalence.

Nearly two decades ago, Barlow et al. (2004) introduced “emotional disorders,” initially encompassing anxiety and unipolar mood disorders, later expanded to emo­ tion-regulation problems more generally (Bullis et al., 2019). These arise when individuals attempt to control emotions through concealment or avoidance, which paradoxically intensifies them (Barlow et al., 2004). This shared regulatory dysfunction explains their frequent comorbidity (Castro et al., 2025; Harvey et al., 2004). Global data show that among those with anxiety dis­ orders, comorbidity is the rule, with depression as the most common comorbid diagnosis (Javaid et al., 2023). In generalized anxiety disorder, lifetime comorbidity exceeds 80%, mainly with mood disorders (> 60%; Rus­ cio et al., 2017). For social anxiety disorder, comorbidity is higher with other anxiety disorders (60%) than with mood disorders (47%; Stein et al., 2017). According to WHO (2023), anxiety symptoms usual­ ly begin in childhood or adolescence and, if untreated, persist in adulthood. Both anxiety and depression are more prevalent in women. Despite effective treatments, only about one in four affected individuals receive care. When untreated, these disorders cause persistent dis­ tress and significant impairment across domains such as social roles and relationships.

Given their impact, accurate evaluation of emotion­ al disorders is essential to determine need and guide treatment. Numerous instruments exist with adequate psychometric properties (Beck et al., 1988; Kroenke et al., 2001). However, Osma et al. (2019, 2022) highlight that ambas escalas. Además, se identificaron puntajes de corte de ≥ 6 para ODSIS y ≥ 7 para OASIS, indicativos de depresión y ansiedad significativas. Finalmente, se obtuvo invarianza multigrupo de nivel escalar para comparaciones entre hombres y mujeres. Conclusiones: Ambas escalas son válidas para su uso en población chilena, con un buen ajuste y con puntos de corte que permiten categorizar según severidad. Por su brevedad y sencillez, se recomienda su uso en diversos contextos de atención clínica.

Palabras clave: Trastornos emocionales, depresión, ansiedad, propiedades psicométricas, validación © 2026 Fundación Universitaria Konrad Lorenz. Este es un artículo Open Access bajo la licencia CC BY-NC-ND (https://creativecommons.org/licenses/ by-nc-nd/4.0/).

most rely on categorical diagnoses and symptom check­ lists, estimating severity by symptom count while ne­ glecting functional impairment. This approach is also impractical in high-demand, resource-limited con­ texts, as it requires multiple disorder-specific scales. To overcome these issues, the Overall Anxiety Sever­ ity and Impairment Scale (OASIS; Norman et al., 2006) and Overall Depression Severity and Impairment Scale (ODSIS; Bentley et al., 2014) provide efficient transdiag­ nostic assessment using two five-item measures. Both evaluate symptom severity and associated impairment independent of diagnosis.

These scales have been validated across populations, including community (Rojas et al., 2023; Sandora et al., 2021), student (Norman et al., 2011; Osma et al., 2022), and clinical samples (Hermans et al., 2015; Mira et al., 2019; Osma et al., 2019), and in several countries such as the United States, Iran, the Netherlands, Argentina, Spain, Japan, and the Czech Republic. Evidence supports their robust unidimensional structures, though some stud­ ies report correlated errors between OASIS items 1–2 and 1–3 (Bragdon et al., 2016; González-Robles et al., 2018; Moore et al., 2015; Norman et al., 2011; Osma et al., 2019; Sandora et al., 2021). Both show adequate internal con­ sistency and validity, with cutoffs ranging from 5–12 (ODSIS) and 4–15 (OASIS; Rojas et al., 2023). Although research on their psychometric properties is extensive, only one study has been conducted in Lat­ in America (Rojas et al., 2023), which did not examine test–retest reliability, cutoffs, or measurement invari­ ance. The present study provides the first comprehen­ sive Latin American validation in a Chilean community sample. It evaluates factorial structure and convergent validity, estimates temporal reliability and cutoffs for case classification, and tests measurement equivalence between women and men.

Method Participants The sample consisted of 740 Chilean residents from across the country, recruited via social media (72.3% women; M = 35.3 years, SD = 12.2, range = 18–71). Partic­ ipants were distributed across all 16 Chilean regions, with the highest concentration in the Metropolitan Region (51.9%), followed by Valparaiso (9.6%), Biobio (6.5%), with the remaining regions each contributing less than 5% of the sample. Additional demographics are reported in Table 1. Participants provided written informed consent before beginning the survey. Proce­

3/9 Psychometric properties of the Overall Anxiety and Depression Severity and Impairment Scales in a Chilean sample dures were approved by the Research Ethics Committee of the Faculty of Medicine, the University of Chile (Ap­ proval N°073), and adhered to the Declaration of Helsin­ ki (World Medical Association, 1964).

Table 1. Demographic information (n = 740) Demographic n (%) Sex Male Female Transexual (male and female)

182 (24.59%)

550 (74.32%)

7 (0.94%)

Gender Man Woman Transgender (man and woman) Othera

182 (24.59%)

535 (72.29%)

3 (0.41%)

20 (2.7%)

Educational Level Elementary completeb Secondary completeb College degree or technical studies Graduate degree

5 (0.67%)

320 (43.24%)

306 (41.35%)

109 (14.73%)

Employment situation/status* Working/having a job Unemployed Retired

354 (47.8%)

371 (50.14%)

23 (3.11%)

a This includes people who prefer not to answer and others who do not identify with the given options. b In Chile, Elementary corresponds from 1st to 8th grade, and Secondary from 9th to 12th grade.

