AIM: To investigate the associations between hearing loss and prevalent and incident mild cognitive impairment (MCI), dementia and MCI or dementia (all cases). METHODS: Cross-sectional and longitudinal analyses of baseline and follow-up data were performed in a population-based cohort. The baseline sample of 2,599 adults aged ≥55 included 1,515 cognitively normal subjects who were followed up to 8 years. Hearing loss at baseline was determined by the whispered voice test, and MCI and dementia by Mini-Mental State Examination screening, Clinical Dementia Rating scale, neurocognitive tests, MRI, and panel consensus diagnosis. RESULTS: Hearing impairment was associated with increased prevalence of dementia (odds ratio = 3.63, 95% confidence interval [CI] 1.16-11.4, p = 0.027) but not MCI alone or all cases of MCI or dementia, adjusted for sex, age, ethnicity, education, central obesity, hypertension, diabetes, dyslipidemia, smoking, alcohol, leisure time activity, cardiac diseases, and depressive symptoms. Among participants who were cognitively normal at baseline, those with hearing impairment were more likely to develop MCI or dementia (hazard ratio [HR] = 2.30, 95% CI 1.08-4.92, p = 0.032). Hearing loss was associated with elevated but statistically nonsignificant estimates of adjusted HR (1.85, 95% CI 0.78-4.40) for incident MCI alone. CONCLUSIONS: Hearing loss is independently associated with prevalent dementia and incident MCI or dementia.