Viewpoint: do we have enough data to estimate the current burden of tuberculosis? The example of Bangladesh

Abstract

National tuberculosis (TB) programmes need accurate estimates of the burden of TB in order to assess case detection, and progress towards the millennium development goals (MDGs). We reviewed nationwide epidemiological data on TB burden in Bangladesh to decide whether additional information is needed, in order to assess the current burden of TB. We collected all nationwide epidemiological information: notification data for the period 1995-2004; reports of TB prevalence surveys in 1964-1966 and 1987-1988 and a tuberculin survey in 1964-1966. Data were evaluated for usefulness to provide information about the current burden of TB. The TB prevalence surveys from 1964 to 1966 and 1987 to 1988 provide very different results, respectively, 127/100 000 and 471/100 000 smear positives. In 1964-1966, the annual risk of infection was 0.6% in the 5- to 9-year age group and 1.2% in those aged between 10 and 14 years. The notification data show an increase from 18/100 000 in 1995 to 46/100 000 smear positives in 2004. Interpretation and extrapolation of the existing epidemiological data of Bangladesh to estimate the current burden of TB is difficult. Many of the 22 high-burden countries have as few or even less survey data than Bangladesh. To enable the evaluation of progress towards the MDGs and to be able to assess the global burden of TB, epidemiological surveys are needed.