Abstract
BACKGROUND: Constipation is correlated with cognitive impairment; however, the association of constipation symptoms with cognitive domains remains unclear. This study aimed to investigate this association. METHODS: Participants aged 65 and older underwent neuropsychological, clinical, and laboratory examinations. Clinicians diagnosed constipation using the Rome IV criteria. Multivariate logistic regression assessed the odds ratios (ORs) and 95% confidence intervals (CIs) for mild cognitive impairment (MCI) and multi-domain cognitive impairments in relation to constipation and its specific symptoms. Mediation analysis was conducted to examine the effects of depressive symptoms. RESULTS: Constipation was diagnosed in 9,417 participants without dementia [mean (standard deviation) age: 72.0 (5.6) years], while constipation symptoms were recorded in 3,344 individuals [mean (standard deviation) age: 72.6 (5.5) years]. Of the overall population, 1,895 (20.1%) were diagnosed with constipation. Constipation was associated with a higher MCI risk (OR: 1.177, 95% CI: 1.047-1.323), worse performance on language (OR: 1.133, 95% CI: 1.011-1.270), and executive function (OR: 1.386, 95% CI: 1.130-1.701). A higher MCI risk was associated with constipation symptoms: bowel movements every 3 or more days (OR: 1.391, 95% CI: 1.011-1.914), defecation difficulty (OR: 1.278, 95% CI: 1.002-1.629), and dry stools (OR: 1.326, 95% CI: 1.027-1.711). Prolonged bowel movements increased the risk of both memory and language impairment, but not MCI. Defecation difficulty was associated with memory impairment (OR: 1.309, 95% CI: 1.003-1.709), dry stools with language impairment (OR: 1.396, 95% CI: 1.088-1.791), and bowel movements every other day with executive impairment (OR: 1.761, 95% CI: 1.151-2.693). Depression mediated the association of constipation with global cognitive and language function. CONCLUSION: In the non-demented stage, constipation and its symptoms were associated with MCI and multi-domain cognitive impairments. These associations, along with depressive symptoms, should be further evaluated in large-scale population screenings to benefit cognitive impairment management.