To build a dementia classification model for middle-aged and elderly Brazilians, implemented in Python, combining variable selection and multivariable analysis, using low-cost variables with modification potential. Observational study with a predictive modeling approach using a cross-sectional design, aimed at estimating the chances of developing dementia, using data from the Brazilian Longitudinal Study of Aging (ELSI-Brazil), involving 9,412 participants. Dementia was determined based on neuropsychological assessment and informant-based cognitive function. Analyses were performed using Random Forest (RF) and multivariable logistic regression to estimate the risk of dementia in the middle-aged and elderly populations of Brazil. The prevalence of dementia was 9.6%. The highest odds of dementia were observed in illiterate individuals (Odds Ratio (OR) = 7.42), individuals aged 90 years or older (OR = 11.00), low weight (OR = 2.11), low handgrip strength (OR = 2.50), self-reported black skin color (OR = 1.47), physical inactivity (OR = 1.61), self-reported hearing loss (OR = 1.65), and presence of depressive symptoms (OR = 1.72). Higher education (OR=0.44), greater life satisfaction (OR=0.72), and being employed (OR=0.78) were protective factors. The RF model outperformed logistic regression, achieving an area under the ROC curve of 0.776, with a sensitivity of 0.708, a specificity of 0.702, an F1-score of 0.311, a G-means of 0.705, and an accuracy of 0.703. Conclusion: The findings reinforce the multidimensional nature of dementia and the importance of accessible factors for identifying vulnerable individuals. Strengthening public policies focused on promoting brain health can contribute significantly to the efficient allocation of resources in primary care and dementia prevention in Brazil
Dementia is defined by the World Health Organization (WHO) (1) as a progressive neurological condition characterized by impairment in multiple cognitive functions, including memory, reasoning, language, and the ability to perform everyday activities, exceeding the decline expected from normal aging. It is a clinical syndrome that may be triggered by various neurodegenerative pathologies, ultimately leading to the gradual and irreversible loss of nerve cells and to structural and functional brain impairment. Cognitive deterioration is often accompanied by behavioral, emotional, and motivational changes, which substantially affect individual autonomy and quality of life. The impact of dementia extends beyond the affected individual, generating physical, emotional, social, and economic consequences for families, caregivers, and healthcare systems.
In Brazil, the Ministry of Health estimates that approximately two million Brazilians are living with some form of dementia. However, preliminary data indicate that over 70% of those affected remain undiagnosed, which further heightens concern (2), especially considering that recent estimates show an overall dementia prevalence of 5.8% among people aged 60 and older. Prevalence is higher among women (6.8%) than men (4.6%), ranging from 3.2% in the 60-64 age group to 42.8% among individuals aged 90 and above, and is more prevalent among illiterate individuals (16.5%) compared to those with a university-level education or higher (2.1%) (3).
A transitional state between normal cognitive functioning and dementia is known as cognitive impairment, which is characterized by a decline greater than expected for a person’s age and educational level. It affects memory, attention, language, and reasoning abilities, but does not significantly interfere with daily activities (4). Estimates indicate a prevalence of 8.1% among individuals aged 60 and older. As with dementia, cognitive impairment is less common in men (6.8%) than in women (9.1%). Its prevalence remains stable across age groups from 60 to 79 but increases to 11.8% and 12% in the 80-84 and 85-89 age groups, respectively. Interestingly, it decreases to 10% among those aged 90 and over. Regarding educational level, prevalence ranges from 9.4% among illiterate individuals to 10.8% among those with a university-level education or higher (3).
In recent years, efforts have been made to identify potential risk factors and early intervention strategies aimed at reducing the likelihood of dementia onset and delaying its progression. It is plausible to assume that anticipating action regarding certain modifiable risk factors could postpone the onset of the disease, given that dementia has prodromal characteristics and may begin to manifest decades before clinical symptoms appear (5). In a global context, it is vital to consider the significant rise in dementia incidence in Low-and Middle-Income Countries (LMICs) compared to High-Income Countries (HICs) (6). The early and accurate identification of individuals at high risk for dementia plays a crucial role in the effective implementation of preventive strategies.
It is imperative to have an assessment that is clinically feasible for identifying individuals at high risk, especially considering that in Latin American countries such as Brazil, there are significant challenges related to dementia development indicators (6). These challenges are reflected in key characteristics, such as: (i) low income; and (ii) high levels of social inequality (6). These socioeconomic factors often pose barriers to accessing high-cost diagnostic tools such as neuroimaging or biomarkers, which directly affect early detection and effective monitoring. In light of these limitations, there is a pressing need to consider alternative approaches, using accessible variables for both clinical use and large-scale community-based studies.
The current epidemiological context demands a strong prioritization of dementia across all levels, from local settings to the global stage (7). Identifying modifiable factors that can be addressed in advance helps to promote brain health, especially considering that approximately 45% of dementia cases could be delayed or prevented through risk factor modification (6). This underscores the urgent need for innovation and research in monitoring these variables (8,9). Evidence shows that machine learning algorithms outperform traditional models in predictive performance within medical contexts, due to their ability to manage complex and non-linear relationships. However, traditional models offer advantages such as transparency and interpretability, which are essential in clinical research settings (10)(11)(12). In this regard, combining both approaches allows us to leverage the strengths of each when it comes to predicting health outcomes.
Currently, a number of models have been developed for predicting cognitive impairment and dementia (13)(14)(15)(16), particularly using a
This content is AI-processed based on open access ArXiv data.