A new semi-parametric mixture model for interval censored data, with applications in the field of antimicrobial resistance

Abstract

Antimicrobial resistance has become one of the main public health burdens of the last decades, and monitoring the development and spread of non-wild-type isolates has therefore gained increased interest. Monitoring is performed, based on the minimum inhibitory concentration (MIC) values, which are collected through the application of dilution experiments. For a given antimicrobial, it is common practice to dichotomize the obtained MIC distribution according to a cut-off value, in order to distinguish between susceptible wild-type isolates and non-wild-type isolates exhibiting reduced susceptibility to the substance. However, this approach hampers the ability to further study the characteristics of the non-wild type component of the distribution as information on the MIC distribution above the cut-off value is lost. As an alternative, a semi-parametric mixture model is presented, which is able to estimate the full continuous MIC distribution, thereby taking all available information into account. The model is based on an extended and censored-adjusted version of the penalized mixture approach often used in density estimation. A simulation study was carried out, indicating a promising behaviour of the new semi-parametric mixture model in the field of antimicrobial susceptibility testing.