Consistencia interna, estructura factorial y confiabilidad del constructo de la Escala de Yesavage para depresión geriátrica (GDS-15) en Cartagena (Colombia) / Internal consistency and exploratory factorial analysis of the Yesavage Geriatric Depression Scale (GDS-15) in Cartagena (Colombia)

Objetivo: Determinar la consistencia interna, la estructura factorial y la confiabilidad del constructo de la escala de Yesavage para depresión geriátrica (GDS-15) en adultos mayores asistentes a centros comunitarios de Cartagena (Colombia). Materiales y métodos: Estudio de validación de una escala para tamizaje sin criterio de referencia. Participaron 185 adultos mayores sin deterioro cognoscitivo, entre 65 y 95 años, con media de 75,6 (DT=6,9); y 57,3% mujeres. Se determinó consistencia interna mediante la fórmula 20 de Kuder-Richardson (KR-20) y se realizó análisis factorial exploratorio con el método de componentes principales. Resultados: La consistencia interna fue 0,783, la solución factorial mostró dos factores (estado de ánimo deprimido y capacidad de disfrute) que explicaban el 37,3% de la varianza y la confiabilidad del constructo fue 0,700. Conclusiones: La GDS - 15 muestra una aceptable consistencia interna y confiabilidad del constructo. No obstante, la estructura factorial bidimensional no es completamente satisfactoria. Es posible que la utilidad de la GDS-15 en personas de baja escolaridad sea limitada.

Objective: To determine the internal consistency, factor structure and construct reliability of the Yesavage Geriatric Depression Scale (GDS-15) among older people a ending community centers in Cartagena (Colombia). Materials and method: Validation study of a screening scale without a gold standard. A group of 185 adults participated in this research. Ages were between 65 and 95 years (Mean = 75.6, SD=6.9); and 57.3% were women. Internal consistency was determined by Kuder-Richardson’s formula 20, factor structure was explored by factor analysis with principal component method, and construct reliability was computed by Anderson-Gerbing’s formula. Results: Internal consistency was 0,783 and the factor solution showed two factors (depressed mood and life enjoy) that explained 37.3% of the variance, and construct reliability was 0,700. Conclusions: The GDS-15 showed acceptable internal consistency and construct reliability. However, its two-dimensional factor structure is not satisfactory. It is possible that the GDS- 15 exhibits low ability to identify depressive episodes in elders with low scholarship...