Physician Use of Stigmatizing Language in Patient Medical Records

Neighborhood socioeconomic factors and characteristics correlated with avoidable emergency department visits: A spatial analysis of a Canadian academic hospital

INTRODUCTION

The influence of neighborhood characteristics and socioeconomic status (SES) factors on avoidable emergency department (ED) utilization is not well understood in a universal healthcare system. We examined correlations between these factors and avoidable ED visits at a Canadian academic hospital.

MATERIALS AND METHODS

We conducted a retrospective cohort study using administrative ED data from a hospital in Hamilton, Canada from April 1, 2018 to August 31, 2023, and neighborhood data from the Statistics Canada Census of Population 2021. Avoidable visits were classified using the Emergency Department Avoidability Classification (EDAC), and mapped to neighborhoods using Canadian postal codes. SES was defined primarily based on education attained, household income, employment and housing security. The top 20 postal codes with the highest avoidable ED visits were categorized into quartiles and analyzed for trends using chi-squared tests of spatial association and Spearman rank correlations.

RESULTS

A consistent ordinal trend across quartiles was observed throughout the study period, with quartile 1 representing the lowest avoidable ED visits and quartile 4 the highest. The quartiles were unevenly distributed spatially, though there was a significant association between close proximity to the ED and avoidable visits (X2 = 7.07, p <0.05). The quartile with the highest avoidable ED visits (quartile 4) had the greatest proportion of one-person households (35.5%) and one-parent families (37.8%), and showed statistically significant positive correlations with male sex, living alone and having an indigenous identity. Quartile 4 had the highest rates of individuals not completing high school (18.6%, p < 0.05), unemployment (13.7%), households spending greater than 30% of their income on shelter (26.5%), and households earning less than $30,000 annually (16.6%, compared to 8.7% in quartile 1 with the lowest avoidable ED visits).

DISCUSSION

In a universal healthcare setting, lower SES neighborhoods were correlated with higher rates of avoidable ED visits. Targeted interventions that address social determinants of health disparities in neighborhoods with lower SES could reduce the burden of avoidable ED visits, and promote more equitable healthcare utilization.

Opportunities for General Internal Medicine to Promote Equity in Obesity Care

The number and complexity of obesity treatments has increased rapidly in recent years. This is driven by the approval of new anti-obesity medications (AOMs) that produce larger degrees of weight loss than previously approved AOMs. Unfortunately, access to these highly effective therapies and to integrated team-based obesity care is limited by intra-/interpersonal patient, institutional/practitioner, community, and policy factors. We contextualized these complexities and the impact of patients' social drivers of health (SDOH) by adapting the social ecological model for obesity. Without multi-level intervention, these barriers to care will deepen the existing inequities in obesity prevalence and treatment outcomes among historically underserved communities. As General Internal Medicine (GIM) physicians, we can help our patients navigate the complexities of evidence-based obesity treatments. As care team leaders, GIM physicians are well-positioned to (1) improve education for trainees and practitioners, (2) address healthcare-associated weight stigma, (3) advocate for equity in treatment accessibility, and (4) coordinate interdisciplinary teams around non-traditional models of care focused on upstream (e.g., policy changes, insurance coverage, health system culture change, medical education requirements) and downstream (e.g., evidence-based weight management didactics for trainees, using non-stigmatizing language with patients, developing interdisciplinary weight management clinics) strategies to promote optimal obesity care for all patients.

Qualitative Examination of the Experience of Perceived Injustice Following Disabling Occupational Injury

PURPOSE

The primary objective of this study was to explore individuals' perspectives on the factors, situations or events that contributed to their perceptions of injustice following occupational injury.

MATERIALS AND METHODS

The study sample consisted of 30 participants (18 women, 12 men) who had submitted a time-loss claim for a work-related musculoskeletal injury. Participants with elevated scores on a measure of perceived injustice were interviewed about the factors that contributed to their sense of injustice. A thematic analysis was conducted to identify the broad classes of situations or events that participants experienced as unjust in the weeks following occupational injury.

RESULTS

Three dominant themes emerged from the interviews: (1) Invalidation, (2) Undeserved suffering and (3) Blame. Inductively derived subthemes reflected specific dimensions of post-injury experiences that contributed to participants' sense of injustice.

CONCLUSIONS

Given that suffering and invalidating communication are potentially modifiable factors, there are grounds for optimism that intervention approaches can be developed to prevent or reduce perceptions of injustice in the aftermath of debilitating injury. The development of intervention approaches that are effective in preventing or reducing perceptions of injustice holds promise of contributing to more positive recovery outcomes in individuals who have sustained debilitating work injuries.

Equity in Pediatric Hospital-Based Safety and Quality Improvement

There are well-documented inequities in the quality of care and health outcomes of minoritized youth. Patient safety and quality improvement (QI) work with an equity focus has been identified as an important strategy to remedy these existing inequities. In this article, we will present evidence of inequities in pediatric hospital-based care, describe root causes with a focus on structural racism, highlight existing frameworks for applying equity principles to patient safety and QI, and provide best practices and recommendations on evaluating patient safety and QI data towards advancing equity in pediatric hospital-based care.

Leveraging large language models to foster equity in healthcare

OBJECTIVES

Large language models (LLMs) are poised to change care delivery, but their impact on health equity is unclear. While marginalized populations have been historically excluded from early technology developments, LLMs present an opportunity to change our approach to developing, evaluating, and implementing new technologies. In this perspective, we describe the role of LLMs in supporting health equity.

MATERIALS AND METHODS

We apply the National Institute on Minority Health and Health Disparities (NIMHD) research framework to explore the use of LLMs for health equity.

RESULTS

We present opportunities for how LLMs can improve health equity across individual, family and organizational, community, and population health. We describe emerging concerns including biased data, limited technology diffusion, and privacy. Finally, we highlight recommendations focused on prompt engineering, retrieval augmentation, digital inclusion, transparency, and bias mitigation.

CONCLUSION

The potential of LLMs to support health equity depends on making health equity a focus from the start.

