Early Identification of Maternal Cardiovascular Risk Through Sourcing and Preparing Electronic Health Record Data: Machine Learning Study

Abstract

Background Health care data are fragmenting as patients seek care from diverse sources. Consequently, patient care is negatively impacted by disparate health records. Machine learning (ML) offers a disruptive force in its ability to inform and improve patient care and outcomes. However, the differences that exist in each individual’s health records, combined with the lack of health data standards, in addition to systemic issues that render the data unreliable and that fail to create a single view of each patient, create challenges for ML. Although these problems exist throughout health care, they are especially prevalent within maternal health and exacerbate the maternal morbidity and mortality crisis in the United States. Objective This study aims to demonstrate that patient records extracted from the electronic health records (EHRs) of a large tertiary health care system can be made actionable for the goal of effectively using ML to identify maternal cardiovascular risk before evidence of diagnosis or intervention within the patient’s record. Maternal patient records were extracted from the EHRs of a large tertiary health care system and made into patient-specific, complete data sets through a systematic method. Methods We outline the effort that was required to define the specifications of the computational systems, the data set, and access to relevant systems, while ensuring that data security, privacy laws, and policies were met. Data acquisition included the concatenation, anonymization, and normalization of health data across multiple EHRs in preparation for their use by a proprietary risk stratification algorithm designed to establish patient-specific baselines to identify and establish cardiovascular risk based on deviations from the patient’s baselines to inform early interventions. Results Patient records can be made actionable for the goal of effectively using ML, specifically to identify cardiovascular risk in pregnant patients. Conclusions Upon acquiring data, including their concatenation, anonymization, and normalization across multiple EHRs, the use of an ML-based tool can provide early identification of cardiovascular risk in pregnant patients.

Use of Machine Learning for Early Detection of Maternal Cardiovascular Conditions: Retrospective Study Using Electronic Health Record Data

“…Specifically, this study offers a novel tool for monitoring cardiovascular risk during pregnancy using real-time trends, thereby assisting health care professionals in the provision of perinatal care for high-risk patients. Consistent with earlier inquiries on the deployment of ML within obstetrics, our ML model has undergone technical validation [ 10 ]. Nonetheless, the imperative remains for further research to establish the clinical validity of the model.…”

“…Therefore, the study design prioritized tracing outcomes for individuals flagged by the algorithm to assess its accuracy in predicting risk compared to diagnoses documented in medical records. For more extensive technical detail related to the development of HOPE-CAT, including data extraction, specifications of the computational systems, the development of the data set, and access to relevant systems, please see our previous work, Early Identification of Maternal Cardiovascular Risk Through Sourcing and Preparing Electronic Health Record Data: Machine Learning Study [ 10 ].…”

“…Within a virtual server environment created for this study, we systematically cleaned and standardized the EHR data (eg, organized and matched fields across tables and databases) before the deployment of an analysis by HOPE-CAT [ 10 ]. While the data set was sourced within a single health care system, standardization was necessary as data were sourced from multiple EHRs.…”

“…Data from each encounter included patient demographics, physical findings, symptoms, and medical history and were analyzed by HOPE-CAT to identify changes and trends. In instances where HOPE-CAT detected any sign of risk, based on the training criteria, a risk profile was generated for the given patient and encounter [ 10 ]. Due to the nature of the available data, it was not possible to determine whether a patient’s pregnancy was their first.…”

“…Hoffman et al [8] used ML to accurately predict maternal readmission due to the complications of hypertensive disorders. ML models have also shown promising results for predicting maternal risk of hypertensive disorders of pregnancy and other cardiovascular conditions [10][11][12], including improved prediction accuracy when gestational age, epidemiology, hemodynamics, and biochemistry data are incorporated [9]. However, a fundamental gap remains to identify pregnant individuals at higher risk of morbidity and mortality using the power of ML and big data including electronic health records (EHRs).…”