Using Domain Knowledge and Data-Driven Insights for Patient Similarity Analytics.

Ronald Wihal Oei,Wei-Ying Tan,Wynne Hsu,Hao Sen Andrew Fang,Mong-Li Lee,Ngiap-Chuan Tan

doi:10.3390/jpm11080699

Abstract

Patient similarity analytics has emerged as an essential tool to identify cohorts of patients who have similar clinical characteristics to some specific patient of interest. In this study, we propose a patient similarity measure called D3K that incorporates domain knowledge and data-driven insights. Using the electronic health records (EHRs) of 169,434 patients with either diabetes, hypertension or dyslipidaemia (DHL), we construct patient feature vectors containing demographics, vital signs, laboratory test results, and prescribed medications. We discretize the variables of interest into various bins based on domain knowledge and make the patient similarity computation to be aligned with clinical guidelines. Key findings from this study are: (1) D3K outperforms baseline approaches in all seven sub-cohorts; (2) our domain knowledge-based binning strategy outperformed the traditional percentile-based binning in all seven sub-cohorts; (3) there is substantial agreement between D3K and physicians (κ = 0.746), indicating that D3K can be applied to facilitate shared decision making. This is the first study to use patient similarity analytics on a cardiometabolic syndrome-related dataset sourced from medical institutions in Singapore. We consider patient similarity among patient cohorts with the same medical conditions to develop localized models for personalized decision support to improve the outcomes of a target patient.

Highlights

Diabetes, hypertension, and dyslipidaemia (DHL) are three of the most prevalent chronic diseases
We develop a patient similarity measure called D3K, which stands for data-driven and domain knowledge; our D3K approach takes into consideration domain knowledge and data-driven insights to retrieve patients that are clinically similar to a target patient
Our proposed method, which discretizes variables based on current clinical guidelines and domain knowledge, is and evaluate which 10 patients in the list were most similar to the index patient

Summary

Introduction

Hypertension, and dyslipidaemia (DHL) are three of the most prevalent chronic diseases. The prevalence of these three conditions is about 8.5%, 25%, and. Diabetes alone was estimated to contribute USD 760 billion in global health expenditure in 2019, and this is projected to grow to USD 825 billion by 2030 and USD 845 billion by 2045 [4]. Many studies have analyzed large populations to provide statistical summaries of an “average” patient. These studies are expensive, time-consuming, and often subject to selection bias [6]. They may not be applicable to patients whose conditions differ from this “average” patient [7]

Objectives

Methods

Results

Discussion

Conclusion