Social Determinants of Health in EMS Records: A Mixed-methods Analysis Using Natural Language Processing and Qualitative Content Analysis.

Introduction: Social determinants of health (SDoH) are known to impact the health and well-being of patients. However, information regarding them is not always collected in healthcare interactions, and healthcare professionals are not always well-trained or equipped to address them. Emergency medical services (EMS) professionals are uniquely positioned to observe and attend to SDoH because of their presence in patients' environments; however, the transmission of that information may be lost during transitions of care. Documentation of SDoH in EMS records may be helpful in identifying and addressing patients' insecurities and improving their health outcomes. Our objective in this study was to determine the presence of SDoH information in adult EMS records and understand how such information is referenced, appraised, and linked to other determinants by EMS personnel. Methods: Using EMS records for adult patients in the 2019 ESO Data Collaborative public-use research dataset using a natural language processing (NLP) algorithm, we identified free-text narratives containing documentation of at least one SDoH from categories associated with food, housing, employment, insurance, financial, and social support insecurities. From the NLP corpus, we randomly selected 100 records from each of the SDoH categories for qualitative content analysis using grounded theory. Results: Of the 5,665,229 records analyzed by the NLP algorithm, 175,378 (3.1%) were identified as containing at least one reference to SDoH. References to those SDoH were centered around the social topics of accessibility, mental health, physical health, and substance use. There were infrequent explicit references to other SDoH in the EMS records, but some relationships between categories could be inferred from contexts. Appraisals of patients' employment, food, and housing insecurities were mostly negative. Narratives including social support and financial insecurities were less negatively appraised, while those regarding insurance insecurities were mostly neutral and related to EMS operations and procedures. Conclusion: The social determinants of health are infrequently documented in EMS records. When they are included, they are infrequently explicitly linked to other SDoH categories and are often negatively appraised by EMS professionals. Given their unique position to observe and share patients' SDoH information, EMS professionals should be trained to understand, document, and address SDoH in their practice.

A 7 Year Summary of Emergency Department Visits by Patients With Mental Health Disorders.

OBJECTIVES: Emergency departments (EDs) have been increasingly utilized over time for psychiatric care. While multiple studies have assessed these trends in nationally representative data, few have evaluated these trends in state-level data. This investigation seeks to understand the mental health-related ED burden in North Carolina (NC) by describing trends in ED visits associated with a mental health diagnosis (MHD) over time. METHODS: Using data from NC DETECT, this investigation describes trends in NC ED visits from January 1, 2008 through December 31, 2014 by presence of a MHD code. A visit was classified by the first listed MHD ICD-9-CM code in the surveillance record and MHD codes were grouped into related categories for analysis. Visits were summarized by MHD status and by MHD category. RESULTS: Over 32 million ED visits were recorded from 2008 to 2014, of which 3,030,746 (9.4%) were MHD-related visits. The average age at presentation for MHD-related visits was 50 years (SD 23.5) and 63.9% of visits were from female patients. The proportion of ED visits with a MHD increased from 8.3 to 10.2% from 2008 to 2014. Annually and overall, the largest diagnostic category was stress/anxiety/depression. Hospital admissions resulting from MHD-related visits declined from 32.2 to 18.5% from 2008 to 2014 but remained consistently higher than the rate of admissions among non-MHD visits. CONCLUSION: Similar to national trends, the proportion of ED visits associated with a MHD in NC has increased over time. This indicates a need for continued surveillance, both stateside and nationally, in order to inform future efforts to mitigate the growing ED burden.

Characteristics and Experiences of Women Physicians and Professionals in NAEMSP.

Introduction: One of the six guiding principles of the EMS Agenda 2050 is to foster a socially equitable care delivery system. A specific recommendation within this principle is that "local EMS leadership, educators and clinicians [should] reflect the diversity of their communities." Research has shown that women comprise a minority of emergency medicine services (EMS) field clinicians. In academic settings, women are represented at lower rates among experienced EMS faculty than within Emergency Medicine clinicians or faculty at large. The reasons for these differences are also unknown. Little data exist describing the number or experience of female physicians and professionals in EMS.Purpose: Our objective was to describe the composition and experiences of EMS physicians, researchers and professionals who participate in the Women in EMS group of the National Association of EMS Physicians (NAEMSP).Methods: We performed a cross-sectional, mixed-methods descriptive study of women belonging to the Women in EMS Committee of NAEMSP. A survey was sent to the 143 members of this group using a list-serve, and the data was collected in Redcap.Results: Seventy-four people completed the survey. Respondents were 96% female, 82% Caucasian, 11% underrepresented minorities (URM), and 7% LGBTQI. Of the 88% that are physicians, 78% are board certified in Emergency Medicine, compared to 55% in EMS. Forty-eight percent reported they received some form of mentorship. Among these respondents, a minority reported female mentorship, which was usually from a remote rather than local mentor (41% vs. 15%). Eighty-three percent of respondents had experienced some form of discrimination or harassment in their career, but only 68% reported their workplace culture discourages such behavior. Thirty-three percent of respondents report receiving unequal recognition because of gender. Thematic evaluation of the qualitative responses showed that respondents felt there were fewer barriers to mentorship and professional advancement opportunities in local work versus national engagement.Conclusions: In a survey evaluating representation of female professionals in EMS, participants reported on their career representations, and experiences of gender-based inequity within their EMS career settings. Several opportunities exist to improve diversity, equity, and inclusion for women in EMS based on our findings.

