Abstract
SETTING: In the United States, tuberculosis (TB) control is increasingly focusing on the identification of persons with latent tuberculous infection (LTBI). OBJECTIVE: To characterize the local epidemiology of LTBI in Connecticut, USA. METHODS: We used spatial analyses 1) to identify census tract-level clusters of reported LTBI and TB disease in Connecticut, 2) to compare persons and populations in clusters with those not in clusters, and 3) to compare persons with LTBI to those with TB disease. RESULTS: Significant census tract-level spatial clusters of LTBI and TB disease were identified. Compared with persons with LTBI in non-clustered census tracts, those in clustered census tracts were more likely to be foreign-born and less likely to be of white non-Hispanic ethnicity. Populations in census tract clusters of high LTBI prevalence had greater crowding, persons living in poverty, and persons lacking health care insurance than populations not in clustered census tracts. Persons with LTBI were less likely than those with TB disease to be of Asian ethnicity, and persons with LTBI were more likely than those with TB disease to reside in a clustered census tract. CONCLUSIONS: Characterizing fine-scale populations at risk for LTBI supports effective and culturally accessible screening and treatment programs.