Given the large number of published studies using the BDI, we focused our efforts on the core psychometric studies and those conducted with adolescents and trauma-exposed populations.

1. Numerous studies have established the reliability and validity of the BDI-II in different populations and cultures. In adults, the BDI-II has been found to correlate with multiple measures of depression including the Center for Epidemiological Studies of Depression Scale (CES-D), Zung Self-Rating Depression Scale, the Beck Hopelessness Scale, and the Revised Hamilton Psychiatric Rating Scale for Depression (Beck, Steer, & Brown, 1996).
2. The BDI-II discriminates depressed from non-depressed patients (Beck, Steer, & Brown, 1996; Sprinkle et al., 1992).
3. It has also been found to be sensitive to change with treatment, including in randomized trials with individuals who have experienced a trauma (Bryant, Moulds, Guthrie & Nixon, 2005) and those diagnosed PTSD (Ehlers et al., 2005).
4. Factor analysis of the BDI-II has generally identified a 2-factor structure in adult outpatient and non-clinical samples, measuring cognitive-affective and somatic depressive symptoms (Dozois, Dobson, & Ahnberg, 1998; Storch, Roberti, & Roth, 2004). Analyses with adult inpatients have identified a single hierarchical depression factor (Cole, Grossman, Prillman, & Hunsaker, 2003). Analyses of adolescents have identified different but related factor solutions (see below). One study involving a confirmatory factor analysis of the CES-D and the original BDI, failed to validate a single-factor model (Skorikov & Vandervoort, 2003). The authors suggested that the measures assess different underlying aspects of the construct of depression, with the CES-D assessing more of an affective component and the BDI assessing more of a cognitive component. The authors suggested that the measures not be used interchangeably since they may be assessing different aspects of depression. They also interpreted their findings as suggesting that the CES-D may be more effective in non-clinical populations.
5. A number of studies report that females score significantly higher than males do on the BDI in adult (Beck, Steer, & Brown, 1996) and adolescent populations (Kumar, Steer, Teitelman, & Villacis, 2002; Osman, Kopper, Guttierez, Barrios, & Bagge, 2004; Steer, Kumar, Ranieri, & Beck, 1998).
6. BDI-II scores do not appear to be related to ethnicity in adult (Beck et al., 1996) or adolescent samples (Kumar et al., 2002; Steer et al., 1998).

STUDIES WITH ADOLESCENTS

1. Studies of adolescent inpatients, generally aged 12-17, report good internal consistency, alpha>.90 for the total scale and >.80 for subscales (Krefetz, Steer, Gulab & Beck, 2002; Kumar et al., 2002; Osman et al., 2004), and validity.
2. BDI-II scores are correlated with scores on the Reynolds Adolescent Depression Scale, the Beck Hopelessness Scale, the Beck Anxiety Inventory, the MMPI-A, and the Suicidal Behaviors Questionnaire-Revised; and BDI-II scores discriminate between adolescents who do and do not meet DSM-IV criteria for a major depressive disorder (Krefetz et al., 2002; Kumar et al., 2002).
3. Confirmatory factor analyses with adolescent psychiatric inpatients (Osman et al., 2004) identified a 2-factor solution as the most parsimonious and interpretable. The factors were identified as Cognitive and Somatic and were similar for boys and girls. The authors report that the solution differed from that reported for adults in that the first factor contained both cognitive and affective symptoms.
4. Steer et al. (1998) examined the psychometrics of the BDI-II with adolescent outpatients and found good internal consistency. Through principal factor analysis, they identified a single second-order dimension of self-reported depression and three first-order factors. The authors claimed that only two of the first-order factors, Cognitive and Somatic-Affective, were generalizable. These two factors have been identified using the BDI-II with adult outpatients. They found no differences between Caucasians and non-Caucasians but did report significant correlations between age and BDI-II scores.

STUDIES WITH TRAUMA-EXPOSED INDIVIDUALS

The BDI has been used in numerous studies with trauma-exposed individuals. A PsychInfo search of “Beck Depression Inventory” or “BDI” AND “trauma” yielded 681 peer-reviewed journal articles (6/05). It has been used in samples of combat veterans, women who have experienced intimate partner violence and sexual abuse, and in numerous treatment outcome studies for PTSD.

