Modeling factors critical for implementation of precision medicine at health systems-level: an IRT approach

Background: Through recent advances in omics technologies, precision medicine (PM) promises to fundamentally change the way we approach health, disease and illness. Imperative applications of omics-based biomarkers are gradually moving from research to clinical settings, with huge long-term clinical and public health implications. Whereas much of research in PM is mainly focused on basic biomedical discoveries, currently there is little research on the clinical implementation of omics biomarkers, especially at health systems level. Aim and Methods: This study investigated the application of multidimensional item response theory (IRT) models to validate a hypothesized PM implementation measurement model. This is a contribution to PM implementation at health systems level. Data obtained through an item-sort procedure involving 496 observations from 124 study participants formed the basis of a 22-item PMI measurement model. Conclusion: Statistical significance of the bifactor model suggests PM implementation may have to be examined using factors that reflect a single common underlying implementation construct, as well as factors that reflect unique variances for the identified four content-specific factors.