Sexual and Gender Minority Status and Suicide Mortality: An Explainable Artificial Intelligence Analysis.

Objectives: Suicide risk is elevated in lesbian, gay, bisexual, and transgender (LGBT) individuals. Limited data on LGBT status in healthcare systems hinder our understanding of this risk. This study used natural language processing to extract LGBT status and a deep neural network (DNN) to examine suicidal death risk factors among US Veterans. Methods: Data on 8.8 million veterans with visits between 2010 and 2017 was used. A case-control study was performed, and suicide death risk was analyzed by a DNN. Feature impacts and interactions on the outcome were evaluated. Results: The crude suicide mortality rate was higher in LGBT patients. However, after adjusting for over 200 risk and protective factors, known LGBT status was associated with reduced risk compared to LGBT-Unknown status. Among LGBT patients, black, female, married, and older Veterans have a higher risk, while Veterans of various religions have a lower risk. Conclusion: Our results suggest that disclosed LGBT status is not directly associated with an increase suicide death risk, however, other factors (e.g., depression and anxiety caused by stigma) are associated with suicide death risks.

Relationship Between Pain and LGBT Status Among Veterans in Care in a Retrospective Cross-Sectional Cohort.

Background: Pain assessment is performed in many healthcare systems, such as the Veterans Health Administration, but prior studies have not assessed whether pain screening varies in sexual and gender minority populations that include individuals who identify as lesbian, gay, bisexual, and/or transgender (LGBT). Objective: The purpose of this study was to evaluate pain screening and reported pain of LGBT Veterans compared to non-LGBT Veterans. Methods: Using a retrospective cross-sectional cohort, data from the Corporate Data Warehouse, a national repository with clinical/administrative data, were analyzed. Veterans were classified as LGBT using natural language processing. We used a robust Poisson model to examine the association between LGBT status and binary outcomes of pain screening, any pain, and persistent pain within one year of entry in the cohort. All models were adjusted for demographics, mental health, substance use, musculoskeletal disorder(s), and number of clinic visits. Results: There were 1,149,486 Veterans (218,154 (19%) classified as LGBT) in our study. Among LGBT Veterans, 94% were screened for pain compared to 89% among those not classified as LGBT (non-LGBT) Veterans. In adjusted models, LGBT Veterans' probability of being screened for pain compared to non-LGBT Veterans was 2.5% higher (95% CI 2.3%, 2.6%); risk of any pain was 2.1% lower (95% CI 1.6%, 2.6%); and there was no significant difference between LGBT and non-LGBT Veterans in persistent pain (RR = 1.00, 95% CI (0.99, 1.01), p = 0.88). Conclusions: In a nationwide sample, LGBT Veterans were more likely to be screened for pain but had lower self-reported pain scores, though adjusted differences were small. It was notable that transgender and Black Veterans reported the greatest pain. Reasons for these findings require further investigation.

Personal Health Libraries for People Returning From Incarceration: Protocol for a Qualitative Study.

BACKGROUND: Individuals released from carceral facilities have high rates of hospitalization and death, especially in the weeks immediately after their return to community settings. During this transitional process, individuals leaving incarceration are expected to engage with multiple providers working in separate, complex systems, including health care clinics, social service agencies, community-based organizations, and probation and parole services. This navigation is often complicated by individuals' physical and mental health, literacy and fluency, and socioeconomic status. Personal health information technology, which can help people access and organize their health information, could improve the transition from carceral systems to the community and mitigate health risks upon release. Yet, personal health information technologies have not been designed to meet the needs and preferences of this population nor tested for acceptability or use. OBJECTIVE: The objective of our study is to develop a mobile app to create personal health libraries for individuals returning from incarceration to help bridge the transition from carceral settings to community living. METHODS: Participants were recruited through Transitions Clinic Network clinic encounters and professional networking with justice-involved organizations. We used qualitative research methods to assess the facilitators and barriers to developing and using personal health information technology for individuals returning from incarceration. We conducted individual interviews with people just released from carceral facilities (n=~20) and providers (n=~10) from the local community and carceral facilities involved with the transition for returning community members. We used rigorous rapid qualitative analysis to generate thematic output characterizing the unique circumstances impacting the development and use of personal health information technology for individuals returning from incarceration and to identify content and features for the mobile app based on the preferences and needs of our participants. RESULTS: As of February 2023, we have completed 27 qualitative interviews with individuals recently released from carceral systems (n=20) and stakeholders (n=7) who support justice-involved individuals from various organizations in the community. CONCLUSIONS: We anticipate that the study will characterize the experiences of people transitioning from prison and jails to community settings; describe the information, technology resources, and needs upon reentry to the community; and create potential pathways for fostering engagement with personal health information technology. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/44748.

Temporal and Geographic Patterns of Documentation of Sexual Orientation and Gender Identity Keywords in Clinical Notes.

