Data Harmonization Via Regularized Nonparametric Mixing Distribution Estimation

TL;DR

This paper introduces a novel non-parametric approach for data harmonization in dementia research, enabling conversion between different cognitive test scores and accounting for measurement error, outperforming standard methods.

Contribution

The paper develops a regularized nonparametric latent trait model with algorithms for estimation and goodness-of-fit assessment, advancing data harmonization techniques in cognitive measurement.

Findings

Effective conversion between cognitive test scores.

Outperforms standard dementia measurement techniques.

Provides tools for assessing model fit.

Abstract

Data harmonization is the process by which an equivalence is developed between two variables measuring a common trait. Our problem is motivated by dementia research in which multiple tests are used in practice to measure the same underlying cognitive ability such as language or memory. We connect this statistical problem to mixing distribution estimation. We introduce and study a non-parametric latent trait model, develop a method which enforces uniqueness of the regularized maximum likelihood estimator, show how a nonparametric EM algorithm will converge weakly to its maximizer, and additionally propose a faster algorithm for learning a discretized approximation of the latent distribution. Furthermore, we develop methods to assess goodness of fit for the mixing likelihood which is an area neglected in most mixing distribution estimation problems. We apply our method to the National Alzheimer's Coordination Center Uniform Data Set and show that we can use our method to convert between score measurements and account for the measurement error. We show that this method outperforms standard techniques commonly used in dementia research. Full code is available at https://github.com/SteveJWR/Data-Harmonization-Nonparametric.