The setting is Homeless and other fringe groups are a priority in the global strategies of TB prevention and control in big cities, as a consequence of their generally poor adherence to treatment and concurrent multiple social and health problems. The objective is to evaluate a social care and health follow-up programme targeting homeless TB patients in Ciutat Vella District, Barcelona, which covered 210 patients from 1987 to 1992. During directly observed treatment, primary health care and, if necessary, accommodation was provided. The design of the study is the differential TB incidence rate between Ciutat Vella and the other districts of Barcelona, the percentage of successfully completed treatments and the days of hospitalization saved by the programme were measured.

There was a significant decrease in the TB incidence rate among homeless patients in Ciutat Vella (from 32.4 per 105 inhabitants in 1987, to 19.8 per 105 in 1992, P=0.03), compared to an unchanged rate elsewhere (1.6 per 105 inhabitants in 1987, compared to 1.7 per 105 in 1992, P=0.34). A smaller than expected proportion, 19.6%, of patients failed to complete their treatment, and a decrease in the mean period of hospitalization for TB in the district hospital was recorded, falling from a mean 27.1 days in 1986 to a mean 15.7 days in 1992. The programme appears to be both effective and efficient, as it has enabled a large number of homeless patients to complete their treatment successfully, at the same time saving twice the amount of funds invested.