* The total is more than 100 percent since there are people who

are retired but working.

Instruments Overall Depressive Severity and Impairment Scale (ODSIS; Bentley et al., 2014; Spanish validation by Osma et al., 2019). This 5-item self-report scale screens depressive symptoms and their interference and is also used to track treatment progress. Items are rated 0–4 (0 = “I did not feel depressed,” 4 = “Constant depression”); totals range 0–20, with higher scores indicating greater impairment. It shows excellent internal consistency ( = .94) and strong convergent and discriminant validity (Osma et al., 2019).

Overall Anxiety Severity and Impairment Scale (OASIS; Norman et al., 2006; Spanish validation by Osma et al., 2019). This is a five-item self-report meas­ ure of anxiety symptoms and their interference. Items are rated on a five-point Likert scale from 0 (“I did not feel anxious”) to 4 (“constant anxiety”) across different situations; higher scores indicate greater impairment. The total score (0–20) is the sum of all items. The in­ strument shows adequate internal consistency (Cron­ bach’s alpha of .87) and good convergent and discrimi­ nant validity (Osma et al., 2019).

Depression, Anxiety and Stress Scale (DASS-21; Lovibond & Lovibond, 1995; Chilean validation by An­ tunez & Vinet, 2012). The DASS-21 is a 21-item self-re­ port measure of stress, anxiety, and depression across three factors. Items use a four-point Likert scale from

0 (Nothing applicable to me) to 3 (Very applicable to

me/most of the time), with higher scores indicating greater symptomatology. Factor scores are obtained by summing the respective items. In the Chilean samples, internal consistency was acceptable (depression Cron­ bach’s alpha = .85; stress = .83; anxiety = .73), and con­ vergent validity was adequate (Antunez & Vinet, 2012). Rosenberg Self-esteem Scale (RSS; Rosenberg, 1965; Chilean validation by Rojas-Barahona et al., 2009). This 10-item questionnaire assesses self-evalu­ ations across two five-item dimensions—positive and negative—rated on a 1–4 Likert scale (strongly disagree to strongly agree; 4). Negatively keyed items are re­ verse-scored; higher totals indicate greater self-esteem. In Chile, internal consistency for the total scale was  = .75 (Rojas-Barahona et al., 2009).

Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004; Hervás & Jódar, 2008; Chile­ an validation by Guzmán-González et al., 2014). The DERS assesses clinically relevant emotion-regulation difficulties via 25 items across five factors—Emotion­ al Dyscontrol (6), Everyday Interference (4), Emotional Disattention (5), Emotional Confusion (3), and Emo­ tional Rejection (7). Items are rated 1 (Almost never) to

5 (Almost always); higher scores indicate greater diffi­

culty, with factor totals computed by summation. The Chilean version showed adequate reliability ( = .71–.88 across subscales;  = .92 total) and acceptable conver­ gent validity (Guzmán-González et al., 2014). Analysis plan Sample characteristics were assessed by analysing the sociodemographic variables using frequency distribu­ tions and percentages.

Confirmatory factorial analysis (CFA) and Mul­ tigroup Invariance Measurement. We used CFA to examine whether each scale was essentially unidi­ mensional. We tested three specifications: for both the ODSIS and the OASIS, a five-item single-factor model; and, for the OASIS, a variant allowing a residual cor­ relation between Items 1 and 2, given their conceptual overlap. Models were estimated with WLSMV (Weighted Least Squares Mean and Variance Adjusted), appropri­ ate for ordinal Likert responses.

To judge model fit, we relied on the Comparative Fit Index (CFI), the Tucker–Lewis Index (TLI), the Root Mean Square Error of Approximation (RMSEA), and the Standardised Root Mean Square Residual (SRMR). Fol­ lowing conventional criteria, we treated CFI/TLI ≥ .95 as evidence of good fit and RMSEA/SRMR ≤ .08 as accept­ able (Cheung & Rensvold, 2002; Hu & Bentler, 1999). We examined the stability of the factor structure across sex using multigroup CFA. A configural mod­ el was first tested, allowing loadings and intercepts to vary. Metric invariance was then evaluated by con­ straining loadings across groups, and scalar invariance by additionally constraining intercepts. Comparisons among these nested models assessed the stability of the measurement structure.

Model equivalence was tested using chi-square dif­ ference tests between nested models (e.g., configural vs. metric, metric vs. scalar). A significant difference in­

4/9 M. Soto et al.

dicated that the more constrained model had a worse fit and that invariance was not achieved. Additionally, alternative fit indices were used for comparison, with differences in CFI and RMSEA smaller than .01 taken as evidence of equivalence (Cheung & Rensvold, 2002; Kline, 2016).

Reliability and Convergent Validity. Internal con­ sistency was assessed with Cronbach’s alpha and Mc­ Donald’s omega (Dunn et al., 2014), and temporal reli­ ability with the intraclass correlation coefficient (ICC; Shrout & Fleiss, 1979). The mean retest interval was 25 days.

To assess construct validity, we examined associ­ ations between the ODSIS and the OASIS and external measures (DASS-21 Anxiety, Depression, Stress; DERS subscales: Emotional Dyscontrol, Everyday Interfer­ ence, Emotional Disattention, Emotional Confusion, Emotional Rejection; and self-esteem). In the prima­ ry analyses, the ODSIS and the OASIS were specified as latent variables and external measures as manifest variables. For comparability with prior work, we also computed manifest correlations using ODSIS and OASIS total scores.

