Evaluating the Effectiveness of Complementary and Integrative Health Therapies in Preventing Postpartum Depression: A Target Trial Emulation Study

2025 Annual Symposium On Demand

Presentation Time: 02:00 PM - 02:12 PM

Abstract Keywords: Real-World Evidence Generation, Natural Language Processing, Patient / Person Generated Health Data (Patient Reported Outcomes)
Primary Track: Applications
Programmatic Theme: Clinical Informatics

This study assessed Complementary and Integrative Health (CIH) therapies for postpartum depression (PPD) using electronic health records (EHR) and target trial emulation. We analyzed four CIH interventions (acupuncture, chiropractic, aromatherapy, omega-3 fatty acids) identified via NLP from clinical notes, with propensity score matching to address confounding. The primary outcome PPD incidence within 12 months showed no significant reduction with any therapy (all p>0.05). However, secondary analysis revealed omega-3 supplementation and chiropractic significantly improved PHQ-9 scores (omega-3: p<0.001; chiropractic: p=0.021), while aromatherapy exhibited mixed effects and acupuncture demonstrated no measurable impact (p>0.05). These results suggest select CIH therapies may alleviate PPD symptoms despite lacking preventive effects, highlighting their potential as adjunctive treatments. By leveraging real-world data and target trial emulation methods, this study provides a framework for evaluating non-pharmacological interventions in perinatal mental health. Further research should validate these findings and explore mechanisms of action.

Speaker:
Huixue Zhou, PhD
University of Minnesota

Authors:
Huixue Zhou, PhD - University of Minnesota; Yiye Zhang, PhD - Weill Cornell Medicine; Zhenxing Xu, PhD - Weill Cornell Medicine; Chang Su, PhD - Weill Cornell Medicine; Kelvin Lim, MD - University of Minnesota; Andrea Johnson, MD - University of Minnesota; Nili Solomonov, PhD - Weill Cornell Medicine; Fei Wang, PhD - Weill Cornell Medicine; Rui Zhang, PhD, FAMIA, FACMI - University of Minnesota, Twin Cities;

---

Identifying Latent Patterns of Intimate Partner Violence Using Electronic Health Records and Association Rule Mining

2025 Annual Symposium On Demand

Presentation Time: 02:12 PM - 02:24 PM

Abstract Keywords: Clinical Decision Support, Public Health, Data Mining
Primary Track: Applications
Programmatic Theme: Public Health Informatics

Intimate partner violence (IPV) remains a significant public health concern, yet its identification in emergency department (ED) settings is often challenging, as IPV-related ICD-10 are infrequently and inconsistently used. This study employs electronic health record (EHR) data and association rule mining to identify latent patterns indicative of IPV among female patients in ED. We applied the Apriori algorithm to identify associations between co-occurring health conditions in confirmed IPV survivors and non-IPV patients. The analysis revealed distinct patterns in both groups, with IPV survivors showing stronger links between physical injuries, mental health conditions (e.g., depression, suicidal ideation), and socioeconomic stressors. Non-IPV cases were primarily associated with anxiety disorders and family-related stressors. These findings highlight the potential of informatics-driven approaches to improve IPV detection by revealing subtle clinical signals that may not be overtly disclosed. Our work paves the way for developing data-driven tools to enhance IPV screening and clinical decision-making in ED.

Speaker:
Azade Tabaie, PhD
MedStar Health Research Institute

Authors:
Azade Tabaie, PhD - MedStar Health Research Institute; Shelby Wyand, MPH - MedStar Health Research Institute; Fan Cao, BSc - Georgetown University; Lia Losonczy, MD, MPH - George Washington University; Sonita Bennett - MedStar Health; Leah Blackall, MPH - MedStar Health Research Institute; Alina Potts, MPH - George Washington University; Allan Fong, MS - MedStar Health;

---

Understanding Stigmatizing Language Lexicons: A Comparative Analysis in Clinical Contexts

2025 Annual Symposium On Demand

Presentation Time: 02:24 PM - 02:36 PM

Abstract Keywords: Health Equity and Social Determinants of Health (SDoH), Health Equity, Quantitative Methods, Racial disparities, Fairness and elimination of bias, Natural Language Processing
Primary Track: Foundations
Programmatic Theme: Public Health Informatics

Stigmatizing language results in healthcare inequities, yet there is no universally accepted or standardized lexicon defining which words, terms, or phrases constitute stigmatizing language in healthcare. We conducted a systematic search of the literature to identify existing stigmatizing language lexicons and then analyzed them comparatively to examine: 1) similarities and discrepancies between these lexicons, and 2) the distribution of positive, negative, or neutral terms based on an established sentiment dataset. Our search identified four lexicons. The analysis results revealed moderate semantic similarity among them, and that most stigmatizing terms are related to judgemental expressions by clinicians to describe perceived negative behaviors. Sentiment analysis showed a predominant proportion of negatively classified terms, though variations exist across lexicons. Our findings underscore the need for a standardized lexicon and highlight challenges in defining stigmatizing language in the clinical texts.

