Burden of tuberculosis in Kampala, Uganda

To determine the prevalence and incidence of tuberculosis in one of Uganda's poor peri-urban areas.

Multi-stage sampling was used to select a sample of households whose members were evaluated for presence of signs and/or symptoms of active tuberculosis; history of tuberculosis treatment; and relevant demographic, socioeconomic, and household environment characteristics. Patients with suspected tuberculosis underwent standardized evaluation for active disease.

A sample of 263 households with 1142 individuals was evaluated. Nineteen people were classified as having had tuberculosis during the one-year reference period (May 2001-April 2002): nine (47%) cases already had been diagnosed through the health care system, while 10 cases (53%) were diagnosed through the survey. The prevalences for all forms of tuberculosis and for sputum smear-positive tuberculosis were 14.0 (95% confidence interval (CI) 7.8-20.3) and 4.4 (CI = 0.83-7.89) per thousand, respectively. The incidences for all forms of tuberculosis and for sputum smear-positive tuberculosis were 9.2 (CI = 3.97-14.4) and 3.7 (CI = 0.39-6.95) per thousand per year, respectively.

The rate of tuberculosis in this peri-urban community was exceptionally high and may be underestimated by current surveillance systems. The need for interventions aimed at reducing tuberculosis transmission in this, and other similar communities with high case rates, is urgent.