The last word: An analysis of power dynamics in clinical notes documenting against-medical-advice discharges

Against Medical Advice (AMA) discharges pose significant challenges to the healthcare system, straining patient-clinician relationships while contributing to avoidable morbidity and mortality. Furthermore, though these discharges culminate in patients' departure from hospitals, their effects reverberate long after, propagated by clinician notes stored in patients' medical records. These notes capture exceptionally fraught interactions between patients and providers, describing the circumstances surrounding breakdowns in clinical relationships. Additionally, they represent just one side of complex, contentious social interactions, for in describing AMA discharges, clinician notewriters quite literally have the last word. For these reasons, notes documenting AMA discharges provide insight into the ways in which clinicians conceptualize, characterize, and propagate power differentials in the contemporary healthcare system. Here, we present a qualitative thematic analysis of 185 notes documenting AMA discharges from a large urban US medical center, interpreting note dynamics through three sociological models of power analysis: (i) the distributive model of power promulgated by Max Weber, (ii) the collectivist power model characterized by Talcott Parsons and Hannah Arendt, and (iii) structural interpretations of power developed by Michel Foucault. We argue that in documenting AMA discharges, clinicians appear to conceive of their relationship with patients in almost exclusively distributive terms, which in turn contributes to an adversarial dynamic whereby both patients and clinicians ultimately suffer disempowerment. We furthermore argue that by facilitating clinicians' recognition of power's collectivist and structural dimensions, we may help transform breakdowns in patient-clinician relationships into opportunities for collaboration.

The Relationship Between Speech-Language Pathologists' Perceptions of Clinical Terminology and Its Use in Voice Therapy with Adults

OBJECTIVES

The terminology used by speech-language pathologists (SLPs) during voice therapy is important for treatment outcomes because it can impact both patient self-efficacy and adherence. However, little is known about what factors influence the terminology that SLPs choose to use. Understanding this gap is important to ultimately optimize voice therapy outcomes. Therefore, the primary aims of this study were to (1) assess the relationship between reported clinician-perceived positivity and (2) assess the relationship between clinician-perceived positivity and clinical endorsement for use. We hypothesized that clinician-perceived positivity would be one important factor driving how frequently clinicians use or avoid specific terms, and if they think the term should be used by other SLPs in clinical practice.

DESIGN/METHODS

This prospective study was conducted as an online survey of SLPs and SLP clinical fellows who evaluate and treat adult voice patients. The survey presented respondents with a total of 46 voice-related terms and prompted respondents to rate: (1) how frequently they use a specific voice-related term ("frequency of use"); (2) how positive or negative they perceive a specific voice-related term to be ("perceived positivity"); and (3) if they feel a specific voice-related term should versus should not be used in clinical practice ("clinical endorsement"). Inferential statistics were used to examine the relationship between perceived positivity and frequency of use, and perceived positivity and clinical endorsement. Summary statistics were used to describe frequency of use across all terms.

RESULTS

One hundred twelve respondents completed the survey. Clinician-perceived positivity of voice-related terminology was significantly related to its reported self-reported frequency of use (β = 1.946; 95% CI: 1.701-2.191; P < 0.0005) and clinical endorsement of use by others (β = 4.103; 95% CI: 3.750-4.456; P < 0.0005). Both of these relationships exhibited relatively large effect sizes.

CONCLUSIONS

This study was an important first step at identifying factors that influence SLP's use of specific terminology in voice therapy. Specifically, an SLP's perceived positivity of clinical terminology strongly influenced the frequency with which they reported using that term in voice therapy and whether or not they thought that term should be used with patients by other SLPs in voice therapy. Future work should investigate clinician characteristics that might affect terminology use, include more diverse sampling, utilize self-selected terminology or audio recordings of therapy interactions, and assess direct effects of terminology use on patient outcomes.

Disparities in Documentation: Evidence of Race-Based Biases in the Electronic Medical Record

Personal implicit biases may contribute to inequitable health outcomes, but the mechanisms of these effects are unclear at a system level. This study aimed to determine whether stigmatizing subjective terms in electronic medical records (EMR) reflect larger societal racial biases. A cross-sectional study was conducted using natural language processing software of all documentation where one or more predefined stigmatizing words were used between January 1, 2019 and June 30, 2021. EMR from emergency care and inpatient encounters in a metropolitan healthcare system were analyzed, focused on the presence or absence of race-based differences in word usage, either by specific terms or by groupings of negative or positive terms based on the common perceptions of the words. The persistence ("stickiness") of negative and/or positive characterizations in subsequent encounters for an individual was also evaluated. Final analyses included 12,238 encounters for 9135 patients, ranging from newborn to 104 years old. White (68%) vs Black/African American (17%) were the analyzed groups. Several negative terms (e.g., noncompliant, disrespectful, and curse words) were significantly more frequent in encounters with Black/African American patients. In contrast, positive terms (e.g., compliant, polite) were statistically more likely to be in White patients' documentation. Independent of race, negative characterizations were twice as likely to persist compared with positive ones in subsequent encounters. The use of stigmatizing language in documentation mirrors the same race-based inequities seen in medical outcomes and larger sociodemographic trends. This may contribute to observed healthcare outcome differences by disseminating one's implicit biases to unknown future healthcare providers.

A comparison of written case notes and the delivery of care in dementia specialist mental health wards

Introduction: Stigmatising language concerning people living with dementia can cause potentially harmful and dehumanising consequences. Language used about people living with dementia in mental health wards may focus on medical perspectives and suggest custodial relationships with patients rather than person-centred accounts of individuals. This language could have a devastating impact on the provision of person-centred care. This study investigated the relationship between accounts of people living with dementia written in healthcare case notes and clinical practice at three dementia specialist wards in Wales, UK. Language guidance was provided to ward staff to assess whether stigmatising language could be reduced and whether this influenced the provision of person-centred care.Methodology: Dementia Care Mapping was adapted to analyse case note entries for enhancing and detracting accounts of people living with dementia at three data collection points. These were compared to the results of routine DCM observations of care across the three wards. The healthcare case notes of 117 people living with dementia, encompassing 4, 522 entries over ten months were analysed. DCM observations of 38 people living with dementia within the three wards were compared against the case note results. Person-centred language guidance was shared with care staff following each data collection point.Results: Following the provision of person-centered language guidance, the use of personally enhancing language was observed to increase across all three wards. Non-person-centred case note entries predominantly focussed on Labelling language, whilst language concerning Invalidation and Objectification also occurred frequently compared to other DCM domains. Person centred language typically concerned Acknowledgement. A relationship between case note entries and practice was evident in some domains although findings were inconsistent.Discussion and Implications: The findings highlight the importance of addressing stigmatising language in healthcare and suggest that further studies to support the anti-stigma agenda in dementia care are required.

"Let Us Take Care of the Medicine": A Qualitative Analysis of Physician Communication When Caring for Febrile Infants

BACKGROUND

Guidelines for the management of febrile infants emphasize patient-centered communication. Although patient-centeredness is central to high-quality health care, biases may impact physicians' patient-centeredness. We aimed to 1) identify physicians' assumptions that inform their communication with parents of febrile infants and 2) examine physicians' perceptions of bias.