Estimation of optimal cutoff. To appraise the sen­ sitivity and specificity of the ODSIS and OASIS, we used the Chilean DASS-21 as the reference standard. In line with Román et al. (2016), scores > 5 on the Depression subscale and > 4 on the Anxiety subscale were taken as indicators of clinically significant symptomatology. To evaluate discriminatory precision, we estimated receiver operating characteristic (ROC) curves and the area under the curve (AUC). Following Rice and Harris (2005), AUCs ≥ .70 denote fair discrimination, ≥ .80 good, and ≥ .90 very good. Optimal cutoffs were derived using multiple criteria, the Youden index (J), Euclidean dis­ tance to (0,1; ER), the Concordance Probability Method (CZ), and the Index of Union (IU), consistent with prior applications (González-Robles et al., 2018; Mira et al., 2019). The threshold satisfying most criteria was adopt­ ed as the optimal cutoff.

We also report positive predictive value (PPV) and negative predictive value (NPV), defined as the prob­ abilities that positive and negative test results corre­ spond to the true presence or absence of the condition, respectively. For both the ODSIS and OASIS, stratified bootstrap confidence intervals were computed for AUC, sensitivity, and specificity while preserving the propor­ tion of cases and controls, with 2000 resamples. Analyses were conducted in R and Mplus 8.9. In R, we computed demographics, ROC analyses, and sex mean differences using pROC (v1.18.5; Robin et al., 2011) and verification (v1.42; NCAR-RAL, 2015) in RStudio (Posit Team, 2024). CFA, correlations, and multigroup meas­ urement invariance were run in Mplus (Muthén & Muthén, 2017).

Results Demographics and descriptive data As previously stated, most participants identified themselves as female, and more than 55% completed studies pursuant to school. The sample is almost equal­ ly divided between unemployed (possibly studying) and employed participants. The mean scores of the ODSIS and OASIS scales were 6.9 (SD = 4.96) and 7.4 (SD = 4.7), re­ spectively. The Wilcoxon test revealed a significant dif­ ference in the OASIS scores by sex (W = 43883, p < 0.05), with females scoring higher (7.7, and males = 6.5). No differences were found in the ODSIS (W = 47570, p = 0.14, females = 7.1, and males = 6.5).

Models fit and reliability The unidimensional model of 5 items estimated for the ODSIS achieves acceptable fit indices overall (2 (5) = 55.539, p < .000, CFI = .996, TLI = .993, RMSEA = .117, SRMR = .011). Although the RMSEA value exceeds the conven­ tional threshold of .08, this should be interpreted with caution: RMSEA is known to be upwardly biased in mod­ els with few degrees of freedom and ordinal indicators estimated via WLSMV (Xia & Yang, 2019), a condition that applies here (df = 5). Considering the remaining in­ dices (CFI, TLI, and SRMR), which all fall within accept­ able ranges, the overall model fit is deemed adequate. Turning to the OASIS, the first model specified, equiva­ lent to the ODSIS structure, shows a similar pattern (2 (5) = 212.407, p < .000, CFI = .982, TLI = .964, RMSEA = .237, SRMR = .040). Given the conceptual overlap between Items 1 (frequency of anxiety) and 2 (severity of anxie­ ty), which both tap the subjective experience of anxious affect rather than its behavioural consequences (avoid­ ance, work/social interference), a correlated residual was specified between these two items, consistent with prior CFA studies of the OASIS (Bragdon et al., 2016; Nor­ man et al., 2011; Osma et al., 2019; Sandora et al., 2021). With this specification, the fit improves substantially (2 (4) = 13.723, p < .05, CFI = .999, TLI = .998, RMSEA = .057, SRMR = .007). Therefore, the following analyses were conducted using the 5-item unidimensional model for the ODSIS and, for the OASIS, the model that includes the residual variance correlation between items 1 and 2. Polychoric correlations between the items of each scale, estimated according to the chosen models, are presented in Table 2, showing a strong association among the indicators. For the ODSIS, values range from .74 to .83; for the OASIS, they range from .60 to .86 (all p-values are significant). Regarding factor loadings, all values for both scales are high (> .75), equivalent and significant. Internal consistency with both parameters was excellent (s and ωs ≥ .90), and temporal stability was adequate for both scales (> .75).

Convergent validity The results were all significant regarding the relation­ ship between the ODSIS, OASIS, and other variables, with the values and directions of the relationships aligning with literature. As shown in Table 3, the highest cor­ relation values for both scales were with the DASS-21 subscales (all > .70). Subsequently, a negative moderate correlation was observed with the positive and nega­ tive RSS subscales. Therefore, higher scores in the OD­ SIS and OASIS correspond to lower self-esteem. Finally, related to the DERS, the highest correlation values were with the Emotional Dyscontrol, Emotional Rejection,

