Methodology
For this study, the Dallas Fort Worth Area is defined as Collin, Dallas, Denton, and Tarrant Counties. The Houston-Galveston Area is defined as the 13 counties represented in the Houston Galveston Area Council: Austin, Brazoria, Chambers, Colorado, Fort Bend, Galveston, Harris, Liberty, Matagorda, Montgomery, Walker, Waller, and Wharton (HGAC, 2008). The Texas Secretary of State (TSOS) defines colonias counties as counties containing a colonia within 150 miles of the Texas-Mexico border. The 29 counties that meet this definition are: Aransas, Brewster, Brooks, Cameron, Culberson, Dimmit, Duval, Edwards, El Paso, Frio, Hidalgo, Hudspeth, Jeff Davis, Jim Hogg, Kinney, Kleberg, La Salle, Maverick, Nueces, Pecos, Presidio, Reeves, Starr, Terrell, Uvalde, Val Verde, Webb, Willacy, Zapata. The counties containing a majority of the colonias are El Paso, Hidalgo, Cameron, and Willacy (Wallisch, 1998). In total, the study encompasses 46 counties. Refer to Figure 1 for counties.
Data on the independent variables, income and race/ethnicity in these counties, is obtained from the 2000 U.S. Census. The percent of uninsured population, unemployed persons and primary healthcare providers per 100,000, are from the Texas Health Ranking Data. The 1980-2009 data on the dependent variables, pediatric HIV mortality rates and HIV average age at diagnosis in women, are provided by the Texas Department of State Health Services. CDC defines pediatric HIV as HIV in children under 13 years of age (CDC, 2010d). The median age and median age of females in each county are retrieved from the US Census.
The female HIV rate is calculated by dividing the total female cases by the total female population in the county, and multiplying the quotient by 100,000. Pediatric HIV rate is calculated similarly. The population under 13 years of age is also from the 2000 U.S. Census data.
Maps are created in ArcGIS and statistical analysis is completed with SPSS 17.0. Maps have differing scales corresponding to their respective data. The pediatric HIV data cannot be mapped using the same scale as that for HIV due to the extremely small number of cases in Texas. There are also a much smaller number of female HIV cases than total cases in the population at large, since men still comprise the majority of HIV cases in Texas.
Variables
The variables in this study include HIV diagnoses per 100,000 children under 13, female HIV diagnosis per 100,000 women, and age at HIV diagnosis in females as dependent variables. They will be compared to several independent variables including percent of uninsured population, unemployment rate, and primary care provider rate per 100,000, used as measures of access to health care. In addition, the study will examine race/ethnicity and median income as other possible risk markers.