METHODS

We recruited physicians from 3 academic pediatric emergency departments (EDs) for semistructured interviews. We applied a constant comparative method approach to conduct a thematic analysis of interview transcripts. Two coders followed several analytical steps: 1) discovery of concepts and code assignment, 2) identification of themes by grouping concepts, 3) axial coding to identify thematic properties, and 4) identifying exemplar excerpts for rich description. Thematic saturation was based on repetition, recurrence, and forcefulness.

RESULTS

Fourteen physicians participated. Participants described making assumptions regarding 3 areas: 1) the parent's affect, 2) the parent's social capacity, and 3) the physician's own role in the parent-physician interaction. Thematic properties highlighted the importance of the physician's assumptions in guiding communication and decision-making. Participants acknowledged an awareness of bias and specifically noted that language bias influenced the assumptions that informed their communication.

CONCLUSIONS

ED physicians described subjective assumptions about parents that informed their approach to communication when caring for febrile infants. Given the emphasis on patient-centered communication in febrile infant guidelines, future efforts are necessary to understand how assumptions are influenced by biases, the effect of such behaviors on health inequities, and how to combat this.

Racial Differences in Stigmatizing and Positive Language in Emergency Medicine Notes

OBJECTIVE

Language used by providers in medical documentation may reveal evidence of race-related implicit bias. We aimed to use natural language processing (NLP) to examine if prevalence of stigmatizing language in emergency medicine (EM) encounter notes differs across patient race/ethnicity.

METHODS

In a retrospective cohort of EM encounters, NLP techniques identified stigmatizing and positive themes. Logistic regression models analyzed the association of race/ethnicity and themes within notes. Outcomes were the presence (or absence) of 7 different themes: 5 stigmatizing (difficult, non-compliant, skepticism, substance abuse/seeking, and financial difficulty) and 2 positive (compliment and compliant).

RESULTS

The sample included notes from 26,363 unique patients. NH Black patient notes were less likely to contain difficult (odds ratio (OR) 0.80, 95% confidence interval (CI), 0.73-0.88), skepticism (OR 0.87, 95% CI, 0.79-0.96), and substance abuse/seeking (OR 0.62, 95% CI, 0.56-0.70) compared to NH White patient notes but more likely to contain non-compliant (OR 1.26, 95% CI, 1.17-1.36) and financial difficulty (OR 1.14, 95% CI, 1.04-1.25). Hispanic patient notes were less likely to contain difficult (OR 0.68, 95% CI, 0.58-0.80) and substance abuse/seeking (OR 0.78, 95% CI, 0.66-0.93). NH NA/AI patient notes had twice the odds as NH White patient notes to contain a stigmatizing theme (OR 2.02, 95% CI, 1.64-2.49).

CONCLUSIONS

Using an NLP model to analyze themes in EM notes across racial groups, we identified several inequities in the usage of positive and stigmatizing language. Interventions to minimize race-related implicit bias should be undertaken.

Promoting diversity, equity and inclusion awareness in clinical documentation through postgraduate year one training

BACKGROUND AND PURPOSE

As healthcare providers increasingly focus on emerging issues of diversity, equity and inclusion (DEI) in patient care, less is known about the training in postgraduate year one (PGY1) pharmacy residency on DEI clinical documentation considerations. This pilot project explored whether training, discussion and self-reflection within a peer review activity promoted DEI self-awareness in clinical documentation through a centralized curriculum of a multisite PGY1.

EDUCATIONAL ACTIVITY AND SETTING

Building upon an established peer review of clinical documentation activity, PGY1 pharmacy residents practicing in ambulatory care settings received training on DEI considerations and completed small and large group discussions, a post-activity mixed methods survey with self-reflection prompts, and a three-month follow-up survey.

FINDINGS

Twenty-two residents participated in the peer review of clinical documentation activity, DEI training and discussions. Twelve residents completed the post-activity survey with reflection prompts; 6 (50%) reported similar previous DEI training prior to residency. After the DEI training and discussions, 12 (100%) agreed or strongly agreed that their awareness of DEI documentation considerations increased; 10 (83%) would document their submitted notes differently, while one resident was unsure and one would not make changes. Twelve residents completed the follow-up survey three months following the activity. Themes from the free-text responses on key learnings collected post-activity and three-month post (respectively) included: 1) new knowledge, increased self-awareness, and intended action and 2) increased self-awareness and changes in note-making convention.

SUMMARY

Integrating DEI training, discussion, and self-reflection prompts into a peer review clinical documentation activity increased self-awareness and knowledge of DEI considerations and promoted intended changes in patient care documentation for pharmacy residents. Regardless of previous training, residents reported continued self-awareness and changes in documentation conventions continued three months later.

Identifying stigmatizing language in clinical documentation: A scoping review of emerging literature

BACKGROUND

Racism and implicit bias underlie disparities in health care access, treatment, and outcomes. An emerging area of study in examining health disparities is the use of stigmatizing language in the electronic health record (EHR).

OBJECTIVES

We sought to summarize the existing literature related to stigmatizing language documented in the EHR. To this end, we conducted a scoping review to identify, describe, and evaluate the current body of literature related to stigmatizing language and clinician notes.

METHODS

We searched PubMed, Cumulative Index of Nursing and Allied Health Literature (CINAHL), and Embase databases in May 2022, and also conducted a hand search of IEEE to identify studies investigating stigmatizing language in clinical documentation. We included all studies published through April 2022. The results for each search were uploaded into EndNote X9 software, de-duplicated using the Bramer method, and then exported to Covidence software for title and abstract screening.

RESULTS

Studies (N = 9) used cross-sectional (n = 3), qualitative (n = 3), mixed methods (n = 2), and retrospective cohort (n = 1) designs. Stigmatizing language was defined via content analysis of clinical documentation (n = 4), literature review (n = 2), interviews with clinicians (n = 3) and patients (n = 1), expert panel consultation, and task force guidelines (n = 1). Natural language processing was used in four studies to identify and extract stigmatizing words from clinical notes. All of the studies reviewed concluded that negative clinician attitudes and the use of stigmatizing language in documentation could negatively impact patient perception of care or health outcomes.

DISCUSSION

The current literature indicates that NLP is an emerging approach to identifying stigmatizing language documented in the EHR. NLP-based solutions can be developed and integrated into routine documentation systems to screen for stigmatizing language and alert clinicians or their supervisors. Potential interventions resulting from this research could generate awareness about how implicit biases affect communication patterns and work to achieve equitable health care for diverse populations.

