OBJECTIVE: Descriptive study of molecular epidemiologic patterns of tuberculosis cases among ethnic minorities in Houston, Texas. DESIGN: Population-based, prospective, active surveillance, and molecular epidemiology study. PATIENTS: Tuberculosis cases reported to the City of Houston Tuberculosis Control Office between October 1995 and September 1998. RESULTS: During the study period, 1,139 culture-positive patients were enrolled for whom isolates of their culture specimen were available. Of these, 910 were part of an ethnic minority. Molecular characterization identified 689 of 1,139 isolates to be clonally related. Factors significantly associated with tuberculosis strain clustering in a multivariable logistic regression analysis were: birth in the United States, a history of homelessness, infection with the human immunodeficiency virus (HIV), pulmonary disease, infection with a tuberculosis strain from principal genetic group 1 or 3, living in a residence with five or more persons present, and use of public transportation more than once weekly. Asian ethnicity and increasing age were associated with decreased odds of clustering. CONCLUSIONS: Ethnicity was not a significant covariate for strain clustering after adjustments for factors related to socio-economic status.
OBJECTIVE: Descriptive study of molecular epidemiologic patterns of tuberculosis cases among ethnic minorities in Houston, Texas. DESIGN: Population-based, prospective, active surveillance, and molecular epidemiology study. PATIENTS: Tuberculosis cases reported to the City of Houston Tuberculosis Control Office between October 1995 and September 1998. RESULTS: During the study period, 1,139 culture-positive patients were enrolled for whom isolates of their culture specimen were available. Of these, 910 were part of an ethnic minority. Molecular characterization identified 689 of 1,139 isolates to be clonally related. Factors significantly associated with tuberculosis strain clustering in a multivariable logistic regression analysis were: birth in the United States, a history of homelessness, infection with the human immunodeficiency virus (HIV), pulmonary disease, infection with a tuberculosis strain from principal genetic group 1 or 3, living in a residence with five or more persons present, and use of public transportation more than once weekly. Asian ethnicity and increasing age were associated with decreased odds of clustering. CONCLUSIONS: Ethnicity was not a significant covariate for strain clustering after adjustments for factors related to socio-economic status.