5/9 Psychometric properties of the Overall Anxiety and Depression Severity and Impairment Scales in a Chilean sample Everyday Interference, and Emotional Confusion sub­ scales, indicating a moderate correlation with the ODSIS and OASIS, ​in contrast the​ correlation with Emotional Dysattention was weak (for details of correlation values between manifest variables see Online Resource 1). Multigroup measurement invariance In Table 4, the results of the multigroup invariance analysis between men (n = 186) and women (n = 551) for the ODSIS and OASIS are presented. The three estimated invariance models show good fit in both groups, with fit index values within the expected ranges. Specifically, the configural invariance obtained the following val­ ues: ²(10) = 37.513, p < .000, CFI = .985, TLI = .969, RMSEA = .086, SRMR = .017 for ODSIS and ²(8) = 10.304, p < .000, CFI = .999, TLI = .997, RMSEA = .028, SRMR = .015 for OASIS. The metric invariance, which assesses the equality of the factor loadings, was supported with non-significant p-values (p > 0.05) when compared with the configural model and small ² differences (Δ² = 4.292, p = .368 for the ODSIS and Δ² = 4.224, p = .376 for the OASIS). Finally, the scalar invariance, which examines the equality of intercepts, also showed good fit for both instruments and non-significant differences when compared to the metric invariance (Δ² = 2.300, p = .681 for ODSIS and Δ² = 8.879, p = .064 for the OASIS). These results indicate that in the estimated models both the factor loadings and intercepts are equivalent between men and wom­ en, allowing for comparisons of the mean scores of the ODSIS and OASIS scales between these groups. Optimal cutoff To assess the sensitivity and specificity of the ODSIS and OASIS as screening tests, ROC curves were calculat­ ed. The total sample was composed of 740 participants, with 291 and 349 in each control group. For the ODSIS, the AUC obtained was .92 (CI 95% = [0.91, 0.94]; p < .01; see Figure 1.a), and for the OASIS the AUC obtained was .86 (CI 95% = [0.84, 0.89]; p < .01; see Figure 1.b), indicating that both tests have excellent performance discrim­ inating among the groups created by dichotomising participants based on the cutoff of > 6 and > 4 estab­ lished by Antunez and Vinet (2012) in the depression and anxiety DASS-21 subscales, respectively. Table 2. Standardised factor loadings and items correlation for the ODSIS and OASIS

ODSIS

OASISa Items FL

1

2

3

4

FL

1

2

3

4

1 Frequency of symptoms

.90

.75

2 Intensity of symptoms

.89

.83

.80

.86

3 Degree of avoidance/loss of interest

.91

.81

.83

.84

.61

.86

4 Interference in work, school life

.87

.75

.81

.81

.90

.70

.61

.76

5 Interference in social life

.86

.76

.75

.79

.77

.90

.67

.70

.77

.80

Cronbach´s Alpha ()

.93

.92

McDonald´s Omega (ω)

.92

.90

ICC

.81

.78

Note. All factor loadings and correlation values are significant. FL: Factor loadings; ICC: Inter-class correlation. a The model was estimated using the variance error of the correlation between items 1 and 2. Table 3. Variable means and correlations Manifest

ODSIS

Manifest

OASIS

DASS

-A

DASS

-D

DASS

-S

RSS-POS

RSS-

NEG

DERS-

DYS

DERS-

REJ

DERS-

INT

DERS-DIS DERS-

CON

Latent

ODSIS

-

.77

.70

.87

.72

-.56

-.60

.50

.59

.59

.31

.48

Latent

OASIS

.80

-

.78

.72

.78

-.45

-.52

.59

.56

.58

.27

.44

Mean (SD)

6.99 (4.96)

7.43 (4.71)

5.8

(4.71)

8.03 (5.8)

8.96

(5.02)

14.82

(3.34)

11.96

(3.74)

11.66

(5.61)

18.02

(8.58)

11.66

(3.14)

13.33

(5.23)

6.68

(3.14)

Note. OASIS: Overall Anxiety Severity and Impairment Scale; ODSIS: Overall Depression Severity and Impairment Scale; DASS-A: Anxiety subscale of the Depression, Anxiety and Stress Scales-21 (DASS-21); DASS-D: Depression subscale of the DASS-21; DASS-S: Stress subscale of the DASS-21; RSS-POS: Positive subscale of the Rosenberg Self-esteem Scale (RSS); RSS-NEG: Negative subscale of the RSS; DERS-DYS: Emotional Dyscontrol subscale of the Difficulties in Emotion Regulation Scale (DERS); DERS-REJ; Emotional Rejection subscale of the DERS; DERS-INT: Everyday Interference subscale of the DERS; DERS-DIS: Emotional Disattention subscale of the DERS; DERS-CON: Emotional Confusion subscale of the DERS; SD: Standard deviation. All values are significant.

6/9 M. Soto et al.

Figure 1. ODSIS and OASIS ROC curves a)

1 − Specificity

Sensitivity

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

b)

1 − Specificity

Sensitivity

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

Note. The ROC curve for the ODSIS is shown on the left side (a), while the curve for the OASIS is displayed on the right (b). Moreover, when scrutinising the four methods for esti­ mating the best thresholds in the present study (i.e., J, IU, ER and CZ), ODSIS 5.5 converges as the best option as cutoff point, with good sensitivity (0.87; CI 95% = [0.84, 0.90]) and specificity values (0.82; CI 95% = [0.78, 0.87]). PPV was 0.88 (CI 95% = [0.86, 0.91]) and NPV was 0.80 (CI 95% = [0.77, 0.84]; for more details see Online Resource 2). In practice, a score of 6 or more points is considered indicative of relevant depression. For the OASIS, the best cutoff was 6.5, with good sensitivity (.80; CI 95% = [0.76, 0.84]) and specificity (.75; CI 95% = [0.71, 0.80]) values. PPV was 0.78 (CI 95% = [0.75, 0.82]) and NPV was

