ABSTRACT
OBJECTIVES: To evaluate the feasibility of the Zero TB Indicator Framework as a tool for assessing the quality of tuberculosis (TB) case-finding, treatment and prevention services in Mongolia. SETTING: Primary health centres, TB dispensaries, and surrounding communities in four districts of Mongolia. DESIGN: Three retrospective cross-sectional cohort studies, and two longitudinal studies each individually nested in one of the cohort studies. PARTICIPANTS: 15 947 community members from high TB-risk populations; 8518 patients screened for TB in primary health centres and referred to dispensaries; 857 patients with index TB and 2352 household contacts. PRIMARY AND SECONDARY OUTCOME MEASURES: 14 indicators of the quality of TB care defined by the Zero TB Indicator Framework and organised into three care cascades, evaluating community-based active case-finding, passive case-finding in health facilities and TB screening and prevention among close contacts; individual and health-system predictors of these indicators. RESULTS: The cumulative proportions of participants receiving guideline-adherent care varied widely, from 96% for community-based active case-finding, to 79% for TB preventive therapy among household contacts, to only 67% for passive case-finding in primary health centres and TB dispensaries (range: 29%-80% across districts). The odds of patients completing active TB treatment decreased substantially with increasing age (aOR: 0.76 per decade, 95% CI: 0.71 to 0.83, p<0.001) and among men (aOR: 0.56, 95% CI: 0.36 to 0.88, p=0.013). Contacts of older index patients also had lower odds of initiating and completing of TB preventive therapy (aOR: 0.60 per decade, 95% CI: 0.38 to 0.93, p=0.022). CONCLUSIONS: The Zero TB Framework provided a feasible and adaptable approach for using routine surveillance data to evaluate the quality of TB care and identify associated individual and health system factors. Future research should evaluate strategies for collecting process indicators more efficiently; gather qualitative data on explanations for low-quality care; and deploy quality improvement interventions.