Association Between Dual Sensory Impairment, Hospitalization, and Burden of Disease

To the Editor: Hearing impairment (HI) and vision impairment (VI) in older adults are independently associated with negative health outcomes, including physical disability,1, 2 cognitive decline,3, 4 hospitalization,5, 6 and mortality.7, 8 Although sensory impairments (SIs) are highly prevalent in older adults, they have largely been perceived as an inevitable, and hence relatively inconsequential, aspect of aging, but two ongoing Institute of Medicine (IOM) consensus studies are now focusing on SIs as potential targets for improving public health and promoting healthy aging. The prevalence of combined HI and VI, termed dual sensory impairment (DSI), was recently shown to be higher than previously assumed, reaching a prevalence of greater than 10% in adults aged 80 and older.9 Although DSI has been associated with greater mortality10 than HI or VI alone, the relationship between DSI and hospitalization and burden of disease is unknown. Participants were selected from the combined 2005–06 and 2009–10 cycles of the National Health and Nutrition Examination Survey (NHANES), a nationally representative study assessing health in the noninstitutionalized civilian population. Objective air-conduction pure-tone audiometric data were collected in these cycles, and hearing loss was defined as the pure tone average (PTA) of hearing thresholds of 0.5 to 4 kHz in the better-hearing ear (normal, <25 dB hearing level (HL); mild, 25–39 dB HL, moderate or worse, ≥40 dB HL). VI was defined as self-reported vision difficulty (yes/no) while wearing corrective lenses (objective vision data were not gathered in both the 2005–06 and 2009–10 study cycles). Participants, who were aged 70 and older with available hearing and vision data (N = 1,669), were classified as having no impairment, HI only, VI only, or DSI. Consistent with a previous NHANES study,5 hospitalization was self-reported and categorized as any hospitalization (yes/no) and number of hospitalizations (0, 1, >1) during the past 12 months. Burden of disease was self-reported and grouped as 0 to 10 days or more than 10 days of the previous 30 days in the domains of self-rated poor physical health, poor mental health, and inactivity because of health. The association between SI and any hospitalization and burden of disease was investigated using multivariate logistic regression, and the association between SI and number of hospitalizations was investigated using ordinal logistic regression, with an interaction term between HI and VI included to investigate the differential effect of DSI versus HI or VI alone. NHANES-provided survey weights were used to account for the complex survey design. Data were analyzed using Stata version 12 (Stata Corp., College Station, TX). Statistical significance was set at P < .05. DSI was significantly associated with more days of hospitalization (odds ratio (OR) = 1.98, 95% confidence interval (CI) = 1.09–3.59), poorer physical health (OR = 2.41, 95% CI = 1.14–5.10), and poorer mental health (OR = 4.80, 95% CI = 2.16–10.65) than no impairment (Table 1). The estimated odds of all outcomes in participants with DSI with moderate or worse HI were greater than the odds in participants with moderate or worse HI alone, but these differences were not significant. In this nationally representative study of 1,669 older adults, DSI was found to be significantly and independently associated with more days of hospitalization, poorer physical health, and poorer mental health than no SI. Although stronger associations were generally found between DSI and all outcomes than between HI or VI alone and all outcomes, these differences were not statistically significant. Nevertheless, the number of events in individuals with DSI was small, and hence this study may be underpowered. In addition, subjective vision data may not correlate well with objective vision measures. To the knowledge of the authors, this is the first study to investigate the association between DSI and hospitalization and burden of disease and adds to a growing body of literature demonstrating the negative health effects of SI. Whether one type of SI is the primary determinant of these associations or whether individuals with dual sensory impairment are at compounded risk of adverse outcomes is unclear. Longitudinal studies with objectively measured SI are needed to prospectively examine this relationship. Addressing SIs could have substantial implications for public health given that these impairments are highly prevalent, undertreated, and amenable to treatment with established, low- to no-risk rehabilitative interventions. If individuals with DSI are at greatest risk of adverse health outcomes, then strategies targeting this at-risk cohort may be particularly effective. This research was supported by National Institutes of Health Grant K23DC011279 and the Eleanor Schwartz Charitable Foundation.