238463 A Novel Rural Classification System at the Small-Area (ZIP code) Level: Adapting the Index of Relative Rurality (IRR) to Estimate Degree of Rurality in Health Services Research

Monday, October 31, 2011: 1:20 PM

Sanae Inagami, MD MPH , General Internal Medicine, University of Pittsburgh, Pittsburgh, PA
Shasha Gao, PhD , Center for Health Equity Research Promotion-Bioinformatics, VA Pittsburgh Healthcare System, Pittsburgh, PA
Roslyn A. Stone, PhD , Department of Biostatistics, University of Pittsburgh Graduate School, Pittsburgh, PA
Martine Shendge, MS , Center for Health Equity Research Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA
Hassan Karimi, PhD , School of Information Science, University of Pittsburgh, Pittsburgh, PA
Janice C. Probst, PhD , University of South Carolina, South Carolina Rural Health Research Center, Columbia, SC
     
Background: We (1) describe the development and demonstrate the application of a novel rural classification system at the small-area level, the ZIP code-level Index of Relative Rurality (IRR), and (2) compare the differences in how geographic areas and populations are defined as rural when using the IRR versus the standard rural classification system, Rural Urban Commuting Area codes. Methods: Data from the 2000 Census and Economic Research Service (ERS) were linked to Veteran data extracted from the Veteran's Administration Pittsburgh Healthcare System. The ERS data provided the ZIP code-level RUCA codes associated with each Veteran address. A ZIP code-level IRR classification system was developed using 4 Census-derived measures: residential area-level population, population density, degree of urbanization, and distance to Metropolitan Statistical Area (MSA) from Veteran address. Results: The IRR for the total sample of 7,608 Veterans with diabetes ranged from 0.01 to 0.77 with a mean of 0.24. Kernel density curves for IRR by RUCA 5-tier category demonstrated inconsistencies in how rurality was defined between the two classification measures. Using RUCA codes, fewer than 5% (137/7570) of Veterans with diabetes lived in rural areas. In contrast, when using a very conservative cut-off for rurality (derived from Kernel density plots), the IRR demonstrated that 21% (1597/7570) of Veterans with diabetes lived in rural areas. Conclusions: The ZIP code-level IRR and RUCA classification system demonstrate inconsistencies in how rurality is defined. The IRR is more flexible methodologically and easier to interpret than using RUCA codes.

Learning Areas:
Epidemiology
Public health or related research

Learning Objectives:
1) To define the components used to construct a novel rural classification system, the Index of Relative Rurality (IRR), at the ZIP code level; 2) To discuss the methodological advantages of using the IRR; 3) To demonstrate the utility in applying the modified ZIP code-level IRR to existing data; 4) To compare how rural areas and populations differ when using the IRR to the standard traditional rural classification system, Rural Urban Commuting Area (RUCA) codes.

Keywords: Epidemiology, Rural Health

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am a research investigator and faculty member at University of Pittsburgh.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.

See more of: Health Disparities
See more of: Epidemiology