Contact investigation is a key component of tuberculosis (TB) control in developed, but not developing, countries. We aimed to measure the prevalence of TB among household contacts of sputum-smear-positive TB cases in The Gambia and to assess the sensitivity of an enzyme-linked immunospot (ELISPOT) assay in this regard. Household contacts of adult smear-positive TB patients were assessed by questionnaire, purified protein derivative (PPD) skin test, ELISPOT assay, physical examination, chest X-ray and sputum/gastric aspirate. Thirty-three TB cases were identified from 2174 of 2381 contacts of 317 adult smear-positive pulmonary TB patients, giving a prevalence of 1518/100000. The cases identified tended to have milder disease than those passively detected. The sensitivity of ESAT-6/CFP-10 ELISPOT test as a screening test for TB disease was estimated as 71%. Fifty-six per cent of contacts with a PPD skin test result >or=10mm induration had detectable responses to ESAT-6/CFP-10 by ELISPOT; 11% with a negative PPD skin test (<10mm) had a positive ESAT-6/CFP-10 response. Active screening for TB among contacts of TB patients may have a role in TB control in The Gambia. These individuals are a high-risk group, and the disease identified is less advanced than that found through passive case detection. An ELISPOT assay was relatively insensitive as a screening test for TB.