Estimating MIC distributions and cutoffs through mixture models: an application to establish M. Tuberculosis resistance.

Abstract

Antimicrobial resistance is becoming a major threat to public health throughout the world. Researchers from around the world are attempting to contrast it by developing both new antibiotics and patient-specific treatments. It is, therefore, necessary to study these treatments, via phenotypic tests, and it is essential to have robust methods available to analyze the resistance patterns to medication, which could be applied to both new treatments and to new phenotypic tests. A general method is here proposed to study minimal inhibitory concentration (MIC) distributions and fixed breakpoints in order to separate sensible from resistant strains. The method has been applied to a new 96-well microtiter plate.