Kernel Assisted Learning for Personalized Dose Finding

TL;DR

This paper introduces a kernel assisted learning method for estimating optimal personalized dose rules, reducing patient risk and costs by improving dose decision accuracy with robust statistical inference.

Contribution

The paper presents a novel kernel assisted learning approach for personalized dose finding that is robust to model misspecification and applicable to various continuous decision problems.

Findings

Successfully identifies optimal dose rules in simulations

Produces favorable expected outcomes in population studies

Demonstrates effectiveness on warfarin dosing data

Abstract

An individualized dose rule recommends a dose level within a continuous safe dose range based on patient level information such as physical conditions, genetic factors and medication histories. Traditionally, personalized dose finding process requires repeating clinical visits of the patient and frequent adjustments of the dosage. Thus the patient is constantly exposed to the risk of underdosing and overdosing during the process. Statistical methods for finding an optimal individualized dose rule can lower the costs and risks for patients. In this article, we propose a kernel assisted learning method for estimating the optimal individualized dose rule. The proposed methodology can also be applied to all other continuous decision-making problems. Advantages of the proposed method include robustness to model misspecification and capability of providing statistical inference for the estimated parameters. In the simulation studies, we show that this method is capable of identifying the optimal individualized dose rule and produces favorable expected outcomes in the population. Finally, we illustrate our approach using data from a warfarin dosing study for thrombosis patients.