Natural language processing of lifestyle modification documentation

Kimberly Shoenbill,Eneida A Mendonca,Maureen Smith,Heather Johnson,Yiqiang Song,Lisa Gress

doi:10.1177/1460458218824742

Abstract

Lifestyle modification, including diet, exercise, and tobacco cessation, is the first-line treatment of many disorders including hypertension, obesity, and diabetes. Lifestyle modification data are not easily retrieved or used in research due to their textual nature. This study addresses this knowledge gap using natural language processing to automatically identify lifestyle modification documentation from electronic health records. Electronic health record notes from hypertension patients were analyzed using an open-source natural language processing tool to retrieve assessment and advice regarding lifestyle modification. These data were classified as lifestyle modification assessment or advice and mapped to a coded standard ontology. Combined lifestyle modification (advice and assessment) recall was 99.27 percent, precision 94.44 percent, and correct classification 88.15 percent. Through extraction and transformation of narrative lifestyle modification data to coded data, this critical information can be used in research, metric development, and quality improvement efforts regarding care delivery for multiple medical conditions that benefit from lifestyle modification.

Highlights

Background and significanceNational guidelines have recommended the use of lifestyle modification in the treatment and prevention of prevalent disorders plaguing the US today including hypertension, obesity, coronary artery disease, diabetes, peripheral vascular disease, and cancer.[1,2,3,4,5,6,7,8,9] The top seven US health risksThere is a need for automated methods that can facilitate the extraction and integration of lifestyle behavior factors for use in research
Results from testing the natural language processing (NLP) tool refinement process for combined lifestyle modification retrieval were excellent with 99.27 percent recall and 94.44 percent precision and an F-measure of 96.79 percent
concept unique identifiers (CUIs) mapping for lifestyle modification was very good with phrases correctly classified as advice or assessment 88.15 percent

Summary

Introduction

National guidelines have recommended the use of lifestyle modification in the treatment and prevention of prevalent disorders plaguing the US today including hypertension, obesity, coronary artery disease, diabetes, peripheral vascular disease, and cancer.[1,2,3,4,5,6,7,8,9] The top seven US health risks. There is a need for automated methods that can facilitate the extraction and integration of lifestyle behavior factors for use in research. In order to accomplish this task more efficiently, this study used natural language processing (NLP) software tools and processes that can automatically extract text-based information. NLP tools can process many thousands of notes per hour.[22] This technology makes larger chart abstractions feasible and allows a more comprehensive evaluation of documentation of lifestyle modification

Objectives

Methods

Results

Conclusion