Changing sociodemographic and clinical characteristics of tuberculosis among HIV-infected patients, New York City, 1992-2005

Background. Although highly active antiretroviral therapy (HAART) has decreased human immunodeficiency virus (HIV)-related morbidity, tuberculosis remains an important disease among HIV-infected individuals. Methods. By use of surveillance data, sociodemographic and clinical changes among HIV-infected and HIVuninfected tuberculosis patients in New York City were evaluated using the Cochran-Armitage trend test and multivariate logistic regression across 3 periods: 1992-1995 (pre-HAART), 1996-2000 (early HAART), and 20012005 (late HAART). Results. Among tuberculosis patients with known HIV status, 4345 (60%) of 7224 were HIV-infected in preHAART, 1943 (33%) of 5933 in early HAART, and 851 (22%) of 3815 in late HAART (P< .001 for trend). During the study period, the age of HIV-infected tuberculosis patients increased, and greater proportions were female, non-Hispanic black, Asian, and foreign born; the proportion that was non-Hispanic white decreased. The proportion that was culture-negative for Mycobacterium tuberculosis increased (from 7% pre-HAART to 21% late HAART; P< .001 for trend; early HAART vs pre-HAART adjusted odds ratio [aOR], 1.68; 95% confidence interval [CI], 1.38-2.04), and the proportion with extrapulmonary disease also increased (from 32% to 46%; P< .001 for trend). The proportion with multidrug-resistant tuberculosis decreased (from 16% to 4%; P<.001 for trend), especially from pre-HAART to early HAART (aOR, 0.31; 95% CI, 0.25-0.40). The proportion who died before tuberculosis treatment decreased (from 12% to 7%), and the proportion who died during tuberculosis treatment also decreased (from 29% to 11%) (both, P<.001 for trend). Over time, HIV-infected tuberculosis patients had AIDS longer before the diagnosis of tuberculosis (P<.001 for trend). Similar trends for culture, site of disease, and drug resistance were seen for HIV-uninfected tuberculosis patients. Conclusions. The sociodemographic and clinical characteristics changed substantially among HIV-infected tuberculosis patients in New York City. Awareness of these changes may speed diagnosis of tuberculosis. Future studies should evaluate HAART's effect on tuberculosis presentation among HIV-infected patients.