Purchase this article with an account.
K. M. Daum, A. L. Irby, M. Sanspree, M. R. O’Neal, C. M. Brezausek, C. Allison, D. Pevsner, L. Moses; Modeling Risk Factors for Intraocular Pressure in Rural Alabama. Invest. Ophthalmol. Vis. Sci. 2008;49(13):5052. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Residents of the relatively isolated and economically-disadvantaged population of the rural Alabama Black Belt are afflicted with high levels of diabetes, hypertension and other conditions and also have poor access to health care. The aim of this study is to develop a model to assess the relationship of demographic variables, health history, socioeconomic status (SES) and a variety of systemic and ocular health characteristics to intraocular pressure (IOP) in an adult population.
The Rural Alabama Diabetes and Glaucoma Initiative (RADGI) performed comprehensive vision and health evaluations on adult volunteers from rural Alabama. The vision and health assessment included 58 variables related to the demographic, health history, SES and physical (systemic and ocular) status of the patients. We used multiple imputation and linear regression models to assess risk factors for IOP. The RADGI process was approved by the UAB IRB. All participants completed informed consent prior to participating in the evaluations.
The results included 2,237 participants (mean 49.9 yrs, Std dev, 15.2, range 19 to 84 yrs). The group was heavily female (33.2 males/100 females) and African-American (80%). Univariate analysis of demographic, SES, history and physical (ocular & systemic) demonstrated many variables significantly associated with IOP. A linear regression model (18 variables) was constructed to predict IOP in the right eye (OD) of participants using a multiple imputation technique from the data as categorized into the 5 areas noted. The model (n=1193) demonstrated 18 variables significantly related to IOP (overall model r-square = 0.0850). The partial r-square and proportions for the variables grouped by categories were: physical, systemic (0.0524; 61.7%), SES (0.0146; 17.2%), health history (0.0135; 15.9%), demographics (0.0041; 4.8%) and, physical, ocular (0.0003; 0.4%).
Many factors were related to IOP. Optimal control of IOP may require understanding factors beyond those traditionally considered in the clinic. Physical systemic factors provided the greatest contribution to the model while SES contributed relatively greater than health history, demographic or physical ocular groups of variables.
This PDF is available to Subscribers Only