'Against Medical Advice' Discharges After Respiratory-Related Hospitalizations: Strategies for Respectful Care

Against medical advice (AMA) discharges are practically and emotionally challenging for both patients and clinicians. Moreover, they are common after admissions for respiratory conditions such as COPD and asthma, and they are associated with poor outcomes. Despite the challenges presented by AMA discharges, clinicians rarely receive formal education and have limited guidance on how to approach these discharges. Often, the approach to AMA discharges prioritizes designating the discharge as "AMA," whereas effective coordination of discharge care receives less attention. Such an approach can lead to stigmatization of patients and low-quality care. Although evidence for best practices in AMA discharges remains lacking, we propose a set of strategies to improve care in AMA discharges by focusing on respect, in which clinicians treat patients as equals and honor differing values. We describe five strategies, including (1) preventing an AMA discharge; (2) conducting a patient-centered and truthful discussion of risk; (3) providing harm-reducing discharge care; (4) minimizing stigma and bias; (5) educating trainees. Through a case of a patient discharging AMA after a COPD exacerbation, we highlight how these strategies can be applied to common issues in respiratory-related hospitalizations, such as prescribing inhalers and managing oxygen requirements. We argue that, by using these strategies, clinicians can deliver respectful and higher-quality care to an often-marginalized population of patients with respiratory disease.

Examining Inclusive Language in Clinical Narratives in Medical Biochemistry Textbooks to Model Equitable Patient-Centered Care in Preclinical Undergraduate Medical Education

Purpose

When healthcare professionals use biased or stigmatizing language to describe people or conditions, it can impact the quality of care or erode the patient-physician relationship. It is not clear where healthcare professionals acquire biased and stigmatizing language in practice. This study focuses on examining language in educational materials used in training of medical students. Specifically, medical biochemistry textbooks were examined as they are often a first exposure to clinical narratives and communication standards. The aim of this project is to investigate whether medical biochemistry textbooks, widely recommended in preclinical UME, model inclusive language communication in clinical narratives.

Methods

To determine if educational materials follow inclusive writing guidelines, we conducted a modified document analysis on a sample of medical biochemistry textbooks when clinical scenarios were described. Three independent researchers separately reviewed the textbooks, coded the language using NVivo, and generated themes.

Results

Our results show that medical biochemistry textbooks contain language which is not in alignment with the best practices for inclusive language. Our analysis mapped codes to two primary themes of language misalignment. The first theme, "clinical language" (n = 92), included the following codes: difficult patient, general negative descriptive language, patient as failure, and questioning patient credibility. The second primary theme, "identity-first labeling" (n = 251), included 21 codes.

Conclusion

This study provides early evidence that the language used in medical biochemistry textbooks to describe people and conditions is not in alignment with inclusive language recommendations. This can reinforce the way future healthcare professionals speak to and about their patients.

Measuring Implicit Bias in ICU Notes Using Word-Embedding Neural Network Models

BACKGROUND

Language in nonmedical data sets is known to transmit human-like biases when used in natural language processing (NLP) algorithms that can reinforce disparities. It is unclear if NLP algorithms of medical notes could lead to similar transmissions of biases.

RESEARCH QUESTION

Can we identify implicit bias in clinical notes, and are biases stable across time and geography?

STUDY DESIGN AND METHODS

To determine whether different racial and ethnic descriptors are similar contextually to stigmatizing language in ICU notes and whether these relationships are stable across time and geography, we identified notes on critically ill adults admitted to the University of California, San Francisco (UCSF), from 2012 through 2022 and to Beth Israel Deaconess Hospital (BIDMC) from 2001 through 2012. Because word meaning is derived largely from context, we trained unsupervised word-embedding algorithms to measure the similarity (cosine similarity) quantitatively of the context between a racial or ethnic descriptor (eg, African-American) and a stigmatizing target word (eg, nonco-operative) or group of words (violence, passivity, noncompliance, nonadherence).

RESULTS

In UCSF notes, Black descriptors were less likely to be similar contextually to violent words compared with White descriptors. Contrastingly, in BIDMC notes, Black descriptors were more likely to be similar contextually to violent words compared with White descriptors. The UCSF data set also showed that Black descriptors were more similar contextually to passivity and noncompliance words compared with Latinx descriptors.

INTERPRETATION

Implicit bias is identifiable in ICU notes. Racial and ethnic group descriptors carry different contextual relationships to stigmatizing words, depending on when and where notes were written. Because NLP models seem able to transmit implicit bias from training data, use of NLP algorithms in clinical prediction could reinforce disparities. Active debiasing strategies may be necessary to achieve algorithmic fairness when using language models in clinical research.

Stigmatizing Language, Patient Demographics, and Errors in the Diagnostic Process

This cohort study assesses the association between stigmatizing language, demographic characteristics, and errors in the diagnostic process among hospitalized adults.

Examining Linguistic Differences in Electronic Health Records for Diverse Patients With Diabetes: Natural Language Processing Analysis

Background

Individuals from minoritized racial and ethnic backgrounds experience pernicious and pervasive health disparities that have emerged, in part, from clinician bias.

Objective

We used a natural language processing approach to examine whether linguistic markers in electronic health record (EHR) notes differ based on the race and ethnicity of the patient. To validate this methodological approach, we also assessed the extent to which clinicians perceive linguistic markers to be indicative of bias.

Methods

In this cross-sectional study, we extracted EHR notes for patients who were aged 18 years or older; had more than 5 years of diabetes diagnosis codes; and received care between 2006 and 2014 from family physicians, general internists, or endocrinologists practicing in an urban, academic network of clinics. The race and ethnicity of patients were defined as White non-Hispanic, Black non-Hispanic, or Hispanic or Latino. We hypothesized that Sentiment Analysis and Social Cognition Engine (SEANCE) components (ie, negative adjectives, positive adjectives, joy words, fear and disgust words, politics words, respect words, trust verbs, and well-being words) and mean word count would be indicators of bias if racial differences emerged. We performed linear mixed effects analyses to examine the relationship between the outcomes of interest (the SEANCE components and word count) and patient race and ethnicity, controlling for patient age. To validate this approach, we asked clinicians to indicate the extent to which they thought variation in the use of SEANCE language domains for different racial and ethnic groups was reflective of bias in EHR notes.

Results

We examined EHR notes (n=12,905) of Black non-Hispanic, White non-Hispanic, and Hispanic or Latino patients (n=1562), who were seen by 281 physicians. A total of 27 clinicians participated in the validation study. In terms of bias, participants rated negative adjectives as 8.63 (SD 2.06), fear and disgust words as 8.11 (SD 2.15), and positive adjectives as 7.93 (SD 2.46) on a scale of 1 to 10, with 10 being extremely indicative of bias. Notes for Black non-Hispanic patients contained significantly more negative adjectives (coefficient 0.07, SE 0.02) and significantly more fear and disgust words (coefficient 0.007, SE 0.002) than those for White non-Hispanic patients. The notes for Hispanic or Latino patients included significantly fewer positive adjectives (coefficient -0.02, SE 0.007), trust verbs (coefficient -0.009, SE 0.004), and joy words (coefficient -0.03, SE 0.01) than those for White non-Hispanic patients.

