Contact tracing and population screening for tuberculosis--who should be assessed?

The aim of the study was to investigate the relative effectiveness of four strategies in detecting and preventing tuberculosis: contact tracing of smear-positive pulmonary disease, of smear-negative pulmonary disease and of non-pulmonary disease, and screening new entrants.

An analysis of patient records and a TB database was carried out for an NHS Trust-based tuberculosis service in a socio-economically deprived area. Subjects were contacts of all patients treated for TB between 1997 and 1999. New entrants were screened in 1999. Outcomes measured were numbers of cases of active tuberculosis detected and numbers of those screened given chemoprophylaxis.

A total of 643 contacts of 227 cases of active TB were seen, and 322 new entrants to the United Kingdom. The highest proportion of contacts requiring full treatment or chemoprophylaxis were contacts of smear-positive index cases (33 out of 263 contacts; 12.5 per cent). Tracing contacts of those with smear-negative pulmonary tuberculosis (12 out of 156; 7.7 per cent) and non-pulmonary disease (14 out of 277; 6.2 per cent) was significantly more effective in identifying individuals requiring intervention (full treatment or chemoprophylaxis) than routine screening of new entrants (10 out of 322; 3.1 per cent).

Screening for TB of new entrants to the United Kingdom is part of the national programme for control and prevention of TB, whereas tracing contacts of those with smear-negative and non-pulmonary disease is not. This study demonstrates that, in our population, the contact-tracing strategy is more effective than new entrant screening. It is not likely that the contacts have caught their disease from the index case, but rather that in high-incidence areas such as ours such tracing selects extended families or communities at particularly high risk.