RELIABILITY

Internal Consistency

Rush et al. (1996) reported the psychometric properties of the IDS-C and IDS-SR using a sample of 456 subjects (338 adult outpatients with MDD, and 118 normal controls). Cronbach's alpha (Cronbach 1951), was 0.94 for both the IDS-C and IDS-SR for the complete sample (n=456), and 0.67 and 0.77 for the IDS-C and IDS-SR, respectively, for the sample of MDD patients (n=338). The results are contrasted with 0.89 for the HRSD₁₇ and 0.94 for the BDI₂₁ for the complete sample, and 0.53 for the HRSD₁₇ and 0.83 for the BDI₂₁ for the sample of MDD patients. In a sample of 68 newly admitted adult inpatients with MDD, Corruble et al. (1999) found a Cronbach's alpha coefficient of 0.75 and 0.79 for the IDS-C and the IDS-SR respectively. The authors also reported a Cronbach's alpha of 0.80 for the MARDS, and 0.95 for SCL₉₀ depression subscale in the same sample. Biggs et al. (2000) reported a Cronbach's alpha of 0.82 for the IDS-C and 0.83 for the IDC-SR in a sample of 62 adult patients with MDD (28 inpatients, 34 outpatients).

In a study of 596 adult outpatients with chronic, nonpyschotic MDD, Rush et al (2003) reported Cronbach's alpha for the IDS-SR, QIDS-SR, HRSD₁₇, HRSD₂₁, and HRSD₂₄. This study revealed high internal consistency for all the scales; 0.92 for the IDS-SR; 0.86 for the QIDS-SR; 0.83 for the HRSD₁₇; 0.84 for the HRSD₂₁; and 0.88 for the HRSD₂₄.

Recently, Trivedi et al. (2004) reported the psychometric properties of the QIDS-C and QIDS-SR, as well as those of the original IDS-C and IDS-SR. The authors assessed 544 patients with MDD and 402 with Bipolar Disorder (BD) using the QIDS-C, QIDS-SR, IDS-C and IDS-SR. Cronbach's alpha was 0.85 (QIDS-C), 0.86 (QIDS-SR), 0.90 (IDS-C), 0.92 (IDS-SR) for the MDD patients, and 0.81 (QIDS-C), 0.89 (IDS-C) for the patients with BD.

GENERALIZABILITY

Both versions of the IDS have been used in nonpsychotic and psychotic MDD (Rush et al. 1986, 1996, 2003, Trivedi et al. 2001, 2004), postpartum depression (Yonkers et al. 2001), dysthymic disorder (Rush et al. 1987, 2005), minor depression (Rappaport et al. 2002, Nina et al. 2002, Judd et al. 2004), bipolar disorder (Rush et al. 1986, Denicoff et al. 2000, Suppes et al. 2002, Trivedi et al. 2004), as well as in patients with depression comorbid with cancer (Jenkins et al. 1998) and asthma (Brown et al. 2001). The IDS and QIDS have been used in a variety of research and clinical settings, including inpatient and outpatient psychiatric clinics, and primary care settings. They are under investigation in elderly and adolescent patients. These assessments have been widely used in both clinical and research settings, and in government, foundation and industry sponsored studies. For a partial list of published uses of the IDS and QIDS see Table 1.