Geospatial analysis to identify census tracts with TB clusters and associated socio-economic factors in a local health department
Abstract
Background Santa Clara County is a large, geographically, ethnically, and socioeconomically diverse county in Northern California with a Tuberculosis (TB) case rate of 8.8 per 100,000, almost three times the national average,... [ view full abstract ]
Background
Santa Clara County is a large, geographically, ethnically, and socioeconomically diverse county in Northern California with a Tuberculosis (TB) case rate of 8.8 per 100,000, almost three times the national average, in 2014. Majority of TB cases were foreign-born Asians. We conducted an ecological study to identify geographic areas with high incidence of TB and associated characteristics.
Methods
Addresses for TB cases during 2010–2014 were geocoded to one of 372 census tracts (CT). CTs with high and low incidence of TB were identified and compared using Getis-Ord GI* geospatial analysis. Univariate and multivariate logistic regressions were applied to examine the relationship between TB clustering and various socioeconomic factors including poverty, nativity, race/ethnicity, household size, and population density.
Results
A total of 893 TB cases occurred during 2010–2014, of which 867 (97%) were successfully geocoded. Nearly half (418/867) of TB cases were spatially clustered in just one-fourth (97/372) of the CTs. Compared to CTs with low incidence of TB, CTs with TB clusters had significantly higher proportions of foreign-born (47.6% vs. 36.6%), Asians (44.8% vs. 35.3%), households with income <200% FPL (31.9% vs. 16.8%), larger household size (3.7 vs. 2.8), and higher population density (8,464 vs. 2,745 people per square mile). After controlling for other factors, CTs with over 40% foreign-born population [Adjusted Odds Ratio (AOR) = 4.9, 95% confidence interval (CI): 1.9 – 12.9], having ≥30% households with income <200% FPL [AOR=11.4, 95% CI: 4.3 – 30.2], and household size >3 [AOR=5.2, 95% CI: 2.6 – 10.3] had significantly higher odds of having TB clusters. Proportion of Asians and population density were not associated factors.
Conclusions
Geospatial analysis helped identify CTs with TB clusters and associated population-level factors. This analysis will guide further investigation of CTs with TB clusters and assist targeted TB screening and treatment efforts.
Authors
-
Doug Schenk
(Santa Clara County Public Health Department, California)
-
Wen Lin
(Santa Clara County Public Health Department, California)
-
Anandi Sujeer
(Santa Clara County Public Health Department, California)
Topic Area
II. Environmental Health 2.1 Disease mapping 2.2 Assessment of the impact of environmental
Session
PS-3 » POSTER SESSION 3 (12:15 - Sunday, 3rd April, TBA)
Paper
2016_International_Planning.doc
Presentation Files
The presenter has not uploaded any presentation files.