Population attributable fractions for risk factors for dementia in seven Latin American countries: an analysis using cross-sectional survey data

Abstract

Background: Approximately 40% of dementia cases worldwide are attributable to 12 potentially modifiable risk factors. However, the proportion attributable to these risks in Latin America remains unknown. We aimed to determine the population attributable fraction (PAF) of 12 modifiable risk factors for dementia in seven countries in Latin America.

Methods: We used data from seven cross-sectional, nationally representative surveys with measurements of 12 modifiable risk factors for dementia (less education, hearing loss, hypertension, obesity, smoking, depression, social isolation, physical inactivity, diabetes, excessive alcohol intake, air pollution, and traumatic brain injury) done in Argentina, Brazil, Bolivia, Chile, Honduras, Mexico, and Peru. Data were collected between 2015 and 2021. Sample sizes ranged from 5995 to 107 907 participants (aged ≥18 years). We calculated risk factor prevalence and communalities in each country and used relative risks from previous meta-analyses to derive weighted PAFs. Pooled PAFs for Latin America were obtained using random effect meta-analyses.

Findings: The overall proportion of dementia cases attributed to 12 modifiable risk factors varied across Latin American countries: weighted PAF 61·8% (95% CI 37·9-79·5) in Chile, 59·6% (35·8-77·3) in Argentina, 55·8% (35·7-71·5) in Mexico, 55·5% (35·9-70·4) in Bolivia, 53·6% (33·0-69·3) in Honduras, 48·2% (28·1-63·9) in Brazil, and 44·9% (25·8-61·2) in Peru. The overall PAF for dementia was 54·0% (48·8-59·6) for Latin America. The highest weighted PAFs in Latin American countries overall were for obesity (7%), physical inactivity (6%), and depression (5%).

Interpretation: The estimated PAFs for Latin American countries were higher than previous global estimates. Obesity, physical inactivity, and depression were the main risk factors for dementia across seven Latin American countries. These findings have implications for public health and individually targeted dementia prevention strategies in Latin America. Although these results provide new information about Latin American countries, demographics and representativeness variations across surveys should be considered when interpreting these findings.