Conclusions

This approach may enable physicians and researchers to identify and mitigate bias in medical interactions, with the goal of reducing health disparities stemming from bias.

The Sociodemographic Biases in Machine Learning Algorithms: A Biomedical Informatics Perspective

Artificial intelligence models represented in machine learning algorithms are promising tools for risk assessment used to guide clinical and other health care decisions. Machine learning algorithms, however, may house biases that propagate stereotypes, inequities, and discrimination that contribute to socioeconomic health care disparities. The biases include those related to some sociodemographic characteristics such as race, ethnicity, gender, age, insurance, and socioeconomic status from the use of erroneous electronic health record data. Additionally, there is concern that training data and algorithmic biases in large language models pose potential drawbacks. These biases affect the lives and livelihoods of a significant percentage of the population in the United States and globally. The social and economic consequences of the associated backlash cannot be underestimated. Here, we outline some of the sociodemographic, training data, and algorithmic biases that undermine sound health care risk assessment and medical decision-making that should be addressed in the health care system. We present a perspective and overview of these biases by gender, race, ethnicity, age, historically marginalized communities, algorithmic bias, biased evaluations, implicit bias, selection/sampling bias, socioeconomic status biases, biased data distributions, cultural biases and insurance status bias, conformation bias, information bias and anchoring biases and make recommendations to improve large language model training data, including de-biasing techniques such as counterfactual role-reversed sentences during knowledge distillation, fine-tuning, prefix attachment at training time, the use of toxicity classifiers, retrieval augmented generation and algorithmic modification to mitigate the biases moving forward.

Reflections on 'common' genetic medical history questions: Time to examine the what, why, and how

OBJECTIVE

A central goal of patient-centered care is to establish a therapeutic relationship. While remaining in tune with patient emotions, genetics providers must ask questions to understand medical histories that will inform the differential diagnosis, evaluation plan, and potential treatments.

METHODS

195 audio-recorded conversations between providers and caregivers of pediatric patients with suspected genetic conditions were coded and analyzed. Coders identified sensitive history-taking questions asked by providers related to exposures and complications during pregnancy; ancestry and consanguinity; educational attainment of the caregiver; and family structure.

RESULTS

We highlight examples of providers: using stigmatizing language about conception or consanguinity; not clarifying the intent behind questions related to caregivers' educational attainment and work history; and making presumptions or assumptions about caregivers' race and ethnicity, family structure, and exposures during pregnancy.

CONCLUSION

Some questions and phrasing considered routine by genetics providers may interfere with patient-centered care by straining attempts to establish a therapeutic, trusting relationship. Additional research is needed to assess how question asking and phrasing impact rapport building and patient experience during genetics encounters.

PRACTICE IMPLICATIONS

Review of the purpose and need for medical history questions common to genetics practice could serve to improve patient-centered care.

A Content Analysis of HIV-Related Stigmatizing Language in the Scientific Literature, From 2010-2020: Findings and Recommendations for Editorial Policy

Despite negative effects of HIV-related stigma on people with HIV, some scientific literature continues to use stigmatizing terms. Our study aimed to explore the use of HIV-related stigmatizing language in the scientific literature between 2010 and 2020 based on 2015 UNAIDS terminology guidelines. We searched for articles with the stigmatizing term "HIV/AIDS-infected" or any variations that were peer-reviewed, published between 2010 and 2020, and in English or with an English translation. Our search yielded 26,476 articles that used the stigmatizing term of interest. Frequencies on the variables of interest (journal, year, and country) were run. The use of these terms increased from 2010 to 2017 and decreased from 2018 to 2020. Most journals using the terms were HIV/AIDS specific or on infectious diseases, but the journal with the greatest frequency of use was on general science and medicine. Thirty-six percent of the articles emanated from the United States. To reduce the use of stigmatizing language in the HIV literature, action should be taken by authors, reviewers, editors,educators, and publishers should create formal policies promoting use of non-stigmatizing language.

Documentation of Trauma-Informed Care Elements for Young People Hospitalized After Assault Trauma

INTRODUCTION

Violent traumatic injury, including firearm violence, can adversely impact individual and community health. Trauma-informed care (TIC) can promote resilience and prevent future violence in patients who have experienced trauma. However, few protocols exist to facilitate implementation of TIC for patients who survive traumatic injury. The purpose of the study is to characterize documentation of TIC practices and identify opportunities for intervention in a single academic quaternary care center.

METHODS

This study is a retrospective chart review analyzing the documentation of trauma-informed elements in the electronic medical record of a random sample of youth patients (ages 12-23) admitted for assault trauma to the pediatric (n = 50) and adult trauma (n = 200) services between 2016 and mid-2021. Descriptive statistics were used to summarize patient demographics, hospitalization characteristics, and documentation of trauma-informed elements. Chi-square analyses were performed to compare pediatric and adult trauma services.

RESULTS

Among pediatric and adult assault trauma patients, 36.0% and 80.5% were hospitalized for firearm injury, respectively. More patients admitted to the pediatric trauma service (96%) had at least one trauma-informed element documented than patients admitted to the adult service (82.5%). Social workers were the most likely clinicians to document a trauma-informed element. Pain assessment and social support were most frequently documented. Safety assessments for suicidal ideation, retaliatory violence, and access to a firearm were rarely documented.

CONCLUSIONS

Results highlight opportunities to develop trauma-informed interventions for youth admitted for assault trauma. Standardized TIC documentation could be used to assess risk of violent reinjury and mitigate sequelae of trauma.

Mental health stigma and its relationship with mental health professionals - A narrative review and practice implications

The extent and magnitude of the mental health stigma are enormous, with substantial clinical and social implications. There is a complex relationship between mental health stigma and mental health professionals (MHPs); MHPs can be anti-stigma crusaders, victims of stigma, and even a source of stigma. Unfortunately, literature is scarce talking about the relationship between stigma and MHPs. Hence, the current review aims to bridge the existing gap in the literature on various aspects of stigma and the role of MHPs. For the current review, we ran a search in PubMed and Google Scholar databases; we restricted our study to records focusing on the interplay of mental health stigma and the MHPs, published during 2012-2022, in English, and having a full text available. We found that MHPs (psychiatrists, psychologists, and psychiatric nurses) can also be the recipients of the stigma. The stigma faced by the MHPs is determined by the negative stereotypes set by the media, or medical students, or other health professionals; the marginal position of psychiatry in the health system; difficult-to-treat mental disorders; MHPs' own experience of stigma; and the attitude or beliefs of various caders of the MHPs, their professional experience, and expertise in managing various mental health conditions. Notably, MHPs can also be a source of stigma (stigmatizers). MHPs need to be sensitized concerning this, and the anti-stigma interventions must incorporate this aspect of stigma. Novel interventions, such as digital-based programs, should be used instead of traditional anti-stigma programs in order to decrease stigma around mental health issues and make anti-stigma initiatives more appealing and scalable. To address the issues of stigma, there has to be more communication between MHPs, other health professionals, service users, and policymakers.

