Health system delay in treatment of multidrug resistant tuberculosis patients in Bangladesh

BACKGROUND Bangladesh is one of the 27 high burden countries for multidrug resistant tuberculosis listed by the World Health Organization. Delay in multidrug resistant tuberculosis treatment may allow progression of the disease and affect the attempts to curb transmission of drug resistant tuberculosis. The main objective of this study was to investigate the health system delay in multidrug resistant tuberculosis treatment in Bangladesh and to explore the factors related to the delay. METHODS Information related to the delay was collected as part of a previously conducted case-control study. The current study restricts analysis to patients with multidrug resistant tuberculosis who were diagnosed using rapid diagnostic methods (Xpert MTB/RIF or the line probe assay). Information was collected by face-to-face interviews and through record reviews from all three Government hospitals providing multidrug resistant tuberculosis services, from September 2012 to April 2013. Multivariable regression analysis was performed using Bootstrap variance estimators. Definitions were as follows: Provider delay: time between visiting a provider for first consultation on MDR-TB related symptom to visiting a designated diagnostic centre for testing; Diagnostic delay: time from date of diagnostic sample provided to date of result; Treatment initiation delay: time between the date of diagnosis and date of treatment initiation; Health system delay: time between visiting a provider to start of treatment. Health system delay was derived by adding provider delay, diagnostic delay and treatment initiation delay. RESULTS The 207 multidrug resistant tuberculosis patients experienced a health system delay of median 7.1 weeks. The health system delay consists of provider delay (median 4 weeks), diagnostic delay (median 5 days) and treatment initiation delay (median 10 days). Health system delay (Coefficient: 37.7; 95 %; CI 15.0-60.4; p 0.003) was associated with the visit to private practitioners for first consultation. CONCLUSIONS Diagnosis time for multidrug resistant tuberculosis was fast using the rapid tests. However, some degree of delay was present in treatment initiation, after diagnosis. The most effective way to reduce health system delay would be through strategies such as engaging private practitioners in multidrug resistant tuberculosis control.