1. The BDI has also been found to be sensitive to intervention effects in and randomized trials with individuals with diagnosed PTSD (e.g., Bryant, Moulds, Guthrie, & Nixon, 2005; Ehlers et al., 2005; Kubany et al., 2004). Individuals treated with interpersonal psychotherapy adapted for PTSD also show decreases in BDI-II scores following treatment (Bleiberg & Markowitz, 2005). Parents of children with PTSD symptoms related to sexual abuse and traumatic bereavement show decreases in BDI symptoms after participating in treatment with their children (Cohen, Deblinger, Mannarino & Steer, 2004; Cohen, Mannarino, & Knudsen, 2004).
2. Among women who have experienced intimate partner violence, those with comorbid PTSD and Major Depression show higher levels of symptomatology on the BDI-II than those with PTSD alone and those with no PTSD or Major Depression (Nixon, Resick, Nishith, 2004).
3. The BDI has also been used with individuals with vicarious traumatization with scores on a Secondary Trauma Scale related to higher levels of depression (Motta, Newman, Lombardo, & Silverman, 2004).

STUDIES WITH OTHER CULTURAL GROUPS AND DIVERSE POPULATIONS

(This is a sampling of the literature in this area. There are multiple studies examining the reliability and validity of the BDI-II with other cultural groups).

1. Leigh & Tolbert (2001) examined the reliability of the BDI-II with deaf college students and found good internal consistency (alpha=.88), split-half reliability (.76), and one-week test-retest reliability (.77).
2. Grothe, K.B., Dutton, G.R., Jones, GN., Bodenlos, J., Ancona, M., & Brantley, P.J. (2005) factor analyzed data from a low-income African American outpatient sample. Consistent with previous research conducted by Beck, they identified 2 first-order factors (somatic and cognitive) and one second-order factor (depression). They also found high internal consistency (alpha=.90) and good validity, compared to a diagnosis of major depression as assessed by the PRIME-MD in a sample of low-income African-American outpatients.
3. Contreras, S., Fernanedez, Senaida, Malcarne, V.L., Ingram, R.E., & Vaccarino, V.R. (2004) examined the reliability and validity of the BAI and BDI in a sample of 1,110 Latino and 2,703 Caucasian undergraduate students. Scales for both groups had good internal consistencies. They also found similar factor structures for both groups, providing evidence of factoral validity. Although they used the original BDI in this study, they suggested that results would generalize to the BDI-II given the overlap between the two.
4. Cardemil, Kim, Pinedo, & Miller (2005) found high internal consistence (alpha was .90-.92) and change in scores over the course of treatment for both English- and Spanish-speaking Latina women from a predominantly low-income sample.
5. Penley, Wiebe, & Nwosu (2003) examined the psychometrics of the Spanish translation of the BDI II in a sample of predominantly Hispanic adults undergoing medical treatment for hemodialysis, many of whom were of lower SES. They found good internal consistency (alpha=.92), and using confirmatory factor analysis, identified two first-order depression factors and one second-order general depression factor, similar to what has been reported in other samples. They reported that BDI-II scores were negatively correlated to SES and acculturation and positively correlated with disease severity. Bilingual participants completed both English and Spanish versions, with comparable scores across language administrations. However, 30% of bilingual participants would be placed in a different depressive category depending on whether their Spanish or English scores are used. These findings are especially important in light of a study using an earlier version of the BDI that reported item bias when Latinos completed a translated version of the BDI (Azocar, Areán, Miranda & Muñoz, 2001).
6. Carmody (2005) examined the psychometrics of the BDI-II with a diverse group of college students. He found similar psychometrics for the non-clinical sample, but results of his confirmatory factor analysis suggested that a 3-factor model, comprised of negative attitude, performance difficulty, and somatic dimensions, provided a better fit than the traditional 2-factor model.
7. Sanz, Perdigón, & Vásquez (2003) examined the psychometrics of the Spanish adaptation of the BDI-II with 470 non-clinical adults. They found good internal consistency and factoral validity, with factor analysis identifying a general dimension of depression and two related factors, cognitive-affective and somatic-motivational, similar to the factor structure reported in the BDI-II manual. The study also provides BDI-II community norms.
8. The BDI has also been found to be related to the Adolescent Dissociative Experiences Survey and to a measure of alexithymia in a sample of Turkish adolescents (Sayar, Kose, Grabe, & Murat, 2005).
9. Byrne, Stewart, & Lee (2004) examined the psychometrics of the Chinese Beck Depression Inventory-II with a sample of Hong Kong community adolescents. They conducted both exploratory and confirmatory factor analysis and found a 2nd order general factor of Depression and three first-order factors: Negative Attitude, Performance Difficulty, and Somatic Elements. Their findings replicate what has been found in Canadian, Swedish, and Bulgarian non-clinical adolescents, but are different from factor analyses conducted with inpatient and outpatient adolescents in the United States. They also reported good internal consistency, test-retest reliability, and convergent validity.
10. The psychometric properties of the Arabic version of the BDI-II has been examined with students aged 18-37 at the University of Bahrain. The authors suggest findings provide support for the BDI-II in this population (Al-Musawi, 2001).