Health status disparities and the role of poverty in health have received global attention in recent years. Health authorities need to know how their current or future policies may cause inequality so that they can make corrections. After investigators measure inequality in a community, the next step is its interpretation. This can be done through decomposition of inequality into its contributing factors.
In the decomposition method ?rst implemented by Blinder and Oaxaca in 1973 to assess labor market discrimination, the outcome gap between 2 groups is divided into 2 components: one component due to differences in magnitudes of the determinants of the outcome between groups (the explained or endowment component) and a second component attributable to group differences in the effects of these determinant(s) (the unexplained or coef?cient component). In recent years, the Oaxaca method has been increasingly implemented in health research. To our knowledge, this is the ?rst article using this method in a study on visual impairment and eye disease.
Studies on disparity in visual impairment are limited to some reports about gender inequality in eye care services and blindness in poor communities. In the Eastern Mediterranean Region, the only study about gender inequality was conducted in Oman, where Khandekar and Mohammed investigated gender disparities in blindness and eye diseases that lead to blindness. In Iran, the prevalence of visual impairment by presenting vision was reported to be 2.5% in Tehran, 33.6% of which was due to refractive errors. Overall, published reports have described visual impairment in Iran to some extent, but similar to other countries, the issue of inequality has been left out. The Shahroud Eye Cohort Study (ShECS) started in 2008, and the ?rst phase was completed in 2009. In this report, we try to explore inequality in visual impairment using ShECS data and the Blinder-Oaxaca decomposition method.
The studied sample comprised participants of the ?rst phase of ShECS, where 6,311 residents aged 40-64 years and living in Shahroud were randomly selected and 5,190 individuals (82.2%) were studied.
During the interview, we inquired about the possession or lack of 11 household items in the form of a binary variable. We also recorded the years of education of the participant and the head of the household as continuous variables. The outcome measure in this study was visual impairment as determined by the presenting (habitual) vision of the participant in daily life situations, whether corrected or uncorrected (20). Presenting vision was measured by trained staff and optometrists using standard logarithm of the minimum angle of resolution (referred to as ''LogMAR'') charts. Visual impairment was de?ned as a presenting vision greater than 0.3 LogMAR (worse than 20/40) in the better eye.
The studied sample was grouped according to their economic status. For this purpose, a principal component analysis was done ?rst on home assets, as described by O'Donnell et al. (3). Then, we constructed the assets index variable based on the weighting of the ?rst component and divided it into ...