• Epidemiology of tuberculosis in Maryland at the turn of the 21st century: Use of DNA fingerprinting of Mycobacterium tuberculosis isolates to enhance a study of disease transmission

      Cronin, Wendy Anne; Strickland, G. Thomas (2000)
      Background. Population-based studies using DNA fingerprinting of Mtb isolates suggest that recent transmission is more frequent than previously suspected in high TB incidence areas of the US. Objective. To describe circumstances and settings for recent TB transmission in a state with low to moderate disease incidence. Methods. The study population included all Maryland TB cases from 1996 though 1998 (N = 975). US Census and other data were collected for risk-specific TB incidence calculation. Isolates from all Mtb culture positive cases were DNA fingerprinted. Record review and interviews were conducted for all DNA clustered cases to identify epidemiologic linkages and transmission settings. Recent TB cases included those aged ≤5years and DNA clustered cases with known sources and settings. Results. Overall TB incidence was greatest among individuals with black or Asian race, foreign birth, urban risks (poverty, substance abuse, homelessness, HIV infection, incarceration), and those aged >64 years and/or residing in long term care. Cases with low IS6110 copy Mtb strains were significantly less likely than those with high copy strains to have identified epidemiologic linkages (13% vs. 48%; RR: 0.30; 95% CI: 0.20,0.45). Among high copy strains, DNA clustering was geographically localized. US birth, black race, urban risks, and residence in Baltimore City and Prince George's County were associated with recent TB acquisition. Recent disease incidence was similar across all age groups. 41% of recent cases acquired TB in non-household settings and 22% of those aged >5 years had no identified TB risk factors. Sources for these low-risk cases were foreign-born, substance abusers, and/or elderly. Compared to non-transmitters, TB transmitters had significantly longer times from symptom onset to treatment initiation (17 vs. 9 weeks, p = 0.005). Demographics and risks associated with DNA clustered cases with unknown linkages (N = 48) suggested that they had recent, rather than reactivation, disease. Conclusions. Low incidence and an excellent treatment program were not enough to prevent recent TB transmission in Maryland. Findings suggest the need for timely case diagnosis and more aggressive contact investigations, in addition to federally recommended screening and treatment for TB infection in high-risk individuals.