Identification of social determinants of health using multi-label classification of electronic health record clinical notes.

OBJECTIVES: Social determinants of health (SDH), key contributors to health, are rarely systematically measured and collected in the electronic health record (EHR). We investigate how to leverage clinical notes using novel applications of multi-label learning (MLL) to classify SDH in mental health and substance use disorder patients who frequent the emergency department. METHODS AND MATERIALS: We labeled a gold-standard corpus of EHR clinical note sentences (N = 4063) with 6 identified SDH-related domains recommended by the Institute of Medicine for inclusion in the EHR. We then trained 5 classification models: linear-Support Vector Machine, K-Nearest Neighbors, Random Forest, XGBoost, and bidirectional Long Short-Term Memory (BI-LSTM). We adopted 5 common evaluation measures: accuracy, average precision-recall (AP), area under the curve receiver operating characteristic (AUC-ROC), Hamming loss, and log loss to compare the performance of different methods for MLL classification using the F1 score as the primary evaluation metric. RESULTS: Our results suggested that, overall, BI-LSTM outperformed the other classification models in terms of AUC-ROC (93.9), AP (0.76), and Hamming loss (0.12). The AUC-ROC values of MLL models of SDH related domains varied between (0.59-1.0). We found that 44.6% of our study population (N = 1119) had at least one positive documentation of SDH. DISCUSSION AND CONCLUSION: The proposed approach of training an MLL model on an SDH rich data source can produce a high performing classifier using only unstructured clinical notes. We also provide evidence that model performance is associated with lexical diversity by health professionals and the auto-generation of clinical note sentences to document SDH.

Identifying Patient Phenotype Cohorts Using Prehospital Electronic Health Record Data.

Objective: Emergency medical services (EMS) provide critical interventions for patients with acute illness and injury and are important in implementing prehospital emergency care research. Retrospective, manual patient record review, the current reference-standard for identifying patient cohorts, requires significant time and financial investment. We developed automated classification models to identify eligible patients for prehospital clinical trials using EMS clinical notes and compared model performance to manual review.Methods: With eligibility criteria for an ongoing prehospital study of chest pain patients, we used EMS clinical notes (n = 1208) to manually classify patients as eligible, ineligible, and indeterminate. We randomly split these same records into training and test sets to develop and evaluate machine-learning (ML) algorithms using natural language processing (NLP) for feature (variable) selection. We compared models to the manual classification to calculate sensitivity, specificity, accuracy, positive predictive value, and F1 measure. We measured clinical expert time to perform review for manual and automated methods.Results: ML models' sensitivity, specificity, accuracy, positive predictive value, and F1 measure ranged from 0.93 to 0.98. Compared to manual classification (N = 363 records), the automated method excluded 90.9% of records as ineligible and leaving only 33 records for manual review.Conclusions: Our ML derived approach demonstrates the feasibility of developing a high-performing, automated classification system using EMS clinical notes to streamline the identification of a specific cardiac patient cohort. This efficient approach can be leveraged to facilitate prehospital patient-trial matching, patient phenotyping (i.e. influenza-like illness), and create prehospital patient registries.

Ability of Older Adults to Report Elder Abuse: An Emergency Department-Based Cross-Sectional Study.

OBJECTIVES: To characterize assessments of a patient's ability to report elder abuse within the context of an emergency department (ED)-based screen for elder abuse. DESIGN: Cross-sectional study in which participants were screened for elder abuse and neglect. SETTING: Academic ED in the United States. PARTICIPANTS: Patients, aged 65 years and older, presenting to an ED for acute care were assessed by trained research assistants or nurses. MEASUREMENTS: All patients completed the four-item Abbreviated Mental Test 4 (AMT4), then completed a safety interview (using the Emergency Department Senior Abuse Identification tool) designed to detect multiple domains of elder abuse and received a physical examination. Based on the cognitive assessment and safety interview, assessors ranked their confidence in the patient's ability to report abuse as absolutely confident, confident, somewhat confident, or not confident. To assess interrater reliability, two assessors independently rated confidence for a subset of patients. RESULTS: Assessors suspected elder abuse in 18 of 276 patients (6.5%). Assessors were absolutely confident in the patient's ability to report abuse for 95.7% of patients, confident for 2.5%, somewhat confident for 1.5%, and not confident for 0.3%. Among patients with an AMT4 of 4 (n = 249), assessors were confident or absolutely confident in 100% of patients. Among patients with an AMT4 of less than 4 (n = 27), they were confident or absolutely confident in the patient's ability to report abuse for 81% of patients, including 11 of 12 patients with mild cognitive impairment and 7 of 11 patients with severe cognitive impairment. For patients receiving paired evaluations (n = 131), agreement between assessors regarding patient ability to report abuse was 97% (κ = 0.5). CONCLUSIONS: In this sample of older adults receiving care in an ED, research assistants and nurses felt that the vast majority were able to report elder abuse, including many patients with cognitive impairment. J Am Geriatr Soc 68:170-175, 2019.