摘要：ABSTRACT Background : The public–private mix (PPM) strategy has strengthened tuberculosis care and control in many countries. Indonesia, a country with a high tuberculosis burden, has a low tuberculosis case detection rate (CDR), despite PPM implementation in 2003. The PPM in Indonesia involves primary healthcare centers, hospitals, and specialized chest clinics. The long-term impact of the strategy is unknown. Objective: We aimed to explore the case detection achievements of the tuberculosis program since PPM implementation in Central Java in 2003. Methods : This retrospective cohort study covered the period 1 January 2000 to 31 December 2014. The data from tuberculosis patients treated in all health facilities in Central Java implementing directly observed treatment short-course, recorded via a standardized form, were analyzed after being validated by the Office of Health of Central Java Province. We evaluated the CDR, case notification rate, and total number of cases, using linear regression to analyze the temporal trends of those indicators in the phases of PPM implementation. Results : The CDR increased during the initial phase (2000–2005), decreased during the mid-phase (2006–2009), and increased slightly during the late phase (2010–2014), ranging from 13 to 61.72. These trends were observed despite a steady increase in the number of participating healthcare facilities. The regression analysis showed that the CDR of referral institutions contributed the most to the total CDR of Central Java Province. Many of the smear-negative tuberculosis cases recorded at primary healthcare centers may have been smear positive; this probable misclassification could have been partially avoided if more specific and sensitive diagnostic tools were available. Conclusions : The CDR remains below the national target (70%). Early awareness of a negative trend in certain program indicators is important to ensure program sustainability. Careful observation of the indicator pattern will secure the long-term success of the program.
