BACKGROUND: Current antimicrobial resistance surveillance (AMR) is mainly laboratory based. This approach can have inherent biases given the potential for selective specimen submission for microbiological analysis and for its inability to map antibiotic susceptibility test results to a clinical syndrome. OBJECTIVES: To discuss the need for population-based surveillance of AMR, and highlight the pros and cons of threshold surveys. SOURCES: Studies on methodology for AMR surveillance published in the last 10 years, obtained through a PubMed search on antimicrobial resistance (all fields) and surveillance/method (MeSH term). CONTENT: We discuss the use of threshold surveys to overcome the challenge of sample size in population-bases AMR surveys, which are a suitable approach in both low- and high-resource settings. IMPLICATIONS: Scale up in the use of population-based threshold survey on the prevalence of AMR will provide necessary information to triangulate the data from routinely-reported laboratory-based AMR surveillance at the local, national, and global level.

RATIONALE: Multidrug-resistant tuberculosis (MDR-TB) is a major burden to public health in Europe. Reported treatment success rates are around 50% or less, and cure rates are even lower. OBJECTIVES: To document the management and treatment outcome in patients with MDR-TB in Europe. METHODS: We performed a prospective cohort study, analyzing management and treatment outcomes stratified by incidence of patients with MDR-TB in Europe. Treatment outcomes were compared by World Health Organization and alternative simplified definitions by the Tuberculosis Network European Trialsgroup (TBNET). MEASUREMENTS AND MAIN RESULTS: A total of 380 patients with MDR-TB were recruited and followed up between 2010 and 2014 in 16 European countries. Patients in high-incidence countries compared with low-incidence countries were treated more frequently with standardized regimen (83.2% vs. 9.9%), had delayed treatment initiation (median, 111 vs. 28 d), developed more additional drug resistance (23% vs. 5.8%), and had increased mortality (9.4% vs. 1.9%). Only 20.1% of patients using pyrazinamide had proven susceptibility to the drug. Applying World Health Organization outcome definitions, frequency of cure (38.7% vs. 9.7%) was higher in high-incidence countries. Simplified outcome definitions that include 1 year of follow-up after the end of treatment showed similar frequency of relapse-free cure in low- (58.3%), intermediate- (55.8%), and high-incidence (57.1%) countries, but highest frequency of failure in high-incidence countries (24.1% vs. 14.6%). CONCLUSIONS: Conventional standard MDR-TB treatment regimens resulted in a higher frequency of failure compared with individualized treatments. Overall, cure from MDR-TB is substantially more frequent than previously anticipated, and poorly reflected by World Health Organization outcome definitions.