mDISH feature in Hepatic Explorer

[JimBuchanan]

Name of the chart

mDISH

Objectives

How to identify possible Hy's Law cases when ALT and/or bilirubin values are abnormal at baseline.

User Interface mockup

Features

The fold change in ALT and bilirubin is calculated based on the "baseline" ALT and bilirubin values. "Baseline" can be defined in one of two ways: 1) the defined baseline lab sample, or 2) the mean of the screening value and the Day 1 pre-dose value. The simplest programmatically would be the single baseline value.

Data Requirement

Only the laboratory data domain is required. For the single lab value method, the field would either be AVISIT="Baseline" or VISIT="Screening"

References

Aithal GP, Watkins PB, Andrade RJ. Case definition and phenotype standardization
in drug-induced liver injury. Clin Pharmacol Ther. 2011;89:806-815.

Lin X, Parks D, Painter J, et al. Validation of multivariate outlier detection analyses used to identify potential drug-induced liver injury in clinical trial populations. Drug Safety 2012;35:865-875.

[jwildfire]

Nice summary, @JimBuchanan - certainly seems doable. Are you thinking this would be a static display? Or are you looking to modify the mDish implementation in hep-explorer from the screenshot above?

[JimBuchanan]

I'm looking at this as a modification of the mDISH implementation. Looking ahead, I can foresee the drop down choices of display type can have eDISH, mDISH, fold-change from a predefined baseline, nR modified Hy's Law and, ultimately, Palt defined Hy's Law.