INTRODUCTION:
With population aging, dementias pose a major challenge to global public health, requiring the identification of potentially modifiable factors associated with their risk¹. Religiosity, particularly religious practice, remains underexplored in population-based studies, especially in Latin America². Thus, this study evaluated the association between religious practice and dementia among Brazilian adults aged 50 years or older.

METHODS:
Data from the second wave of the Brazilian Longitudinal Study of Aging (ELSI-Brazil³; 2019–2021) were used, including 9,856 participants aged ≥50 years. Dementia was defined as a z-score ≤ −1.5 on cognitive tests combined with impairment in at least one activity of daily living, or a cutoff score ≥3.4 on the Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE) for proxy respondents. Religious practice was measured using the Duke University Religion Index (P-DUREL) and categorized according to frequency of participation. Associations were estimated using Poisson regression with robust variance, applying the Wald test, with a 5% significance level (Stata 17.0).

RESULTS:
The prevalence of dementia was 5.73% (n = 444). Religious practice was categorized as: “more than once a week” (27.36%), “once a week” (32.95%), “two or three times a month” (16.94%), and “a few times a year or never” (22.75%). Dementia prevalence and its association with religious practice were as follows: more than once a week (2.04%), once a week (2.70%; PR = 1.32; 95% CI 0.88–1.98), two to three times a month (3.57%; PR = 1.74; 95% CI 1.12–2.71), and a few times a year or never (4.98%; PR = 2.43; 95% CI 1.64–3.60), with a statistically significant difference (p < 0.001).

CONCLUSIONS:
Religious practice was associated with dementia, with a progressive increase in prevalence as the frequency of religious practice decreased.