County characteristics and racial and ethnic disparities in the use of preventive services Maureen R. Benjamins, M.A., a, * James B. Kirby, Ph.D., b and Stephanie A. Bond Huie, Ph.D. c a Population Research Center, The University of Texas at Austin, Austin, TX 78712, USA b Agency for Healthcare Research and Quality, Washington, DC 20850, USA c Southwest Education Development Laboratory, Austin, TX, USA Available online 2 April 2004 Abstract Background. Studies examining predictors of preventive service utilization generally focus on individual characteristics and ignore the role of contextual variables. To help address this gap in the literature, the present study investigates whether county-level characteristics, such as racial and ethnic composition, are associated with the use of preventive services. Methods. Data from the Medical Expenditure Panel Survey and the Area Resource Files (1996 – 1998) are used to identify the individual- and county-level predictors of five types of preventive services (n = 49,063). Results. County racial or ethnic composition is associated with the utilization of certain preventive services, net of individual-level characteristics. Specifically, individuals in high percent Hispanic counties are more likely to report cholesterol screenings, while those in counties with more blacks are more likely to have regular mammograms. Moreover, county racial or ethnic composition modifies the relationship between individual race or ethnicity and preventive use. In particular, Hispanic individuals who reside in high percent black counties report higher levels of utilization for most preventive services compared to Hispanics living in other counties. Conclusions. Physical and social environments are key determinants of health behaviors and outcomes. Future studies should take into account the racial or ethnic composition of an area and how this interacts with individual race or ethnicity when investigating predictors of preventive care use. D 2004 The Institute For Cancer Prevention and Elsevier Inc. All rights reserved. Keywords: Preventive health care; Utilization; Race; Ethnicity; Contextual effects Introduction Researchers in a wide variety of disciplines have studied the use of preventive services extensitwb=.3wvely. The sheer number of studies reflects both a widespread belief in the effectiveness of these services and numerous attempts to understand variation in levels of use. Studies have examined a variety of individual-level social, demo- graphic, and health-related predictors to better understand suboptimal levels of use [1–7]. The most important individual-level factors affecting the use of preventive services include age [7–10], race [2,4,11 – 14], ethnicity [15], marital status [10], income [2,4,8,10 – 12], education [8,11,13], having a usual source of care [16,17], and insurance coverage [8,9,15,18,19]. In addition, studies have examined the role of health status [4], region of residence and urban or rural status [2,4], length of time in the United States [10], acculturation [20], and health beliefs [9]. Despite this wealth of information, a gap in the literature remains. Namely, little is known about how characteristics of the areas in which individuals reside may influence their use of preventive services. In fact, although the importance of using both individual- and community-level data is acknowl- edged in the literature [21], we found only two previous studies that examined the relationship between area charac- teristics and preventive service use [1,22]. The most recent of these studies examined how proximity to mammography facilities influenced mammography use among female Medi- care beneficiaries in Kansas [22]. This study found that population density did not have a significant effect on mammography use, but that women in counties with higher median incomes and a greater percentage of residents with 0091-7435/$ - see front matter D 2004 The Institute For Cancer Prevention and Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2004.02.039 * Corresponding author. Population Research Center, University of Texas at Austin, 1 University Station G1800, Austin, TX 78712. E-mail address: reindl@prc.utexas.edu (M.R. Benjamins). www.elsevier.com/locate/ypmed Preventive Medicine 39 (2004) 704 – 712