Abstract Background Global initiatives call for further understanding of the impact of inequity on aging across underserved populations. Previous research in LMICs presents limitations in assessing combined sources of inequity and outcomes (i.e., cognition and functionality). In this study, we assessed how social determinants of health (SDH), cardiometabolic factors (CMF), and other medical/social factors predict cognition and functionality in an aging Colombian population. Method We ran a cross‐sectional study that combined theory‐ (structural equation models and population attributable risk scores), and data‐driven (machine learning) approaches in a population‐based study (N = 23,694; M = 69.8 years) to assess the best predictors of cognition and functionality. Result We found that a combination of SDH and CMF accurately predicted cognition and functionality, although SDH was the stronger predictor. Cognition was predicted with the highest accuracy by SDH, followed by demographics, CMF, and other factors. A combination of SDH, age, CMF, and additional physical/psychological factors were the best predictors of functional status. Conclusion Results highlight the role of inequity in predicting brain health and advancing solutions to reduce the cognitive and functional decline in LMICs.