RESUMO
OBJECTIVE: To identify factors predictive of rapid cognitive decline (RCD) among elderly subjects aged 75 or over suffering from dementia. METHODS: The analysis concerned 250 patients drawn from the 'Sujet Agé Fragile--Evaluation et Suivi' (SAFES) cohort, presenting a dementia syndrome at inclusion and followed-up for at least 1 year. RCD was defined as the loss of at least 3 points on the Mini-Mental State Examination (MMSE) in the follow-up period of 12 months. All patients underwent a standardised geriatric evaluation. Logistic regression was used to identify factors predictive of RCD. RESULTS: In the study sample, 84 patients (33.6%) presented RCD. The factors identified in multivariate analysis as predictive of RCD were: high level of education (OR = 7.8, 95% CI = [1.9-31.2], p = 0.004), risk of depression (OR = 1.8, 95% CI = [1.02-3.18], p = 0.048, and the initial MMSE score (OR = 1.1, 95% CI = [1.0-1.2], p = 0.002). Among subjects with a main caregiver (n = 177), the predictive factors were malnutrition or risk thereof (OR = 4.2, 95% CI = [1.3-14.1], p = 0.02), risk of a fall (OR = 2.6, 95% CI = [1.1-6.1], p = 0.03, caregiver burden (OR = 2.6, 95% CI = [1.1-6.4], p = 0.04) and initial MMSE score (OR = 1.1, 95% CI = [1.0-1.3], p = 0.004). CONCLUSIONS: As soon as dementia is diagnosed in elderly subjects, information should be collected about the subject's socioeconomic status, nutritional status, risk of falling, mood state, and caregiver burden. This would enable the provision of appropriate therapeutic care, and make it possible to adapt follow-up in case of a risk of accelerated cognitive deterioration.