Tuberculosis (TB) is an infectious disease of immense global public health importance; in 2006 alone, there were more than 9 million incident cases of TB and over 1.7 deaths attributable to this disease. The causative agent of TB, Mycobacterium tuberculosis, is transmitted through the respiratory route. While the majority of TB cases can be cured with relatively long treatment courses (6-9 month) with standard combinations of antibiotics (generally including 4 drugs during the initial months of therapy), the appearance of drug resistant forms of TB compromise the effectiveness of therapy and threaten to undermine the success of standard treatment protocols in areas where the prevalence of drug resistant TB is increasing.
M. tuberculosis initially acquires a drug resistant phenotype through the sporadic appearance of chromosomal mutations. Selection of these rare drug-resistant mutants occurs when individuals receive inadequate drug treatment; this occurs most frequently when adherence to combination drug regimens is not adequately maintained. Individuals who acquire drug-resistance while on therapy may then transmit drug resistant infection to their respiratory contacts. Strains of M. tuberculosis with combined resistance to at least the two most effective first-line drugs (isoniazid and rifampin) are categorized as multidrug resistant (MDR). MDR TB represents an important clinical entity since individuals with MDR TB have a substantially reduced probability of cure with standard combination chemotherapy.
In 2008, the World Health Organization released a report suggesting that the annual incidence of MDR TB may approach 500,000 persons per year and that around 3% of new cases and 19% of recurrent cases had MDR disease. These estimates are reason for concern, but they probably represent an underestimate; the actual global burden of drug resistant TB remains unclear due to geographic gaps in reporting (including in areas likely to be most affected by drug-resistance), failure to include cases being treated in the private sector, and difficulty in detecting and measuring the occurrence of acquired resistance. A highly publicized report of an outbreak of extensively drug resistant TB (XDR TB: MDR TB plus additional resistance to at least one fluoroquinolone and one second-line injectable antibiotic) in South Africa highlighted the importance of MDRTB surveillance, particularly after subsequent analysis of banked specimens indicated that XDRTB is present in at least 50 countries but had gone undetected by current surveillance systems. Expanding surveillance to more locations is clearly needed, but with scarce resources, there is a need for quantitative analyses to improve the efficiency and accuracy of surveillance methods.
As part of MIDAS activities, we plan to develop models and statistical methods to address several limitations of current and proposed approaches for conducting TB drug resistance surveys. In particular we aim to develop methods to i) improve the detection of trends in drug resistant TB; ii) correct for bias due to non-sampling of cases within the private sector; and iii) account for the operating characteristics of different drug resistance detection/testing methodologies.