Exploring the Spatial Determinants of Late HIV Diagnosis in Texas
ORIGINAL RESEARCH — Volume 17 — August 27, 2020
PEER REVIEWED
The figure consists of a map of the study area in Texas. The study area includes the 5 Texas cities with the largest population and the highest HIV morbidity and the areas in between these cities. The map shows a spatially smoothed regional percentage of late HIV diagnoses that ranges from 1% to 43%. The percentage of late HIV diagnosis is higher in less densely populated areas, such as rural areas and suburban areas near urban centers.
Figure 1.
Spatially smoothed regional percentage of late HIV diagnoses in the study area in Texas (5 largest cities, by population and by HIV morbidity: Houston, Dallas, San Antonio, Austin, and Fort Worth).
The figure consists of 4 maps of the Texas study area. Three maps show estimates and their critical t values for the variables poverty, unemployment, and distance to the nearest HIV testing site based on a multiscale geographically weighted regression (MGWR). The map on poverty shows negative significant estimates in regions northwest and southeast within the study area. The map on unemployment shows positive estimates for regions northeast of Dallas and within Houston. Regions at the center of the study area and north of Austin had a negative association. Local significant associations with distance were positive in regions west of Fort Worth, southeast of Dallas, and in the southwest of the study area, including the cities of San Antonio and Austin. In a region north of Houston the association was negative. A fourth map shows the local R square for the MGWR model with a range of explanatory power from 27% to 91%. The model performed best in regions with relatively high rates of late HIV diagnoses in rural and suburban peripheral regions of the largest urban centers.
Figure 2.
Spatial distribution of parameter estimates in study area (5 largest Texas cities, by population and by HIV morbidity: Houston, Dallas, San Antonio, Austin, and Fort Worth) of late HIV diagnosis at the regional level in Texas, 2011–2015. Maps show spatial distribution of parameter estimates for the percentage of people in poverty, percentage of people unemployed, distance to the nearest HIV testing site, and local R2 in a multiscale geographically weighted regression.
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.