Using Natural Language Processing to Identify Stigmatizing Language in Labor and Birth Clinical Notes

INTRODUCTION

Stigma and bias related to race and other minoritized statuses may underlie disparities in pregnancy and birth outcomes. One emerging method to identify bias is the study of stigmatizing language in the electronic health record. The objective of our study was to develop automated natural language processing (NLP) methods to identify two types of stigmatizing language: marginalizing language and its complement, power/privilege language, accurately and automatically in labor and birth notes.

METHODS

We analyzed notes for all birthing people > 20 weeks' gestation admitted for labor and birth at two hospitals during 2017. We then employed text preprocessing techniques, specifically using TF-IDF values as inputs, and tested machine learning classification algorithms to identify stigmatizing and power/privilege language in clinical notes. The algorithms assessed included Decision Trees, Random Forest, and Support Vector Machines. Additionally, we applied a feature importance evaluation method (InfoGain) to discern words that are highly correlated with these language categories.

RESULTS

For marginalizing language, Decision Trees yielded the best classification with an F-score of 0.73. For power/privilege language, Support Vector Machines performed optimally, achieving an F-score of 0.91. These results demonstrate the effectiveness of the selected machine learning methods in classifying language categories in clinical notes.

CONCLUSION

We identified well-performing machine learning methods to automatically detect stigmatizing language in clinical notes. To our knowledge, this is the first study to use NLP performance metrics to evaluate the performance of machine learning methods in discerning stigmatizing language. Future studies should delve deeper into refining and evaluating NLP methods, incorporating the latest algorithms rooted in deep learning.

Pain Catastrophizing: Controversies, Misconceptions and Future Directions

Recent reports have pointed to problems with the term "pain catastrophizing." Critiques of the term pain catastrophizing have come from several sources including individuals with chronic pain, advocates for individuals with chronic pain, and pain scholars. Reports indicate that the term has been used to dismiss the medical basis of pain complaints, to question the authenticity of pain complaints, and to blame individuals with pain for their pain condition. In this paper, we advance the position that the problems prompting calls to rename the construct of pain catastrophizing have little to do with the term, and as such, changing the term will do little to solve these problems. We argue that continued calls for changing or deleting the term pain catastrophizing will only divert attention away from some fundamental flaws in how individuals with pain conditions are assessed and treated. Some of these fundamental flaws have their roots in the inadequate training of health and allied health professionals in evidence-based models of pain, in the use of psychological assessment and intervention tools for the clinical management of pain, and in gender equity and antiracism. Critiques that pain scholars have leveled against the defining, operational, and conceptual bases of pain catastrophizing are also addressed. Arguments for reconceptualizing pain catastrophizing as a worry-related construct are discussed. Recommendations are made for remediation of the problems that have contributed to calls to rename the term pain catastrophizing. PERSPECTIVE: The issues prompting calls to rename the construct of pain catastrophizing have their roots in fundamental flaws in how individuals with pain are assessed and treated. Efforts to address these problems will require more than a simple change in terminology.

Twelve tips for integrating medical students into specialty clinics

The specialty clinic is an excellent educational environment for medical students. However, preceptors face several challenges as they seek to balance treating complex system-specific conditions with effective teaching, including time constraints, clinical tasks, engaging multi-level learners, and perhaps a lack of guidelines for or training in outpatient medical education. We thus propose twelve tips for integrating medical students into specialty clinics in a feasible and mutually fulfilling way. The first three tips focus on planning the session and setting expectations, the next seven tips detail specific, actionable strategies for enhancing learning while maximizing efficiency, and the final two tips discuss how to optimally close the session with feedback and debriefing.

Empirical data drift detection experiments on real-world medical imaging data

While it is common to monitor deployed clinical artificial intelligence (AI) models for performance degradation, it is less common for the input data to be monitored for data drift - systemic changes to input distributions. However, when real-time evaluation may not be practical (eg., labeling costs) or when gold-labels are automatically generated, we argue that tracking data drift becomes a vital addition for AI deployments. In this work, we perform empirical experiments on real-world medical imaging to evaluate three data drift detection methods' ability to detect data drift caused (a) naturally (emergence of COVID-19 in X-rays) and (b) synthetically. We find that monitoring performance alone is not a good proxy for detecting data drift and that drift-detection heavily depends on sample size and patient features. Our work discusses the need and utility of data drift detection in various scenarios and highlights gaps in knowledge for the practical application of existing methods.

Metaphorical use of "headache" and "migraine" in media: A longitudinal study of 1.3 million articles in major publications

BACKGROUND

Stigmatization and trivialization of headache confront individuals with headache disorders, but the degree to which media may contribute is incompletely understood.

OBJECTIVE

The objective of this study was to quantify the frequency of disparaging metaphorical use of the words "headache" and "migraine" in articles and summaries of major publications.

METHODS

This longitudinal study analyzed a dataset of 1.3 million articles and summaries written by authors and editors of 38 major publications. Data cover written publications from 1998 up to 2017. The use of the words "headache" or "migraine" in articles and summaries by major publications was rated by two authors (P.Z. and A.V.) as either "metaphorical" or "medical" based on their contextual application. Pearson's chi-squared test was applied to assess differences in the frequency of metaphorical use of "headache" in comparison to "migraine." Secondary outcomes were the source of publication and time of publication.

RESULTS

A total of 6195 and 740 articles included the words "headache" or "migraine," respectively; 7100 sentences contained the word "headache" and 1652 sentences contained the word "migraine." Among a random sample of 1000 sentences with the word "headache," there was a metaphorical use in 492 (49.2% [95% CI, 46.1-52.3]) sentences. Among a random sample of 1000 sentences with the word "migraine," there was a metaphorical use in 45 (4.5% [95% CI, 3.2-5.8]) sentences. The five most prevalent sources were CNN, Fox News, The New York Times, The Guardian, and The Washington Post. There was an overall increase in the number of articles containing the words "headache" or "migraine" from database inception until analysis (1998 up to 2017). The database included no articles containing either "headache" or "migraine" in 1998; in 2016, this number was 1480 articles.

