Methods: A cross-sectional validation study involving 219 elderlies with cognitive impairment was conducted using universal sampling method. Participants aged 60 years old and above, scored less than 11 for the short mini mental state examination (SMMSE), have sufficient command of the Malay or English language, residing in the government nursing homes and attending memory clinics in government hospitals in Klang Valley were sampled. The GDS-15 English version underwent adaptation, translation, face validation and field testing to produce the Malay version. Psychometric analysis was performed using the exploratory factor analysis and its internal consistency was examined.
Results: Mean age of participants were 72.61 ± 7.79. Majority were male (50.7%), Malay (50.2%), studied at least until primary school (71.7%) and stayed at nursing homes (50.2%). Construct validity showed significant Bartlett's test of sphericity (Chi-square = 1,340.058, P < 0.001) and Kaiser-Meyer-Olkin (KMO) test of 0.90. Factor loadings for each item in the depression domain were satisfactory ranging from 0.42 to 0.83. Factor loadings for each item in the psychosocial activities' domain were satisfactory ranging from 0.53 to 0.76. For the questionnaire reliability analysis, the total Cronbach's alpha for the final model was satisfactory, with and overall Cronbach's alpha of 0.89. Cronbach's alpha value for the depression and psychosocial activities domain was 0.861 and 0.80, respectively.
Conclusion: The GDS (Malay version) is a valid and reliable tool to screen for depression in elderly with cognitive impairment.