Background: The tuberculin skin test is used for tracing of tuberculosis transmission and identifying individuals in need of prophylactic treatment.
Methods: Using a case-control study design, we recruited 220 smear-positive tuberculosis cases and 223 randomly selected healthy community controls in Bissau, Guinea-Bissau, during 1999-2000. Tuberculin skin tests were performed on family members of cases and controls (n = 1059 and n = 921, respectively). Induration of 10 mm or greater was considered positive. Risk factors were calculated for children (<15 years) and adults separately in multivariate logistic regression analysis.
Results: The prevalence of positive tuberculin skin test was 41% in case-contacts compared with 22% in control-contacts, resulting in a prevalence ratio of 1.48 (95% confidence interval = 1.37-1.60). Positive skin tests among case-contacts increased with age for children, as well as with proximity to a case during the night, for both children and adults. A Bacille Calmette Guerin scar increased the likelihood of having a positive tuberculin skin test for adults in case households, but not in other categories of contacts. Among control-contacts the prevalence of positive skin test was associated with older age in children, history of tuberculosis in the family, and a positive tuberculin skin test of the control person.
Conclusions: Risk factors for a positive tuberculin skin test among case- and control-contacts are closely related to tuberculosis exposure. Having a BCG scar did not increase the risk of positive skin test in unexposed individuals. Tuberculin skin testing remains a useful tool for diagnosing tuberculosis infection.