CONCLUSIONS

In this longitudinal study, major publications applied a metaphorical use of "headache" about half of the time. The metaphorical use of "headache" is 11-fold greater than the metaphorical use of "migraine" in the same media sample. These depictions may contribute to the trivialization of headache and the stigmatization of individuals with headache disorders. Studies with individuals affected by headache disorders are needed to clarify potential influences.

Caregivers Attitude Detection From Clinical Notes

Caregivers' attitudes impact healthcare quality and disparities. Clinical notes contain highly specialized and ambiguous language that requires extensive domain knowledge to understand, and using negative language does not necessarily imply a negative attitude. This study discusses the challenge of detecting caregivers' attitudes from their clinical notes. To address these challenges, we annotate MIMIC clinical notes and train state-of-the-art language models from the Hugging Face platform. The study focuses on the Neonatal Intensive Care Unit and evaluates models in zero-shot, few-shot, and fully-trained scenarios. Among the chosen models, RoBERTa identifies caregivers' attitudes from clinical notes with an F1-score of 0.75. This approach not only enhances patient satisfaction, but opens up exciting possibilities for detecting and preventing care provider syndromes, such as fatigue, stress, and burnout. The paper concludes by discussing limitations and potential future work.

Patient Reasoning: Patients' and Care Partners' Perceptions of Diagnostic Accuracy in Emergency Care

OBJECTIVES

In the context of validating a measure of patient report specific to diagnostic accuracy in emergency department or urgent care, this study investigates patients' and care partners' perceptions of diagnoses as accurate and explores variations in how they reason while they assess accuracy.

METHODS

In February 2022, we surveyed a national panel of adults who had an emergency department or urgent care visit in the past month to test a patient-reported measure. As part of the survey validation, we asked for free-text responses about why the respondents indicated their (dis)agreement with 2 statements comprising patient-reported diagnostic accuracy: 1) the explanation they received of the health problem was true and 2) the explanation described what to expect of the health problem. Those paired free-text responses were qualitatively analyzed according to themes created inductively.

RESULTS

A total of 1,116 patients and care partners provided 982 responses coded into 10 themes, which were further grouped into 3 reasoning types. Almost one-third (32%) of respondents used only corroborative reasoning in assessing the accuracy of the health problem explanation (alignment of the explanation with either test results, patients' subsequent health trajectory, their medical knowledge, symptoms, or another doctor's opinion), 26% used only perception-based reasoning (perceptions of diagnostic process, uncertainty around the explanation received, or clinical team's attitudes), and 27% used both types of reasoning. The remaining 15% used general beliefs or nonexplicated logic (used only about accurate diagnoses) and combinations of general reasoning with perception-based and corroborative.

CONCLUSIONS

Patients and care partners used multifaceted reasoning in their assessment of diagnostic accuracy.

IMPLICATIONS

As health care shifts toward meaningful diagnostic co-production and shared decision making, in-depth understanding of variations in patient reasoning and mental models informs use in clinical practice.

HIGHLIGHTS

An analysis of 982 responses examined how patients and care partners reason about the accuracy of diagnoses they received in emergency or urgent care.In reasoning, people used their perception of the process and whether the diagnosis matched other factual information they have.We introduce "patient reasoning" in the diagnostic measurement context as an area of further research to inform diagnostic shared decision making and co-production of health.

Natural Language Processing Applied to Clinical Documentation in Post-acute Care Settings: A Scoping Review

OBJECTIVES

To determine the scope of the application of natural language processing to free-text clinical notes in post-acute care and provide a foundation for future natural language processing-based research in these settings.

DESIGN

Scoping review; reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines.

SETTING AND PARTICIPANTS

Post-acute care (ie, home health care, long-term care, skilled nursing facilities, and inpatient rehabilitation facilities).

METHODS

PubMed, Cumulative Index of Nursing and Allied Health Literature, and Embase were searched in February 2023. Eligible studies had quantitative designs that used natural language processing applied to clinical documentation in post-acute care settings. The quality of each study was appraised.

RESULTS

Twenty-one studies were included. Almost all studies were conducted in home health care settings. Most studies extracted data from electronic health records to examine the risk for negative outcomes, including acute care utilization, medication errors, and suicide mortality. About half of the studies did not report age, sex, race, or ethnicity data or use standardized terminologies. Only 8 studies included variables from socio-behavioral domains. Most studies fulfilled all quality appraisal indicators.

CONCLUSIONS AND IMPLICATIONS

The application of natural language processing is nascent in post-acute care settings. Future research should apply natural language processing using standardized terminologies to leverage free-text clinical notes in post-acute care to promote timely, comprehensive, and equitable care. Natural language processing could be integrated with predictive models to help identify patients who are at risk of negative outcomes. Future research should incorporate socio-behavioral determinants and diverse samples to improve health equity in informatics tools.

The Social Context: Social and Behavioral Factors That Affect Health Outcomes

To achieve understanding and best care, screening and treating patients should consider the patient's social environment. Social and behavioral factors influence both positive and negative health behaviors that influence mental and physical health. Primary care providers continually navigate barriers faced by patients and seek solutions that take into consideration social and behavioral factors. The role of the PCP begins with an understanding of common barriers and community resources, then by assessing and responding to the patient's own challenges, and finally by advocating in the clinic and public for changes to the underlying social and structural causes of morbidity and mortality.

Cute Kid? Patient Obesity Status and the Use of Nonmedical Descriptors in Presentations by Pediatric Residents

Nonmedical descriptors, adjectives that are not related to a medical condition, such as "cute," are often used in presentations in pediatrics. We hypothesize that patterns of their use may reflect obesity bias. Descriptors used by pediatric residents presenting cases of children <9 years in an outpatient clinic during the 2018-2019 and 2019-2020 academic years were recorded. The primary outcome was the association of the use of positive nonmedical descriptors with children's obesity status using logistic regression. Positive descriptors were used in 14% of 994 presentations. Most addressed the appearance of the child with variations of "cute" and "adorable." There was no variation in use of positive descriptors by obesity status. On multivariate logistic regression, the odds of using positive descriptors were higher among female residents, and positive descriptor use declined with patient age. Negative descriptors were rare and often focused on weight.

