FACTORS PREDICTING 2‐YEAR COGNITIVE DECLINE IN NONAGENARIANS WITHOUT COGNITIVE IMPAIRMENT AT BASELINE: THE NONASANTFELIU STUDY

To the Editor: Few studies have prospectively evaluated predictors of cognitive decline in the oldest old.1 This report examines the changes in cognitive function observed over a period of 24 months in a cohort of nonagenarians without cognitive decline at baseline. The data were taken from the NonaSantfeliu study, a population-based study of nonagenarian inhabitants in the town of Sant Feliu de Llobregat (Barcelona, Spain). The survey has been described in detail elsewhere.2,3 In brief, contact was made with all 305 nonagenarian residents, 61% of whom responded (n=186 participants). Cognitive status was evaluated using the version of the Mini-Mental State Examination (MMSE) that has been adapted and validated for use in Spain, known as the Mini-Examen Cognitivo (MEC),4 which provides a score of up to 35 (a score of ≤23 indicates cognitive impairment). Functional status was measured using the Barthel Index (BI)5 for activities of daily living (ADLs) and the Lawton and Brody Index (LI)6 for instrumental ADLs. Presbyopia was determined using the appropriate Snellen charts. Hearing competency was measured using the Whisper test. The Charlson Comorbidity Index was used to measure overall comorbidity.7 Information about diagnosis of hypertension, diabetes mellitus, dyslipidemia, atrial fibrillation, and anemia; the existence of a previous stroke and depression; and chronic drug prescription was collected. Nonagenarians without cognitive decline at baseline were defined as those scoring 24 points or more on the MEC. At baseline, 102 subjects (56%) had MEC values greater than 23. Individuals were followed up for 24 months or until they died, whichever occurred first. Three of 102 nonagenarians were lost to follow-up, and 21 of the remaining 99 died. At 2-year follow-up, unimpaired subjects at baseline were categorized as showing decline if their MEC scores had dropped to less than 24. Categorical variables are reported as proportions. The Student t test, the chi-square or Fisher exact text, and multiple logistic regression analysis adjusted for baseline age and sex were performed. P<.05 indicated statistical significance. The final sample consisted of 59 women (75.6%) and 19 men. At baseline, mean age ± standard deviation was 92.4 ± 2.9, mean BI was 78.3 ± 20.0, and mean MEC score was 30.3 ± 3.0. After a 24-month follow-up period, mean BI score had dropped to 67.1 (P<.0001) and MEC to 23.1 (P<.001). Thirty patients (38.4%) showed a drop in MEC score below 24. MEC score fell by a mean of 14.2 points. In 33 (68%) of the remaining 48 subjects, there was a drop in MEC score but not below 24. Table 1 shows all the variables that were tested for their associations with mortality in a bivariate analysis. Using multiple logistic regression analysis, the risk of cognitive decline was associated with lower MEC scores at baseline (odds ratio=1.23, 95% confidence interval=1.19–1.73; P<.001) and the presence of near vision impairment (odds ratio=4.43, 95% confidence interval=1.2–15.3; P=.01). Individuals with a poorer baseline total MMSE score are more likely to develop cognitive impairment than their counterparts who perform better.8 The results of the current study confirm this finding even in the oldest old, such as nonagenarians, although cognitive scores earlier in their life were not known. Visual impairment is common in nonagenarians (38% in the entire cohort of the NonaSantfeliu study).3 Here, near-vision impairment was significantly associated with cognitive decline, as previously reported.9 Near-vision impairment has also been associated with functional decline when vision-dependent items were omitted from the MMSE.10 In this study, ophthalmological examination was not performed, and thus the cause for the impaired vision could not be identified. Visual impairment has been found to influence the level and quality of interactive experiences in older adults.9 The reduced level of participation in intellectually stimulating activities may involve a decrease in brain reserve to maintain cognitive function in later life.10 Several limitations should be considered when interpreting the results. First, the sample size, particularly in terms of the number of men, was small. Another limitation was that a subanalysis of MEC items was not performed to detect mortality differences according to the areas of cognitive function (e.g., orientation, attention, calculation). Neither was the apolipoprotein E genotype assessed. The exclusion of cases that died during the interval of observation might have led to an underestimate or overestimate of change predictors. Finally, dementia was not formally assessed with more-detailed test battery. In conclusion, near-vision impairment and MEC values were strong independent predictors of 24-month cognitive decline in nonagenarians without cognitive impairment at baseline. Financial Disclosure: None. Author Contributions: Francesc Formiga: study concept and design, drafting of the manuscript, and study supervision. Assumpta Ferrer and Jordi Gascon: acquisition of data. Ramon Rene: drafting of the manuscript. Antoni Riera: study concept and design, drafting of the manuscript. Ramon Pujol: critical revision of the manuscript. Sponsor's Role: None.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.