Validación de la escala de depresión de Glasgow en población colombiana con discapacidad intelectual

Abstract

Objective

To determine the reliability, construct and criterion validity of the Glasgow Depression Scale (GDS) in Colombian people with intellectual disability (ID).

Methods

One hundred and twenty participants who underwent the GDS and a clinical interview for the diagnosis of depressive disorders were included. To determine the validity of the convergent construct, the Hospital Anxiety and Depression Scale (HADS) for people with ID was applied to a subgroup, and for the evaluation of test–retest reproducibility, a new application of the GDS was made 7–10 days later.

Results

In the confirmatory factorial analysis, it was found that the unidimensional structure had an adequate fit, but it was better for the two-dimensional bifactorial structure: “ideas and depressive mood” and “changes in energy”. It had good internal consistency (Cronbach's alpha: 0,85 (95% CI: 0.81–0.89). In test–retest reproducibility, the intraclass correlation coefficient was 0.56 (95% CI: 0.34–0.72) for the total scale. Convergent validity was demonstrated by the correlation of 0.6 of the GDS with the HADS, and there is discriminative validity by the difference in median scores between patients with and without a depressive episode (23 vs. 18, P = 0.0001). The area under the receiver operating characteristic curve for diagnosis of depression was 0.80 (95% CI: 0.70–0.90) and the sensitivity was 95% (95% CI: 75.1–99.9) for a cutoff point of 18.

Conclusion

The GDS is reliable and validated for depression screening in Colombian people with ID.