ChatGPT in Answering Queries Related to Lifestyle-Related Diseases and Disorders

Background Lifestyle-related diseases and disorders have become a significant global health burden. However, the majority of the population ignores or do not consult doctors for such disease or disorders. Artificial intelligence (AI)-based large language model (LLM) like ChatGPT (GPT3.5) is capable of generating customized queries of a user. Hence, it can act as a virtual telehealth agent. Its capability to answer lifestyle-related diseases or disorders has not been explored. Objective This study aimed to evaluate the effectiveness of ChatGPT, an LLM, in providing answers to queries related to lifestyle-related diseases or disorders. Methods A set of 20 lifestyle-related disease or disorder cases covering a wide range of topics such as obesity, diabetes, cardiovascular health, and mental health were prepared with four questions. The case and questions were presented to ChatGPT and asked for the answers to those questions. Two physicians rated the content on a three-point Likert-like scale ranging from accurate (2), partially accurate (1), and inaccurate (0). Further, the content was rated as adequate (2), inadequate (1), and misguiding (0) for testing the applicability of the guides for patients. The readability of the text was analyzed by the Flesch-Kincaid Ease Score (FKES).  Results Among 20 cases, the average score of accuracy was 1.83±0.37 and guidance was 1.9±0.21. Both the scores were higher than the hypothetical median of 1.5 (p=0.004 and p<0.0001, respectively). ChatGPT answered the questions with a natural tone in 11 cases and nine with a positive tone. The text was understandable for college graduates with a mean FKES of 27.8±5.74. Conclusion The analysis of content accuracy revealed that ChatGPT provided reasonably accurate information in the majority of the cases, successfully addressing queries related to lifestyle-related diseases or disorders. Hence, initial guidance can be obtained by patients when they get little time to consult a doctor or wait for an appointment to consult a doctor for suggestions about their condition.

Creating a tool to understand university students' experiences regarding inappropriate, disrespectful, and coercive (IDC) healthcare interactions

Objective: The purpose of this study was to develop a survey tool to capture inappropriate, disrespectful, and coercive (IDC) interactions with healthcare providers among a diverse sample of university students. Participants: Participants were university students at one large Midwestern public university. Methods: An exploratory qualitative approach was used to create a survey tool to capture IDC interactions. Results: In Phase I, 9 focus group discussions (FGDs) and 3 individual interviews were conducted with a total of 38 participants. In Phase II, 18 participants completed cognitive interviews. Themes across all FGDs included: (1) communication; (2) respect for identity; (3) institutional practices; (4) power imbalances; and (5) lack of patient education and empowerment. Queer participants discussed unique considerations of how queer identity influences one's IDC healthcare experiences. Conclusions: This study resulted in the development of a 64-70 item tool, the IDC Survey, to measure the prevalence and characteristics of IDC healthcare interactions.

An Intervention-Based Approach to Achieve Racial Equity in Gynecologic Oncology

Racial inequities within gynecologic oncology exist at every step of the cancer continuum. Although the disparities have been well described, there is a significant gap in the literature focused on eliminating inequities in gynecologic cancer outcomes. The goal of this narrative review is to highlight successful, evidence-based interventions from within and outside of gynecologic oncology that alleviate disparity, providing a call to action for further research and implementation efforts within the field. These solutions are organized in the socioecologic framework, where multiple levels of influence-societal, community, organizational, interpersonal, and individual-affect health outcomes.

The Power of Language in Hospital Care for Pregnant and Birthing People: A Vision for Change

Language is commonly defined as the principal method of human communication made up of words and conveyed by writing, speech, or nonverbal expression. In the context of clinical care, language has power and meaning and reflects priorities, beliefs, values, and culture. Stigmatizing language can communicate unintended meanings that perpetuate socially constructed power dynamics and result in bias. This bias may harm pregnant and birthing people by centering positions of power and privilege and by reflecting cultural priorities in the United States, including judgments of demographic and reproductive health characteristics. This commentary builds on relationship-centered care and reproductive justice frameworks to analyze the role and use of language in pregnancy and birth care in the United States, particularly regarding people with marginalized identities. We describe the use of language in written documentation, verbal communication, and behaviors associated with caring for pregnant people. We also present recommendations for change, including alternative language at the individual, clinician, hospital, health systems, and policy levels. We define birth as the emergence of a new individual from the body of its parent, no matter what intervention or pathology may be involved. Thus, we propose a cultural shift in hospital-based care for birthing people that centers the birthing person and reconceptualizes all births as physiologic events, approached with a spirit of care, partnership, and support.

Patient Factors Associated With Biased Language in Nightly Resident Verbal Handoff

This cross-sectional study uses audio recordings of resident handoff of inpatient general medicine and general pediatrics teams to measure the extent of stigmatizing language and describes associations between patient factors and biased language in handoffs.

Cardiovascular Care Innovation through Data-Driven Discoveries in the Electronic Health Record

The electronic health record (EHR) represents a rich source of patient information, increasingly being leveraged for cardiovascular research. Although its primary use remains the seamless delivery of health care, the various longitudinally aggregated structured and unstructured data elements for each patient within the EHR can define the computational phenotypes of disease and care signatures and their association with outcomes. Although structured data elements, such as demographic characteristics, laboratory measurements, problem lists, and medications, are easily extracted, unstructured data are underused. The latter include free text in clinical narratives, documentation of procedures, and reports of imaging and pathology. Rapid scaling up of data storage and rapid innovation in natural language processing and computer vision can power insights from unstructured data streams. However, despite an array of opportunities for research using the EHR, specific expertise is necessary to adequately address confidentiality, accuracy, completeness, and heterogeneity challenges in EHR-based research. These often require methodological innovation and best practices to design and conduct successful research studies. Our review discusses these challenges and their proposed solutions. In addition, we highlight the ongoing innovations in federated learning in the EHR through a greater focus on common data models and discuss ongoing work that defines such an approach to large-scale, multicenter, federated studies. Such parallel improvements in technology and research methods enable innovative care and optimization of patient outcomes.

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.

Transmission of Bias in the Medical Record Among Physical Medicine and Rehabilitation Trainees

ABSTRACT

Stigmatizing language can negatively influence providers' attitudes and care toward patients, but this has not been studied among physiatrists. An online survey was created to assess whether stigmatizing language can impact physical medicine and rehabilitation trainees' attitudes toward patients. We hypothesized stigmatizing language would negatively impact trainees' attitudes. Participants were randomized to a stigmatizing or neutral language vignette describing the same hypothetical spinal cord injury patient. Questions were asked about attitudes and assumptions toward the patient, pain management based on the vignette, and general views regarding individuals with disabilities. Between August 2021 and January 2022, 75 US physical medicine and rehabilitation residency trainees participated. Thirty-seven (49.3%) identified as women; 52 (69.3%) were White, and half (50.6%) received the stigmatized vignette. Participants exposed to stigmatizing language scored 4.8 points lower ( P < 0.01) on the provider attitude toward patient scale compared with those exposed to neutral language. There were no significant differences in the disability attitude scores between the two groups ( P = 0.81). These findings may indicate that stigmatizing language in the medical record may negatively affect physical medicine and rehabilitation trainees' attitudes toward patients. Further exploration is needed to identify the best way to educate trainees and reduce the propagation of bias in the medical record.