0.77 (CI 95% = [0.73, 0.81]; for more details see Online Re­

source 3). Therefore, a score of 7 or higher is regarded as a threshold for relevant anxiety. Regarding PPV and NPV, in the case of the ODSIS, they indicate that 88% of the persons who tested positive for high depression lev­ els actually present relevant depression signs, and 80% of the participants who tested negative were indeed healthy. For the OASIS, 78% of the individuals with high anxiety were correctly identified, while 77% of those without relevant anxiety symptoms were accurately classified as free from significant anxiety. Discussion This study advances evidence on the ODSIS and OASIS in Latin America. Building on Rojas et al. (2023) with an Argentine sample, it delivers a more comprehen­ sive psychometric appraisal by testing temporal sta­ bility, estimating cutoffs, and evaluating measurement equivalence across sexes. Overall, the findings support the validity and utility of both scales and clarify their performance in a non-clinical Chilean community sample. Importantly, this cross-cultural contribution underscores that the constructs of anxiety and depres­ sion severity, as operationalised by the OASIS and OD­ SIS, appear to function comparably across diverse cul­ tural contexts. Although the transdiagnostic model of emotional disorders (Barlow et al., 2004) is theoretically culture-general, the way individuals interpret, express, and report emotional distress may vary across cultur­ al contexts (Canino & Alegria, 2008), and that should be addressed in other studies exploring potential cultural nuances in item interpretation.

When analysing the means, both scales yielded val­ ues at the upper end of those reported for the general population in previous studies (e.g., Bentley et al., 2014; Sandora et al., 2021), suggesting the importance of ex­ amining clinical samples to more clearly differentiate the two groups in terms of their scale scores in the Chil­ ean sample.

Table 4. Multigroup invariance between males (n = 186) and females (n = 551) Instrument Model 2(df) Δ2 p-value

CFI

TLI

RMSEA

SRMR

Configural

37.513(10)

.985

.969

.086

.017

ODSIS

Metric

44.661(14)

4.292

.368

.983

.976

.077

.023

Scalar

48.376(18)

2.300

.681

.983

.981

.068

.026

Configural

10.304(8)

.999

.997

.028

.015

OASIS

Metric

14.647(12)

4.224

.376

.999

.998

.024

.023

Scalar

23.301(16)

8.879

.064

.996

.995

.035

.024

Note. Fit indices indicate good model fit at the configural, metric, and scalar levels.  2: Chi square; df: degrees of freedom; CFI: Comparative Fit Index; TLI: Tucker-Lewis Index; RMSEA: Root Mean Square Error of Approximation; SRMR: Standardised Root Mean Square Residual.

7/9 Psychometric properties of the Overall Anxiety and Depression Severity and Impairment Scales in a Chilean sample Moderate to high correlations between the ODSIS/ OASIS and external measures support their validity for assessing depression and anxiety. Strong associations with the corresponding DASS-21 subscales replicate pri­ or findings (Bragdon et al., 2016; Osma et al., 2022). Corre­ lations with the stress subscale were notable and these patterns suggest that stress is a key transdiagnostic component not directly captured by the ODSIS/OASIS and warrants explicit assessment.

Moreover, moderate negative correlations with the RSS positive and negative subscales indicate that great­ er depression and anxiety are associated with lower self-esteem. Moderate correlations with DERS subscales further underscore the link between emotion-regula­ tion difficulties and depressive/anxious symptoms. By contrast, the weak association with Emotional Disat­ tention suggests attentional processes are less central; individuals with higher symptom levels may intently attend their symptoms rather than disregard them. The cutoffs identified here align with prior reports (Rojas et al., 2023). For the ODSIS, a score ≥ 6 indicates clinically meaningful depressive impairment (cutoff point estimate = 5.5). For the OASIS, scores ≥ 7 reflect ele­ vated anxiety (threshold = 6.5). To our knowledge, these exact values have not been reported previously; how­ ever, they are close to the ODSIS cutoff of 5 in Spanish students (Osma et al., 2022) and to the OASIS threshold of 6 for moderate anxiety in a clinical sample (Bragdon et al., 2016). Further Chilean studies using diagnostic in­ terviews or confirmed diagnoses should test these cut­ offs to refine case classification.

Multigroup invariance analysis for the ODSIS and OASIS support measurement equivalence across sexes. Configural, metric, and scalar invariance were estab­ lished—factor structures, loadings, and intercepts were comparable for males and females—indicating similar interpretation of the constructs and permitting mean­ ingful comparison of group mean scores. Looking into the mean scores, females showed sig­ nificantly higher OASIS means than males, consistent with WHO (2023) evidence of higher anxiety prevalence in women. For the ODSIS, group means did not differ significantly; however, equivalence between female and male means was not established.

Limitations and future directions Among the limitations, the marked female–male im­ balance, also noted in comparable studies (Osma et al., 2019, 2022; Rojas et al., 2023), may bias estimates by overrepresenting one group’s experiences. Participants were recruited online via a non-probabilistic conven­ ience sampling strategy, which, combined with the predominance of female respondents and the overrep­ resentation of the Metropolitan Region, limits the gen­ eralisability of the findings to the broader Chilean pop­ ulation. Future research should diversify recruitment and assess psychometric properties in clinical samples to strengthen applicability to health settings and ex­ ternal validity. Moreover, the cutoff should be used cau­ tiously: it is not a diagnostic screen but a criterion for categorising impact and severity in this sample. To im­ prove its utility, studies in clinical populations should also estimate indices of reliable change and clinically significant change.

