[Validation of a DSM-5-based screening test using digital phenotyping in the Russian population]

To evaluate the validity of a depression and anxiety screening test based on DSM-5 diagnostic criteria to identify cases of these conditions simultaneously assessed with the validated Hospital Anxiety and Depression Scale (HADS) in a population sample by digital phenotyping.

This cross-validation study included 5.116 respondents (mean age 36.9 (9.8)), of which 49.4% (2526) were women. The depression and anxiety screening test was done in electronic form and based on the DSM-5 diagnostic criteria for major depressive disorder and generalized anxiety disorder. The validated HADS scale was used as a standard test. The categories of depression (HADS-D) and anxiety (HADS-A) phenotypes were formed with a cutoff of ≥8 points and ≥11 points. The main parameters of the validity of the screening test were calculated, including accuracy (Ac), sensitivity (Sn) and specificity (Sp) with their 95% confidence intervals [CI].

The prevalence of current depression and anxiety according to the screening test was 7.8% (400) and 12.5% (639), respectively. The prevalence of lifetime depression was 25.9% (1327). For the HADS-D depression subscale with cut-offs of ≥11 and ≥8 points, the prevalence of depression was 3.4% (174) and 15% (766), respectively. For the HADS-A anxiety subscale with cut-offs of ≥11 and ≥8 points, the prevalence of anxiety was 8.9% (456) and 31.8% (1628), respectively. For HADS-D and HADS-A with a cutoff of ≥11 points, the parameters of current depression were Ac=92%, Sn=47% (CI 95% [39-54]), Sp=94% (CI 95% [93-94]), lifetime depression - Ac=75%, Sn=63% (CI 95% [56-70]), Sp=75% (CI 95% [74-77]) and current anxiety - Ac=88%, Sn=54% (CI 95% [50-59]) and Sp=92% (CI 95% [90-92]). For HADS-D and HADS-A with a cutoff of ≥8 points, the parameters of current depression were Ac=86%, Sn=30% (CI 95% [27-33]), Sp=96% (CI 95% [95-97]), lifetime depression - Ac=74%, Sn=51% (CI 95% [48-55]), Sp=75% 79% (CI 95% [77-80]) and current anxiety - Ac=75%, Sn=31% (CI 95% [29-33]), Sp=96% (95% CI [95-97]).

The high Ac and Sp of this test allows it to be used for screening purposes to identify (but not exclude) cases of depression and anxiety in the population. However, further studies are needed to validate the screening test using a diagnostic interview with a physician.