Education Shows Strongest Link to Cognition in Older Rural Ecuadorians

In a rural corner of the southern Ecuadorian Andes, where more than 70 percent of the rural population lives in multidimensional poverty, researchers have found that education stands out as the factor most strongly and consistently tied to cognitive performance in older adults. The findings come from the EcuAging study, a cross-sectional analysis of 238 adults aged 55 and older living in farming communities in Loja province, published in the journal Ageing International.

The research team, led by Tamara Rodríguez Quintana and colleagues, measured global cognition using a modified Spanish version of the Mini-Mental State Examination (MMSE), a 30-item screening instrument adapted for Latin American populations. In this version, items such as the current season are replaced with the approximate time of day, and the words to be registered are paper, bicycle and spoon. Scores range from 0 to 30, with higher scores indicating better performance on tests of orientation, registration, attention and calculation, recall, language and visuoconstruction.

The participants lived in eight communities across the cantons of Calvas and Gonzanamá, near the Peruvian border. Data were collected over two weeks in June and July 2023 at mobile stations set up in the communities, with residents recruited through public health ministry campaigns, community leaders and snowball sampling. Of 268 unique participants enrolled in the broader cohort, 238 had complete data for all variables in this analysis. People with prior psychiatric or neurological diagnoses, inadequate vision or hearing for testing, or mobility limitations preventing attendance were excluded.

The sample averaged 70.2 years of age, with a range from 55 to 100. Most participants were women (58.0 percent) and had completed only primary education (58.4 percent), and nearly all self-identified as Mestizo (94.5 percent). Economic circumstances were stark: 73.9 percent reported a monthly household income of USD 100 or less, and when household size was taken into account, at least 66.8 percent had a per capita income below Ecuador’s national extreme poverty line for June 2023.

In fully adjusted linear regression models, educational attainment showed the strongest association with MMSE scores. Compared with participants without formal education, those who had completed primary school scored about 3.8 points higher (B = 3.76; 95% CI 2.49 to 5.03), those with secondary education about 4.6 points higher (B = 4.62; 95% CI 2.52 to 6.72), and those with tertiary education about 6.0 points higher (B = 5.98; 95% CI 3.53 to 8.44). Each additional year of age was associated with a score roughly one-fifth of a point lower (B = −0.19), and the fully adjusted model accounted for 46 percent of the variance in MMSE scores.

Mental health also mattered. Each additional depressive symptom, measured with the 15-item Geriatric Depression Scale, was associated with an MMSE score about one-fifth of a point lower (B = −0.22; 95% CI −0.42 to −0.02). Anxiety symptoms, measured with the GAD-7, showed no association with cognition. Among lifestyle factors, only physical activity emerged as significant: participants reporting more than two hours of daily physical activity scored about 2.6 points higher than those reporting none (B = 2.60; 95% CI 0.56 to 4.63). Because most participants were farmers, this item captured total daily activity, including agricultural and domestic work, but not its intensity. Alcohol, tobacco and coffee consumption were not associated with cognition in the adjusted models.

A sensitivity analysis sharpened the picture. When the researchers excluded the ten participants with MMSE scores of 11 or below, a range compatible with moderate-to-severe dementia, the associations with education, age and depressive symptoms persisted, though the education coefficients were attenuated. The physical activity association, however, did not survive. The authors note that reverse causation is plausible: people with better cognition may simply be more able to remain physically active, and the cross-sectional design cannot establish the direction of any of these relationships.

The study’s authors also raise an important measurement caveat. The MMSE, like most brief cognitive screening tests, relies on school-like skills such as reading, writing, serial subtraction and figure copying. People with little or no formal schooling may score lower on these items without being impaired in daily life, and in rural settings cognitive abilities are often expressed in tasks the test does not capture, such as planning a harvest or identifying local plants and animals. Although the Spanish version used was validated for older Latin American adults, it has not been validated in rural Andean populations with very little schooling, and education-adjusted norms do not exist for Ecuador. Part of the education association may therefore reflect measurement bias rather than true differences in cognition, and the authors suggest instruments with lower educational load, such as the Rowland Universal Dementia Assessment Scale, for future work.

Several further limitations temper the conclusions. The cross-sectional design precludes causal inference, exposures were self-reported and may be less accurate among participants with cognitive impairment, and the single-item physical activity measure ignores intensity. The MMSE shows floor and ceiling effects, with 4.2 percent of participants scoring 11 or below and 3.4 percent achieving the maximum. Recruitment through community campaigns and snowball sampling, combined with the fact that excluded participants had lower MMSE scores, suggests the sample may over-represent healthier and more mobile older adults. Only one participant identified as Indigenous, so ethnicity differences could not be examined, and risk factors such as hypertension, diabetes and traumatic brain injury were not accounted for.

Even so, the authors argue the findings point to practical steps within existing rural services. Depressive symptoms could be routinely screened in rural primary care using the GDS-15, which is already part of Ecuador’s national comprehensive geriatric assessment, alongside accessible referral pathways for treatment. Community programmes for older adults could include cognitively stimulating activities adapted to low literacy, such as adult literacy and numeracy classes and intergenerational activities. In the longer term, they suggest, narrowing rural-urban gaps in schooling may reduce the burden of cognitive impairment among future generations of older Ecuadorians, though longitudinal studies using culturally appropriate cognitive measures are needed to confirm these cross-sectional associations.

Subject of Research: Associations of sociodemographic, lifestyle and mental health factors with global cognition in older adults in rural Ecuador

Article Title: Is Global Cognition Associated with Sociodemographic and Lifestyle Factors, and with Mental Health in Older Adults Living in Rural Ecuador? Insights from the EcuAging Study

Article References: Rodríguez Quintana, T., Massa, F., Marroig, A., Grijalva, M. J., Vinueza-Fernández, I., Hurtado-Encalada, A. T., Palacios Piedra, J. D., & Muniz-Terrera, G. (2026). Is Global Cognition Associated with Sociodemographic and Lifestyle Factors, and with Mental Health in Older Adults Living in Rural Ecuador? Insights from the EcuAging Study. Ageing International, 51(4), Article 46. https://doi.org/10.1007/s12126-026-09685-5

Image Credits: AI Generated

DOI: 10.1007/s12126-026-09685-5

Keywords: global cognition, MMSE, older adults, rural Ecuador, education, depressive symptoms, physical activity, EcuAging study, cross-sectional, Loja province, cognitive reserve, low-income populations