Olfactory impairment associated with cognitive decline: findings from cognitive domain assessments among older adults in West Jakarta, Indonesia.

BACKGROUND

Olfactory dysfunction has emerged as a potential early marker of neurodegenerative diseases. Despite growing interest, population-based evidence on its association with cognitive domains remains limited, particularly in low- and middle-income countries. This study examined the association between olfactory impairment and multidomain cognitive performance in older Indonesian adults.

METHODS

We conducted a cross-sectional study among 205 community-dwelling adults aged ≥ 60 years in West Jakarta. The cognitive assessments included the Indonesian version of the Montreal Cognitive Assessment (MoCA-Ina), verbal fluency, a modified Boston Naming Test (BNT), constructional praxis, word list recall, delayed memory, and word list recognition. Multivariable logistic and linear regression analyses were used to examine associations between olfactory function and cognitive performance, adjusting for demographic factors and chronic illnesses.

RESULTS

Probable dementia as assessed by the MoCA-Ina was significantly associated with increased odds of olfactory impairment (OR [95% CI] = 4.29 [1.87-9.85]), as were verbal fluency impairment (2.25 [1.15-4.39]) and the BNT (2.21 [1.12-4.38]). In linear regression models with continuous olfactory scores, cognitive impairment as assessed by the MoCA-Ina, verbal fluency, and the BNT was associated with lower olfactory scores, indicating poorer olfactory function (MoCA-Ina: β= -1.51, p < 0.001; verbal fluency: β= -0.87, p = 0.001; BNT: β= -1.00, p < 0.001).

CONCLUSIONS

Olfactory impairment is independently associated with both global cognitive function and specific deficits in verbal fluency and naming. These tasks primarily reflect language processing but may also involve executive control. These findings suggest that olfactory assessment may be useful in community-based cognitive screening, especially in resource-limited settings.