Finally, some aspects of the measurement model raise concerns regarding potential item redundancy. This issue is reflected in the need to correlate residuals between two OASIS items in order to achieve acceptable model fit, as well as in the high inter-item correlations, strong factor loadings, and very high internal consist­ ency indices ( and ω ≥ .90). Although these results may superficially suggest good psychometric properties, given that both scales consist of only five items, they may instead indicate overlapping item content rather than a well-differentiated representation of the under­ lying constructs.

Additionally, the elevated RMSEA observed in the initial models could reflect some degree of model mis­ specification related to this redundancy, beyond the known limitations of RMSEA in models with few de­ grees of freedom.

Taken together, these patterns suggest that the scales may capture a relatively narrow facet of anxiety and depression severity, warranting further examina­ tion of item distinctiveness, particularly for items as­ sessing frequency and intensity, which appear concep­ tually close. At the same time, although the remaining items seem to cover more differentiated domains (e.g., avoidance behaviours, social interference, and work impairment), their similar grammatical structure and shared response format may still contribute to inflated associations among items and, consequently, to the ob­ served redundancy.

Declarations Funding MCSM was supported by the Agencia Nacional de In­ vestigación y Desarrollo de Chile (ANID) through the fund [Fondecyt # 11250037], JO received support from the Autonomous Government of Aragón [Departamen­ to de Ciencia, Universidad y Sociedad del Conocimiento] (Grant Number Research Team S31_23R).

Competing interests The authors declare they have no related financial interests.

Ethics approval This study was performed in line with the principles of the Declaration of Helsinki.

Data availability statement The data that supports the findings of this study are available from the corresponding author, upon reason­ able request.

Declaration of conflicting interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publi­ cation of this article.

8/9 M. Soto et al.

Author credit statement MS: Conceptualisation, Methodology, Writing - Origi­ nal Draft, Formal analysis, Writing - Review & Editing. LPG: Conceptualisation, Methodology, Software, For­ mal analysis, Writing - Original Draft, Writing - Review & Editing.

PPRP: Conceptualisation, Investigation, Writing

- Original Draft, Writing - Review & Editing. PSM: Con-

ceptualisation, Investigation, Data Curation. RB: Conceptualisation, Writing - Review & Editing. VQS: Supervi­ sion, Writing - Review & Editing. MCSM and JO: Concep­ tualisation, Supervision, Writing - Review & Editing, Funding acquisition.

References Antunez, Z., & Vinet, E. V. (2012). Depression Anxiety Stress Scales (DASS-21): Validation of the abbreviated version in chilean university students. Terapia Psicológica, 30(3), 49-

55. https://doi.org/10.4067/S0718-48082012000300005

Barlow, D. H., Allen, L. B., & Choate, M. L. (2004). Toward a unified treatment for emotional disorders. Behavior Therapy, 35(2), 205-230. https://doi.org/10.1016/S0005-7894(04)80036-4 Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inven­ tory for measuring clinical anxiety: Psychometric proper­ ties. Journal of Consulting and Clinical Psychology, 56(6), 893-

897. https://doi.org/10.1037//0022-006x.56.6.893

Bentley, K. H., Gallagher, M. W., Carl, J. R., & Barlow, D. H. (2014). Development and validation of the Overall Depression Severity and Impairment Scale. Psychological Assessment, 26(3), 815-830. https://doi.org/10.1037/a0036216 Bragdon, L. B., Diefenbach, G. J., Hannan, S., & Tolin, D. F. (2016). Psychometric properties of the Overall Anxiety Severity and Impairment Scale (OASIS) among psychiatric outpa­ tients. Journal of Affective Disorders, 201, 112-115. https://doi. org/10.1016/j.jad.2016.05.005 Bravo, D., Errázuriz, A., Calfucoy, P., & Campos, D. (2024). Ter­ mómetro de la salud mental en Chile ACHS-UC: Octava ron­ da. Pontificia Universidad Católica de Chile. https://www. achs.cl/docs/librariesprovider2/2024/achs-221357/termo­ metro-de-la-salud-mental-en-chile-achs-uc-octava-ron­ da.pdf Bullis, J. R., Boettcher, H., Sauer-Zavala, S., Farchione, T. J., & Barlow, D. H. (2019). What is an emotional disorder? A transdiagnostic mechanistic definition with implications for assessment, treatment, and prevention. Clinical Psy­ chology: Science and Practice, 26(2), Article e12278. https:// doi.org/10.1037/h0101755 Canino, G., & Alegría, M. (2008). Psychiatric diagnosis—Is it universal or relative to culture? Journal of Child Psychology and Psychiatry, 49(3), 237-250. https://doi.org/10.1111/j.1469- 7610.2007.01854.x Castro, J. C., Diez-Canseco, F., Brusco, L. I., Rodríguez Malagón, N., Fung, C., Cepeda-Gil, M., Carbonetti, F. L., Ramírez-Me­ neses, D., Esnal, F., Carbonel, A., Bird, V., Priebe, S., & Gó­ mez-Restrepo, C. (2025). Depresión, ansiedad y comorbili­ dad ansiedad-depresión y su relación con el capital social y la resiliencia en jóvenes de áreas urbanas desfavorecidas: un estudio transversal en América del Sur. Revista Colom­ biana de Psiquiatría, 54(2), 33-42. https://doi.org/10.1016/j. rcp.2025.08.001 Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-offit indexes for testing measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 9(2), 233-255. https://doi.org/10.1207/S15328007SEM0902_5 Dunn, T. J., Baguley, T., & Brunsden, V. (2014). From alpha to omega: A practical solution to the pervasive problem of in­ ternal consistency estimation. British Journal of Psychology, 105(3), 399-412. https://doi.org/10.1111/bjop.12046 González-Robles, A., Mira, A., Miguel, C., Molinari, G., Díaz- García, A., García-Palacios, A., Bretón-López, J. M., Quero, S., Baños, R. M., & Botella, C. (2018). A brief online transdi­ agnostic measure: Psychometric properties of the Overall Anxiety Severity and Impairment Scale (OASIS) among Spanish patients with emotional disorders. Plos One, 13(11), e0206516. https://doi.org/10.1371/journal.pone.0206516 Gratz, K., & Roemer, L. (2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the Difficulties in Emotion Regulation Scale. Journal of Psychopathology and Behavioral Assessment, 26, 41-54. https://doi.org/10.1023/B:­

