Development of a rural-urban classification system for public health research that accommodates structural differences between states.

Rural environments in the United States present challenges to wellness, but there is a lack of tools with which to categorize rurality at the subcounty level. The most common tool, the Food and Drug Administration's 2010 Rural-Urban Commuting Area (RUCA) codes, uses data that are over a decade old and cannot accommodate regional differences in rurality. The purpose of this study was to develop a census-tract classification system of rurality and demonstrate its use in describing HIV outcomes. We transformed census-tract measures (population density, natural resource workforce, walkability index, household type, and air quality) into local scales of rurality using factor analysis. We surveyed public health practitioners to determine cutpoints and compared the resulting categorization with RUCA codes. We characterized the incidence of HIV in Washington State according to rurality category. Our classification system categorized 25% of census tracts as rural, 19% as periurban, and 56% as urban. Our survey yielded cutoffs that were more conservative in categorizing urban areas than RUCA codes. The rate of HIV diagnosis was substantially higher in urban areas. Our rural-urban classification system offers an alternative to RUCA codes that is more responsive to regional differences.