Measuring the Value of a Practical Text Mining Approach to Identify Patients With Housing Issues in the Free-Text Notes in Electronic Health Record: Findings of a Retrospective Cohort Study

Abstract

Introduction: Despite the growing efforts to standardize coding for social determinants of health (SDOH), they are infrequently captured in electronic health records (EHRs). Most SDOH variables are still captured in the unstructured fields (i.e., free-text) of EHRs. In this study we attempt to evaluate a practical text mining approach (i.e., advanced pattern matching techniques) in identifying phrases referring to housing issues, an important SDOH domain affecting value-based healthcare providers, using EHR of a large multispecialty medical group in the New England region, United States. To present how this approach would help the health systems to address the SDOH challenges of their patients we assess the demographic and clinical characteristics of patients with and without housing issues and briefly look into the patterns of healthcare utilization among the study population and for those with and without housing challenges.Methods: We identified five categories of housing issues [i.e., homelessness current (HC), homelessness history (HH), homelessness addressed (HA), housing instability (HI), and building quality (BQ)] and developed several phrases addressing each one through collaboration with SDOH experts, consulting the literature, and reviewing existing coding standards. We developed pattern-matching algorithms (i.e., advanced regular expressions), and then applied them in the selected EHR. We assessed the text mining approach for recall (sensitivity) and precision (positive predictive value) after comparing the identified phrases with manually annotated free-text for different housing issues.Results: The study dataset included EHR structured data for a total of 20,342 patients and 2,564,344 free-text clinical notes. The mean (SD) age in the study population was 75.96 (7.51). Additionally, 58.78% of the cohort were female. BQ and HI were the most frequent housing issues documented in EHR free-text notes and HH was the least frequent one. The regular expression methodology, when compared to manual annotation, had a high level of precision (positive predictive value) at phrase, note, and patient levels (96.36, 95.00, and 94.44%, respectively) across different categories of housing issues, but the recall (sensitivity) rate was relatively low (30.11, 32.20, and 41.46%, respectively).Conclusion: Results of this study can be used to advance the research in this domain, to assess the potential value of EHR's free-text in identifying patients with a high risk of housing issues, to improve patient care and outcomes, and to eventually mitigate socioeconomic disparities across individuals and communities.

The Assessment of Social Determinants of Health in Postsepsis Mortality and Readmission: A Scoping Review

“…Measurement error may explain the mixed findings of independent associations between some SDH and postsepsis outcomes reported among our studies. Because data related to SDH are often not accessible through structured data fields, but embedded in free-text fields (61), continued development and application of novel clinical natural language processing (NLP) methods are needed to harness valuable SDH data from unstructured EHR data (62, 63).…”

“…Measurement error may explain the mixed findings of independent associations between some SDH and postsepsis outcomes reported among our studies. Because data (62,63). However, even with improved extraction techniques such as NLP, EHR-derived SDH data are not likely sufficient to constitute a complete and accurate set of SDH domains, as many social and behavioral determinants that may influence health and mortality such as living arrangement and economic stability are not reliably captured and recorded (64).…”

Occupation Recognition and Exploitation in Rheumatology Clinical Notes: Employing Deep Learning Models for Named Entity Recognition and Knowledge Discovery in Electronic Health Records

“…Although the relevance of this Social Determinant of Health (SDOH) [5] has been identified and studied at multiple health levels (i.e., mental health [6], health inequality [7], or self-rated health [8]), its study is overshadowed by other SDOH including gender, race, or ethnicity [9, 10]; compromising its content [11], variability [12] and quality [13]. Moreover, most SDOH are stored as free-text unstructured data [14] making them difficult to handle and use.…”

Realizing the Potential of Social Determinants Data: A Scoping Review of Approaches for Screening, Linkage, Extraction, Analysis and Interventions

“…Other articles (n = 7) describe rule-based systems paired with traditional machine learning approaches i.e., an ensemble, particularly using NLP systems such as General Architecture for Text Engineering (GATE), Clinical Language Annotation, Modeling, and Processing Toolkit (CLAMP), Extract SDOH from EHRs (EASE), Yale clinical Text Analysis and Knowledge Extraction System (cTAKES), Relative Housing Stability in Electronic Documentation (ReHouSED), and toolkits such as spaCy and medspaCy in conjunction with conditional random fields and support vector machines [110-113]. In contrast, several investigators have leveraged opensource NLP toolkits like spaCy and medspaCy without supervised learners to extractSDoH variables[114][115][116]. Other studies (n = 19) have solely leveraged traditional supervised and unsupervised learning techniques, support vector machines (SVM), logistic regression (LR), Naïve Bayes, Adaboost, Random Forest, XGBoost, Bio-ClinicalBERT, Latent Dirichlet Allocation (LDA), and bidirectional Long Short-Term Memory (BI-LSTM)[16,[117][118][119][120][121] to extract and standardize social and behavioral determinants of health (SBDoH), e.g., alcohol abuse, drug use, sexual orientation, homelessness,substance use, sexual history, HIV status, drug use, housing status, transportation needs, housing insecurity, food insecurity, financial insecurity, employment/income insecurity, insurance insecurity, and poor social support.…”