JOBA.0000007455.08539.94

Guzmán-González, M., Trabucco, C., Urzúa M., A., Garrido, L., & Leiva, J. (2014). Validez y Confiabilidad de la versión Adaptada al Español de la Escala de Dificultades de Re­ gulación Emocional (DERS-E) en población chilena. Tera­ pia Psicológica, 32(1), 19-29. https://doi.org/10.4067/S0718-

48082014000100002

Harvey, A. G., Watkins, E., & Mansell, W. (2004). Cognitive behav­ ioural processes across psychological disorders: A transdiag­ nostic approach to research and treatment. Oxford University Press.

Hermans, M., Korrelboom, K., & Visser, S. (2015). A Dutch ver­ sion of the Overall Anxiety Severity and Impairment Scale (OASIS): Psychometric properties and validation. Journal of Affective Disorders, 172, 127-132. https://doi.org/10.1016/j. jad.2014.09.033 Hervás, G., & Jódar, R. (2008). Adaptación al castellano de la Es­ cala de Dificultades en la Regulación Emocional. Clínica y Salud, 19(2), 139-156.

Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit index­ es in covariance structure analysis: Conventional cri­ teria versus new alternatives. Structural Equation Mod­ eling: A Multidisciplinary Journal, 6(1), 1-55. https://doi. org/10.1080/10705519909540118 Instituto Nacional de la Juventud. (2024). Hablemos de todo. https://hablemosdetodo.injuv.gob.cl/ Javaid, S. F., Hashim, I. J., Hashim, M. J., Stip, E., Samad, M. A., & Ahbabi, A. A. (2023). Epidemiology of anxiety disor­ ders: Global burden and sociodemographic associations. Middle East Current Psychiatry, 30(1), article 44. https://doi. org/10.1186/s43045-023-00315-3 Kline, R. B. (2016). Principles and practice of structural equation modeling (4th ed.). Guilford Press.

Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ- 9: Validity of a brief depression severity measure. Jour­ nal of General Internal Medicine, 16(9), 606-613. https://doi. org/10.1046/j.1525-1497.2001.016009606.x Lovibond, P. F., & Lovibond, S. H. (1995). The structure of nega­ tive emotional states: comparison of the Depression Anx­ iety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behaviour Research and Therapy, 33(3), 335-343. https://doi.org/10.1016/0005-7967(94)00075-u Ministerio de Salud. (2016-2017). Encuesta Nacional de Salud (ENS) 2016-2017: Primeros resultados. Ministerio de Salud Gobierno de Chile.

https://www.minsal.cl/wp-content/ uploads/2017/11/ENS-2016-17_PRIMEROS-RESULTADOS.pdf Mira, A., González-Robles, A., Molinari, G., Miguel, C., Díaz- García, A., Bretón-López, J., García-Palacios, A., Quero, S., Baños, R., & Botella, C. (2019). Capturing the severity and impairment associated with depression: The Overall De­ pression Severity and Impairment Scale (ODSIS) validation

9/9 Psychometric properties of the Overall Anxiety and Depression Severity and Impairment Scales in a Chilean sample in a Spanish clinical sample. Frontiers in Psychiatry, 10, 180. https://doi.org/10.3389/fpsyt.2019.00180 Moore, S. A., Welch, S. S., Michonski, J., Poquiz, J., Osborne, T. L., Sayrs, J., & Spanos, A. (2015). Psychometric evaluation of the Overall Anxiety Severity and Impairment Scale (OA­ SIS) in individuals seeking outpatient specialty treatment for anxiety-related disorders. Journal of Affective Disorders, 175, 463-470. https://doi.org/10.1016/j.jad.2015.01.041 Muthén, L. K., & Muthén, B. O. (2017). Mplus: Statistical Analysis with Latent Variables: User’s Guide (Version 8). Authors. Norman, S. B., Campbell-Sills, L., Hitchcock, C. A., Sullivan, S., Rochlin, A., Wilkins, K. C., & Stein, M. B. (2011). Psychomet­ rics of a brief measure of anxiety to detect severity and im­ pairment: The Overall Anxiety Severity and Impairment Scale (OASIS). Journal of Psychiatric Research, 45(2), 262-268. https://doi.org/10.1016/j.jpsychires.2010.06.011 Norman, S. B., Cissell, S. H., Means-Christensen, A. J., & Stein, M. B. (2006). Development and validation of an Overall Anxiety Severity and Impairment Scale (OASIS). Depression and Anxiety, 23(4), 245-249. https://doi.org/10.1002/da.20182 NCAR - Research Applications Laboratory. (2015). verification: Weather Forecast Verification Utilities [Software de com­ putador]. R package version 1.42. https://CRAN.R-project. org/package=verification Osma, J., Martínez-Loredo, V., Díaz-García, A., Quilez-Orden, A., & Peris-Baquero, Ó. (2022). Spanish Adaptation of the Overall Anxiety and Depression Severity and Impairment Scales in University Students. International Journal of Envi­ ronmental Research and Public Health, 19(1), 345. https://doi. org/10.3390/ijerph19010345 Osma, J., Quilez-Orden, A., Suso-Ribera, C., Peris-Baquero, O., Norman, S. B., Bentley, K. H., & Sauer-Zavala, S. (2019). Psy­ chometric properties and validation of the Spanish ver­ sions of the overall anxiety and depression severity and impairment scales. Journal of Affective Disorders, 252, 9-18. https://doi.org/10.1016/j.jad.2019.03.063 Posit Team. (2024). RStudio: Integrated Development Environment for R [Software de computador]. Posit Software, PBC, Bos­ ton, MA. http://www.posit.co/.