Speaker:
Yiliang Zhou, PhD
University of California, Irvine

Authors:
Yiliang Zhou, PhD - University of California, Irvine; Tianchu Lyu, PhD - University of California, Irvine; Di Hu, Master of Science in Information Systems - University of California - Irvine; Alexandra Beck, Bachelor - University of California, Irvine; Jasmine Dhillon, Bachelor - University of California, Irvine; Gelareh Sadigh, MD - University of California, Irvine; Kai Zheng, PhD - University of California, Irvine;

---

Leveraging consumer wearables to passively and objectively measure antidepressant effectiveness

2025 Annual Symposium On Demand

Presentation Time: 02:36 PM - 02:48 PM

Abstract Keywords: Mobile Health, Personal Health Informatics, Patient / Person Generated Health Data (Patient Reported Outcomes)
Primary Track: Applications
Programmatic Theme: Consumer Health Informatics

Antidepressants are widely used in the U.S., but proper selection for a patient is a trial-and-error process. Using the All of Us dataset, we explored data from wearables to objectively assess antidepressant effectiveness. We observed increased physical activity and sleep after antidepressant use and in those with symptom improvement. These findings support the feasibility of digital biomarkers to measure antidepressant effectiveness, with methods applicable to other medications and mental health disorders.

Speaker:
Eric Hurwitz, PhD
University of North Carolina Chapel Hill

Authors:
Eric Hurwitz, PhD - University of North Carolina Chapel Hill; Evan Connelly, BS - University of North Carolina at Chapel Hill; Danielle Lowe, MD, PhD - University of North Carolina at Chapel Hill; Patrick Sullivan, MD - University of North Carolina at Chapel Hill; Melissa Haendel, PhD - University of North Carolina at Chapel Hill;

---

Poor Detection, Strong Rephrasing: ChatGPT Divergent Performance with Stigmatizing Language in EHRs

2025 Annual Symposium On Demand

Presentation Time: 02:48 PM - 03:00 PM

Abstract Keywords: Fairness and Elimination of Bias, Large Language Models (LLMs), Diversity, Equity, Inclusion, and Accessibility, Natural Language Processing
Primary Track: Applications
Programmatic Theme: Clinical Research Informatics

Stigmatizing language in electronic health records can undermine patient trust and exacerbate health disparities. This study evaluated ChatGPT-4o’s ability to identify and rephrase stigmatizing language in 140 clinical notes from birth admission. While ChatGPT demonstrated strong rephrasing performance (average scores ≥2.7/3 for de-stigmatization, clarity, and faithfulness), its identification accuracy was limited, with low precision (0.41) and occasional hallucinations. These findings underscore the need for clinician oversight and improved detection approaches before clinical integration.

Speaker:
Zhihong Zhang, PhD
Columbia University

Authors:
Zhihong Zhang, PhD - Columbia University; Jihye Kim Scroggins, PhD - School of Nursing, University of North Carolina at Chapel Hill; Sarah Harkins, MPhil, BSN, RN - Columbia University School of Nursing; Ismael Ibrahim Hulchafo, MD, MS - Columbia University School of Nursing; Hans Moen, Department of Computer Science, Aalto University - Aalto University Department of Computer Science; Michele Tadiello, Master's degree - Columbia University Irving Medical Center; Veronica Barcelona, PhD - Columbia University School of Nursing; Maxim Topaz, PhD - Columbia University School of Nursing;

---

Interpretable Machine Learning to Identify Risk Factors for Recidivism in Intimate Partner Violence

2025 Annual Symposium On Demand

Presentation Time: 03:00 PM - 03:12 PM

Abstract Keywords: Machine Learning, Fairness and elimination of bias, Public Health, Data Mining, Healthcare Quality, Clinical Decision Support, Quantitative Methods
Primary Track: Applications

Intimate Partner Violence (IPV) remains a significant global health issue with severe consequences ranging from physical injury to death, with rates rising in recent years. Prediction of recidivism is critical for prevention and treatment. Using data from a four-year clinical study, we develop interpretable machine-learning models to identify features for physical assault recidivism among IPV offenders. To standardize clinician-assigned severity scores and address non-linear associations, we apply filtered target encoding, which reduces subjectivity and bias in assessment. We find that combining self-reported and partner-reported variables enhances predictive power. Through feature importance analyses, we identify factors associated with lower recidivism risk, including decreased substance use and avoiding partner contact, while separation processes correlate with higher reoffending likelihood. These findings advance IPV risk assessment by providing a deeper understanding of risk factors critical for improving treatment effectiveness and addressing disparities in IPV management.

Speaker:
Cerag Oguztuzun, PhD
Case Western Reserve University

Authors:
Cerag Oguztuzun, PhD - Case Western Reserve University; Mehmet Koyuturk, PhD - Case Western Reserve University; Gunnur Karakurt, PhD - Case Western Reserve University;

---