|
Abstract:
|
There have been attempts using personal characteristics to explain barriers/delays in accessing healthcare, and using ecological measures as near-proxies for health access problems. We combined these efforts to develop indices of health access deprivation using 2001-2003 NHIS data, and constructing indices using PCA. We merged indices with census tract level data, and used multi-level logistic modeling to pair wise substitute Census for NHIS variables. We found 10 predictors of health deprivation out of 23 potential from literature. Deprivation is more likely, if one is female, do not own a home, live in a single adult household, have a functional limitation, smoke, have no health insurance or is less than 65. We mapped indices and overlaid them with provider availability statistics. This tool could assist local community assessments, health planning, and identifying populations at risk.
|