Rice, M. E., & Harris, G. T. (2005). Comparing effect sizes in fol­ low-up studies: ROC Area, Cohen’s d, and r. Law and Human Behavior, 29(5), 615-620. https://doi.org/10.1007/s10979-005-

6832-7

Robin, X., Turck, N., Hainard, A., Tiberti, N., Lisacek, F., Sanchez, J.-C., & Müller, M. (2011). pROC: An open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinfor­ matics, 12, 77. https://doi.org/10.1186/1471-2105-12-77 Rojas, R. L., Cremades, C. F., Celleri, M., & Garay, C. J. (2023). Overall Anxiety Severity and Impairment Scale (OASIS) and Overall Depression Severity and Impairment Scale (ODSIS): Adaptation and validation in Buenos Aires, Argen­ tina. Clinical Psychology in Europe, 5(2), e10451. https://doi. org/10.32872/cpe.10451 Rojas-Barahona, C. A., Zegers, P. B., & Förster, C. E. (2009). La escala de autoestima de Rosenberg: validación para Chile en una muestra de jóvenes adultos, adultos y adultos ma­ yores. Revista Médica de Chile, 137(6), 791-800. https://doi. org/10.4067/S0034-98872009000600009 Román, F., Santibáñez, P., & Vinet, E. V. (2016). Uso de las Es­ calas de Depresión Ansiedad Estrés (DASS-21) como ins­ trumento de tamizaje en jóvenes con problemas clínicos. Acta de Investigación Psicológica, 6(1), 2325-2336. https://doi. org/10.1016/S2007-4719(16)30053-9 Rosenberg, M. (1965). Society and the adolescent self-image. Princeton University Press.

Ruscio, A. M., Hallion, L. S., Lim, C. C. W., Aguilar-Gaxiola, S., Al-Hamzawi, A., Alonso, J., Andrade, L. H., Borges, G., Bro­ met, E. J., Bunting, B., Caldas de Almeida, J. M., Demytten­ aere, K., Florescu, S., de Girolamo, G., Gureje, O., Haro, J. M., He, Y., Hinkov, H., Hu, C., de Jonge, P., … & Scott, K. M. (2017). Cross-sectional comparison of the epidemiology of DSM-5 generalized anxiety disorder across the globe. JAMA Psy­ chiatry, 74(5), 465-475. https://doi.org/10.1001/jamapsychia­ try.2017.0056 Sandora, J., Novak, L., Brnka, R., van Dijk, J. P., Tavel, P., & Ma­ linakova, K. (2021). The Abbreviated Overall Anxiety Se­ verity and Impairment Scale (OASIS) and the Abbreviated Overall Depression Severity and Impairment Scale (OD­ SIS): Psychometric properties and evaluation of the Czech versions. International Journal of Environmental Research and Public Health, 18(19), 10337. https://doi.org/10.3390/ ijerph181910337 Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin, 86(2), 420-

428. https://doi.org/10.1037/0033-2909.86.2.420

Stein, D. J., Lim, C. C. W., Roest, A. M., de Jonge, P., Aguilar-Gax­ iola, S., Al-Hamzawi, A., Alonso, J., Benjet, C., Bromet, E. J., Bruffaerts, R., de Girolamo, G., Florescu, S., Gureje, O., Haro, J. M., Harris, M. G., He, Y., Hinkov, H., Horiguchi, I., Hu, C., Karam, A., …& WHO World Mental Health Survey Collabora­ tors. (2017). The cross-national epidemiology of social anx­ iety disorder: Data from the World Mental Health Survey Initiative. BMC Medicine, 15(1), 143. https://doi.org/10.1186/ s12916-017-0889-2 Vicente, B., Saldivia, S., & Pihán, R. (2016). Prevalencias y bre­ chas hoy; salud mental mañana. Acta Bioethica, 22(1), 51-61. https://doi.org/10.4067/S1726-569X2016000100006 World Health Organisation. (2017). Depression and other common mental disorders. Global Health Estimates. World Health Or­ ganisation.

World Health Organisation. (2023). Anxiety disorders. World Health Organisation. https://www.who.int/news-room/ fact-sheets/detail/anxiety-disorders World Medical Association. (1964). Declaration of Helsinki: Eth­ ical principles for medical research involving human subjects. World Medical Association.

Xia, Y., & Yang, Y. (2019). RMSEA, CFI, and TLI in structural equation modeling with ordered categorical data: The story they tell depends on the estimation methods. Behav­ ior Research Methods, 51, 409-428. https://doi.org/10.3758/ s13428-018-1055-2