Observation alone isn't the key

World wide, more adult deaths are attributed to Mycobacterium tuberculosis than to any other infectious agent, and without improvements in control 30 million people are expected to die from tuberculosis in 1990-2000.1 In sub-Saharan Africa alone about 1.5 million new cases arise each year. How can we care for all these patients?

We have known for almost 40 years that most patients with tuberculosis can be treated in the community without increasing the risk of their infecting contacts. Indeed, the strategy of supervised outpatient therapy was developed in poor settings.2 Nevertheless, until recently, the World Health Organisation and the International Union Against Tuberculosis and Lung Disease advocated a strategy of admission to hospital for at least the first two months of treatment, primarily as a way of ensuring adherence. Indeed hospitalisation, together with the other elements of the control programme, was highly effective.3

However, the epidemic of tuberculosis associated with HIV in sub-Saharan Africa is such that hospital based care is no longer feasible. The caseload in Malawi increased from 5334 in 1985 to 19 195 in 1995 (Malawi …