Close contacts of all adults aged ≥15 years with culture-positive pulmonary tuberculosis were prospectively enrolled in a multicenter study from January 2002 to December 2006 at 9 health departments (7 in the United States and 2 in Canada) in the Tuberculosis Epidemiologic Studies Consortium. Close contacts were defined as persons who had shared air space with an individual with pulmonary tuberculosis in the household or other indoor setting for >15 hours per week or >180 hours total during an infectious period, defined as the interval from 3 months before collection of the first culture-positive sputum specimen or the date of onset of cough (whichever was longer) through 2 weeks after the initiation of appropriate antituberculosis treatment.
Contacts were screened as soon as possible after they were identified through interview of patients with tuberculosis and again 10–12 weeks after last exposure to the patient. Screening consisted of a standardized interview and tuberculin skin test (TST), with a positive TST result defined as a ≥5-mm induration. Chest radiography was performed for contacts with positive results of TST. While a standard protocol was used for conducting contact investigations, the staff at the study sites did not use a standard protocol for patient management, which included efforts to prevent secondary cases by investigation and treatment of contacts with LTBI. The Centers for Disease Control and Prevention’s (CDC’s) standard surveillance definitions for a reported case of tuberculosis were used for tuberculosis reporting by all study sites [6 ].
Contacts were cross-matched with state and provincial tuberculosis registries at the end of the enrollment period and annually for 4 years thereafter, with the exception of one study site, which cross-matched contacts annually for 2 years (the final match was in February 2011).
The timing of tuberculosis among contacts was calculated by subtracting the tuberculosis diagnosis date for each index patient from the tuberculosis diagnosis date(s) for their contact(s), and tuberculosis rates per interval were based on the number of contacts with tuberculosis diagnosed in a given interval divided by the number of observed contacts who were disease free at the start of that interval. For contacts with exposure to >1 index case, the earliest index case tuberculosis diagnosis date was used.
Tuberculosis events among contacts with disease diagnosed >30 days after the index cases’ diagnosis were considered incident cases, and tuberculosis events diagnosed before or ≤30 days after the index cases’ diagnosis were considered coprevalent cases.
Survival analysis (Proc Lifetest) was performed using the log-rank test to assess the effect of age group, TST size, and preventive therapy on disease-free survival of contacts. Statistically significant differences for other analyses were assessed using χ2 or Fisher exact tests. All analyses were performed using SAS software, version 9.2 (Statistical Analysis Software Institute, Cary, NC).
Approvals for human subjects research were obtained from the CDC and all project sites.
Contacts were screened as soon as possible after they were identified through interview of patients with tuberculosis and again 10–12 weeks after last exposure to the patient. Screening consisted of a standardized interview and tuberculin skin test (TST), with a positive TST result defined as a ≥5-mm induration. Chest radiography was performed for contacts with positive results of TST. While a standard protocol was used for conducting contact investigations, the staff at the study sites did not use a standard protocol for patient management, which included efforts to prevent secondary cases by investigation and treatment of contacts with LTBI. The Centers for Disease Control and Prevention’s (CDC’s) standard surveillance definitions for a reported case of tuberculosis were used for tuberculosis reporting by all study sites [6 ].
Contacts were cross-matched with state and provincial tuberculosis registries at the end of the enrollment period and annually for 4 years thereafter, with the exception of one study site, which cross-matched contacts annually for 2 years (the final match was in February 2011).
The timing of tuberculosis among contacts was calculated by subtracting the tuberculosis diagnosis date for each index patient from the tuberculosis diagnosis date(s) for their contact(s), and tuberculosis rates per interval were based on the number of contacts with tuberculosis diagnosed in a given interval divided by the number of observed contacts who were disease free at the start of that interval. For contacts with exposure to >1 index case, the earliest index case tuberculosis diagnosis date was used.
Tuberculosis events among contacts with disease diagnosed >30 days after the index cases’ diagnosis were considered incident cases, and tuberculosis events diagnosed before or ≤30 days after the index cases’ diagnosis were considered coprevalent cases.
Survival analysis (Proc Lifetest) was performed using the log-rank test to assess the effect of age group, TST size, and preventive therapy on disease-free survival of contacts. Statistically significant differences for other analyses were assessed using χ2 or Fisher exact tests. All analyses were performed using SAS software, version 9.2 (Statistical Analysis Software Institute, Cary, NC).
Approvals for human subjects research were obtained from the CDC and all project sites.
