Analyzing the symptoms in colorectal and breast cancer patients with or without type 2 diabetes using EHR data

Natural Language Processing Accurately Differentiates Cancer Symptom Information in Electronic Health Record Narratives

PURPOSE

Identifying cancer symptoms in electronic health record (EHR) narratives is feasible with natural language processing (NLP). However, more efficient NLP systems are needed to detect various symptoms and distinguish observed symptoms from negated symptoms and medication-related side effects. We evaluated the accuracy of NLP in (1) detecting 14 symptom groups (ie, pain, fatigue, swelling, depressed mood, anxiety, nausea/vomiting, pruritus, headache, shortness of breath, constipation, numbness/tingling, decreased appetite, impaired memory, disturbed sleep) and (2) distinguishing observed symptoms in EHR narratives among patients with cancer.

METHODS

We extracted 902,508 notes for 11,784 unique patients diagnosed with cancer and developed a gold standard corpus of 1,112 notes labeled for presence or absence of 14 symptom groups. We trained an embeddings-augmented NLP system integrating human and machine intelligence and conventional machine learning algorithms. NLP metrics were calculated on a gold standard corpus subset for testing.

RESULTS

The interannotator agreement for labeling the gold standard corpus was excellent at 92%. The embeddings-augmented NLP model achieved the best performance (F1 score = 0.877). The highest NLP accuracy was observed in pruritus (F1 score = 0.937) while the lowest accuracy was in swelling (F1 score = 0.787). After classifying the entire data set with embeddings-augmented NLP, we found that 41% of the notes included symptom documentation. Pain was the most documented symptom (29% of all notes) while impaired memory was the least documented (0.7% of all notes).

CONCLUSION

We illustrated the feasibility of detecting 14 symptom groups in EHR narratives and showed that an embeddings-augmented NLP system outperforms conventional machine learning algorithms in detecting symptom information and differentiating observed symptoms from negated symptoms and medication-related side effects.

Using natural language processing to analyze unstructured patient-reported outcomes data derived from electronic health records for cancer populations: a systematic review

INTRODUCTION

Patient-reported outcomes (PROs; symptoms, functional status, quality-of-life) expressed in the 'free-text' or 'unstructured' format within clinical notes from electronic health records (EHRs) offer valuable insights beyond biological and clinical data for medical decision-making. However, a comprehensive assessment of utilizing natural language processing (NLP) coupled with machine learning (ML) methods to analyze unstructured PROs and their clinical implementation for individuals affected by cancer remains lacking.

AREAS COVERED

This study aimed to systematically review published studies that used NLP techniques to extract and analyze PROs in clinical narratives from EHRs for cancer populations. We examined the types of NLP (with and without ML) techniques and platforms for data processing, analysis, and clinical applications.

EXPERT OPINION

Utilizing NLP methods offers a valuable approach for processing and analyzing unstructured PROs among cancer patients and survivors. These techniques encompass a broad range of applications, such as extracting or recognizing PROs, categorizing, characterizing, or grouping PROs, predicting or stratifying risk for unfavorable clinical results, and evaluating connections between PROs and adverse clinical outcomes. The employment of NLP techniques is advantageous in converting substantial volumes of unstructured PRO data within EHRs into practical clinical utilities for individuals with cancer.

Symptom clusters in breast cancer survivors with and without type 2 diabetes over the cancer trajectory

Objective

This study aimed to investigate symptoms and symptom clusters in breast cancer survivors (BCS) with and without type 2 diabetes across three crucial periods during the cancer trajectory (0-6 months, 12-18 months, and 24-30 months) post-initial chemotherapy.

Methods

Eight common symptoms in both BCS and individuals with diabetes were identified through natural language processing of electronic health records from January 2007 to December 2018. Exploratory factor analysis was employed to discern symptom clusters, evaluating their stability, consistency, and clinical relevance.

Results

Among the 4601 BCS in the study, 20% (n = 905) had a diabetes diagnosis. Gastrointestinal symptoms and fatigue were prevalent in both groups. While BCS in both groups exhibited an equal number of clusters, the composition of these clusters differed. Symptom clusters varied over time between BCS with and without diabetes. BCS with diabetes demonstrated less stability (repeated clusters) and consistency (same individual symptoms comprising clusters) than their counterparts without diabetes. This suggests that BCS with diabetes may experience distinct symptom clusters at pivotal points in the cancer treatment trajectory.

Conclusions

Healthcare providers must be attentive to BCS with diabetes throughout the cancer trajectory, considering intensified and/or unique profiles of symptoms and symptom clusters. Interdisciplinary cancer survivorship models are essential for effective diabetes management in BCS. Implementing a comprehensive diabetes management program throughout the cancer trajectory could alleviate symptoms and symptom clusters, ultimately enhancing health outcomes and potentially reducing healthcare resource utilization.

Predictors of Unrelieved Symptoms in All of Us Research Program Participants With Chronic Conditions

CONTEXT

Over half of American adults are diagnosed with a chronic condition, with an increasing prevalence being diagnosed with multiple chronic conditions. These adults are at higher risk for having unrelieved, co-occurring symptoms, known as symptom clusters.

OBJECTIVES

To identify symptom phenotypes of patients diagnosed with four common chronic conditions, specifically, cancer, chronic obstructive pulmonary disease, heart failure, and/or type 2 diabetes mellitus, and to understand factors that predict membership in symptomatic phenotypes.

METHODS

We conducted a retrospective, cross-sectional analysis using participant responses (N=14,127) to All of Us Research Program, a National Institutes of Health biomedical database, survey questions. We performed hierarchical clustering to generate symptom phenotypes of fatigue, emotional distress, and pain and used multinomial regression to determine if demographic, healthcare access and utilization, and health-related variables predict symptom phenotype.

RESULTS

Four phenotypes, one asymptomatic or mildly symptomatic and three highly symptomatic (characterized by severe symptoms, severe pain, and severe emotional distress), were identified. The percentage of participants belonging to the severe symptoms phenotype increased with the number of chronic conditions. Most notably, foregoing or delaying medical care and rating mental health as poor or fair increased the odds of belonging to a highly symptomatic phenotype.

CONCLUSION

We found meaningful relationships between demographic, healthcare access and utilization, and health-related factors and symptom phenotypes. With the increasing trends of American adults with one or more chronic conditions and a demand to individualize care in the precision health era, it is critical to understand the factors that lead to unrelieved symptoms.