Purpose
The excess medical care associated with vision impairment and blindness in the United States for individuals aged 40 and over was shown to be $2157 for individuals self-reported as blind and $1037 self reported as visually impaired in 2007. This study’s objective is to use additional data on self-reported vision status and medical care expenditures to compare the original estimates from 1996-2002 with estimates from 2003-2009 to determine whether the impact of blindness and vision impairment at the individual level has changed.
Methods
Medical Expenditure Panel Survey (MEPS) data from 1996-2009 were used. A logistic regression was used to assess the relationship between not having self-reported normal vision and having any medical care expenditures. A separate generalized linear model with a log link-function and an assumption of a gamma family error term distribution was used to assess the relationship between not having self-reported normal vision and the level of positive medical care expenditures.. Survey regression techniques were used to make the estimates as applicable as possible to the population of the United States. Demographic variables including income, health status, age, and insurance, were included in the regressions. Total expenditures were compared as well as 26 other categories of expenditures and total out-of-pocket expenditures that lead to a higher burden on the blind and visually impaired individuals.
Results
Of the 112 differences examined, only 9 were statistically significant. When comparing the chance of any expenditure, significant changes implied a smaller odds ratio in the more recent time period. When comparing levels of expenditures among those with non-zero expenditures, the significant changes implied less spending for blind or visually impaired individuals relative to those with self-reported normal vision in the more recent time period in all cases but one.
Conclusions
Blindness and vision impairment continued to be associated with a total expenditures and various categories of expenditures overall. However, to the degree the differences are not stable, evidence suggests the impact has decreased between 1996-2002 and 2003-2009. This may be more associated with poorer health indicators for the general population rather than any relative advantage gained by visually impaired individuals.
Keywords: 460 clinical (human) or epidemiologic studies: health care delivery/economics/manpower