Background: Disability and a decline in functional capacity are common in old age. Objective: To determine predictors of functional decline in nonagenarians’ basic activities of daily living (ADL) after 1 year of follow-up. Methods: A sample of 97 nonagenarians subjects not previously severely dependent (Barthel Index >59) was evaluated. This sample included 72 women (74.2%) and the mean (SD) age was 93.4 ± 2.7 years. The following data were collected: sociodemographic data, Barthel Index (BI), Lawton-Brody Index (LI), Mental State Examination (MEC), a short version of the Mini-Nutritional Assessment, comorbidity (Charlson Index), lower-extremity function, Gait Rating Scale from the Tinetti Performance-Oriented Mobility Scale and prevalent chronic diseases. Subjects who had a 10-point or higher decline in the BI in 1 year were compared to subjects who had no decline or a maximum decline of 9 points. In addition, subjects whose total BI score fell below 60 were compared to the group of subjects who maintained scores between 60 and 100. The Student’s t test, the χ2 or the Fisher’s exact test, and a multiple logistic regression analysis (with the identified risk factors of age and gender) were performed. Results: 39 nonagenarians experienced a 10-point or higher decline in the BI in 1 year. A lower LI score (p = 0.003) and visual impairment (p = 0.01) were associated with functional decline. The multiple regression analysis showed that there was a significant association with LI (odds ratio (OR) 0.74, confidence interval 95% 0.60–0.91, p < 0.005). The 18 nonagenarians who had a BI <60 had a BI >60 at baseline. Lower scores on the LI (p = 0.004) and on the MEC (p = 0.01), a history of a previous stroke (p = 0.009) and higher Charlson Index scores (p = 0.03) were associated with recently acquired, severe dependency. A multiple regression analysis showed a significant association between LI (OR 0.65, 0.47–0.89, p < 0.008) and a history of previous strokes (OR 3.39, 1.01–11.34, p < 0.04). Conclusions: According to the definition used to describe functional basic ADL decline, poor performance in instrumental ADL at baseline and a history of a stroke appear to be independent risk factors. Prevention strategies could be intensified in this subgroup of nonagenarians.