We found significant variation by socio-demographic characteristics and some variation in state-level estimates. Correlates of visiting an eye care provider included: being older (e.g. 70
years of age or older), being female, having higher educational attainment, having general health and eye care insurance, being diagnosed with diabetes or high blood pressure, reporting an eye condition such as cataract, glaucoma, or age-related macular degeneration, and having near vision difficulties. To our knowledge, no previous studies have used multistate-level data to test the association between sociodemographic characteristics and report of eye care provider visit. Our findings are consistent with studies from the Los Angeles Latino Eye Study, which reported that select socio-demographic characteristics are strongly associated with more frequent eye care provider visits among Latinos, for example: age, educational attainment, general health and eye insurance status, and co-morbidities [17
A visit to an eye care provider within the past two years in this study was associated with a number of variables in our multivariable analysis. Among predisposing variables, older age, female gender, and more education were independently associated with greater use of provider eye care service. These results are consistent with previous research showing that women and older individuals are more likely to use vision health services than their male and younger counterparts [18
]. Previous literature has also shown that education is associated with greater use of eye care [18
]. Nonetheless, these relatively strong independent variables for eye care, such as the social predisposing variable (education) and the enabling variable (insurance status), suggested that the least educated and uninsured were also the least likely to use eye care services. These groups deserve focused attention in any interventions designed to increase eye care utilization rates in these socio-demographic subgroups.
Other factors correlated with greater odds of visiting an eye care provider within the past two years included: having primary ocular disease such as cataract, glaucoma, or age-related macular degeneration, (although the 95% confidence interval for this estimate was large rates (1.06–22.22)). These findings are also similar to results from the Blue Mountains Eye Study in Australia, which included clinical eye examinations [19
]; they reported that blue mountain participants with a history of diabetes, hypertension or with any major eye pathology, including moderate to severe myopia, were significantly more likely to have seen an ophthalmologist in the past 2
years. We found that general health and eye insurance were important enabling variables, therefore, we conducted a stepwise regression analyses to identify indicators of eye care for the subgroup of participants with general health and vision insurance. Significant indicators of eye care in the past 24
0.05) were: (1) Having a larger number of chronic conditions, (2) Having near vision difficulties, (3) Having a higher level of education, (4) Being of female gender, and (5) Being of older age.
We found some variation in state-level estimates of eye care provider visits. Among respondents attending an eye care provider visit within the past two years, adults from Connecticut and New York had the highest estimates for visiting an eye care provider, while respondents from Missouri and New Mexico had the lowest. Studies suggest that state variation in health care visits is driven by underlying economic and demographic factors, such as the employment makeup in the state (e.g., firm size, industry and occupation, and the degree of unionization), eligibility requirements for public programs such as Medicaid, and the demographic/socioeconomic composition of state residents [20
]. State variation in employer-sponsored coverage appears to be driven, in part, by employee characteristics, such as industry and length of time spent with an employer, and local labor market characteristics, such as state-level unionization [23
]. Given that general health and eye care insurance were associated with report of recent eye care visits, all findings consistent were with those reported by Zhang et al. [24
], and variations in economic and labor mixes in the each state could be driving the observed differences.
Strengths and Limitations
This study adds to the literature by being the first to describe the association between eye care provider visits and socio-demographic characteristics using recent population-based data across multiple US states. We were also able to identify the contributions of several important variables (e.g., health and eye insurance status) to these relationships. Although the BRFSS data have been found to provide valid and reliable estimates as compared with the national household surveys [25
], our study has several limitations. First, the cross-sectional design does not allow for causal inferences. Since BRFSS is a telephone based survey, there is the possibility of non-response bias. In addition, the survey used for this study was based on self-reported data and data on the type and quality of health care visits were not available. Studies have shown that self-reported data, particularly of less socially desirable behaviors, are subject to limitations of underreporting and recall bias.