Journal of Gerontology: MEDICAL SCIENCES

AGE-RELATED hearing loss is highly prevalent among older adults ( 1 – 4). It features among the leading causes of years lived with disability and is considered a substantial contributor to global burden of disease ( 5). Cross-sectional studies have identifi ed diabetes ( 6, 7), cardiovascular dis-ease, hypertension, and blood pressure ( 8) as risk factors for hearing loss. Hearing loss has also been linked with poor physical and mental health, falls ( 9), mortality ( 10, 11), and lower cognitive functioning or dementia ( 12 – 19). However, longitudinal analyses have failed to show an association between many of these risk factors with incidence of age-related hearing loss ( 20 – 22). Divergent patterns of predictors for prevalence versus rates of decline in hearing have been suggested to arise from meth-odological factors. These include insuffi cient statistical power, differences in the rate of onset, and age dependency of hearing loss ( 20). Alternatively, the common practice of dividing ranges of averaged hearing thresholds into conventional categories of hearing loss (eg, no impairment, mild impair-ment, moderate impairment) may obscure true associations between risk factors for change in hearing acuity. We address these issues by employing growth curve techniques to examine hearing trajectories in a larger representative sample of older adults than has previously been available. Other studies inves-tigating longitudinal changes in continuous measures have primarily focused on mapping age and sex trajectories of indi-vidual pure-tone frequencies ( 23 – 27). This study aims to extend the current understanding of age-related hearing loss by additionally investigating sociodemographic and health-related risk factors for change in hearing thresholds. M ethod