OBJECTIVE: Disclosure of sexual orientation and gender identity correlates with better outcomes, yet data may not be available in structured fields in electronic health record data. To gain greater insight into the care of sexual and gender-diverse patients in the Veterans Health Administration (VHA), we examined the documentation patterns of sexual orientation and gender identity through extraction and analyses of data contained in unstructured electronic health record clinical notes. METHODS: Salient terms were identified through authoritative vocabularies, the research team's expertise, and frequencies, and the use of consistency in VHA clinical notes. Term frequencies were extracted from VHA clinical notes recorded from 2000 to 2018. Temporal analyses assessed usage changes in normalized frequencies as compared with nonclinical use, relative growth rates, and geographic variations. RESULTS: Over time most terms increased in use, similar to Google ngram data, especially after the repeal of the "Don't Ask Don't Tell" military policy in 2010. For most terms, the usage adoption consistency also increased by the study's end. Aggregated use of all terms increased throughout the United States. CONCLUSION: Term usage trends may provide a view of evolving care in a temporal continuum of changing policy. These findings may be useful for policies and interventions geared toward sexual and gender-diverse individuals. Despite the lack of structured data, the documentation of sexual orientation and gender identity terms is increasing in clinical notes.

Documentation of Screening for Firearm Access by Healthcare Providers in the Veterans Healthcare System: A Retrospective Study.

INTRODUCTION: Presence of a firearm is associated with increased risk of violence and suicide. United States military veterans are at disproportionate risk of suicide. Routine healthcare provider screening of firearm access may prompt counseling on safe storage and handling of firearms. The objective of this study was to determine the frequency with which Veterans Health Administration (VHA) healthcare providers document firearm access in electronic health record (EHR) clinical notes, and whether this varied by patient characteristics. METHODS: The study sample is a post-9-11 cohort of veterans in their first year of VHA care, with at least one outpatient care visit between 2012-2017 (N = 762,953). Demographic data, veteran military service characteristics, and clinical comorbidities were obtained from VHA EHR. We extracted clinical notes for outpatient visits to primary, urgent, or emergency clinics (total 105,316,004). Natural language processing and machine learning (ML) approaches were used to identify documentation of firearm access. A taxonomy of firearm terms was identified and manually annotated with text anchored by these terms, and then trained the ML algorithm. The random-forest algorithm achieved 81.9% accuracy in identifying documentation of firearm access. RESULTS: The proportion of patients with EHR-documented access to one or more firearms during their first year of care in the VHA was relatively low and varied by patient characteristics. Men had significantly higher documentation of firearms than women (9.8% vs 7.1%; P < .001) and veterans >50 years old had the lowest (6.5%). Among veterans with any firearm term present, only 24.4% were classified as positive for access to a firearm (24.7% of men and 20.9% of women). CONCLUSION: Natural language processing can identify documentation of access to firearms in clinical notes with acceptable accuracy, but there is a need for investigation into facilitators and barriers for providers and veterans to improve a systemwide process of firearm access screening. Screening, regardless of race/ethnicity, gender, and age, provides additional opportunities to protect veterans from self-harm and violence.

Criticality: A New Concept of Severity of Illness for Hospitalized Children.

OBJECTIVES: To validate the conceptual framework of "criticality," a new pediatric inpatient severity measure based on physiology, therapy, and therapeutic intensity calibrated to care intensity, operationalized as ICU care. DESIGN: Deep neural network analysis of a pediatric cohort from the Health Facts (Cerner Corporation, Kansas City, MO) national database. SETTING: Hospitals with pediatric routine inpatient and ICU care. PATIENTS: Children cared for in the ICU (n = 20,014) and in routine care units without an ICU admission (n = 20,130) from 2009 to 2016. All patients had laboratory, vital sign, and medication data. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: A calibrated, deep neural network used physiology (laboratory tests and vital signs), therapy (medications), and therapeutic intensity (number of physiology tests and medications) to model care intensity, operationalized as ICU (versus routine) care every 6 hours of a patient's hospital course. The probability of ICU care is termed the Criticality Index. First, the model demonstrated excellent separation of criticality distributions from a severity hierarchy of five patient groups: routine care, routine care for those who also received ICU care, transition from routine to ICU care, ICU care, and high-intensity ICU care. Second, model performance assessed with statistical metrics was excellent with an area under the curve for the receiver operating characteristic of 0.95 for 327,189 6-hour time periods, excellent calibration, sensitivity of 0.817, specificity of 0.892, accuracy of 0.866, and precision of 0.799. Third, the performance in individual patients with greater than one care designation indicated as 88.03% (95% CI, 87.72-88.34) of the Criticality Indices in the more intensive locations was higher than the less intense locations. CONCLUSIONS: The Criticality Index is a quantification of severity of illness for hospitalized children using physiology, therapy, and care intensity. This new conceptual model is applicable to clinical investigations and predicting future care needs.

Sedation, Analgesia, and Neuromuscular Blockade: An Assessment of Practices From 2009 to 2016 in a National Sample of 66,443 Pediatric Patients Cared for in the ICU.

OBJECTIVES: To describe the pharmaceutical management of sedation, analgesia, and neuromuscular blockade medications administered to children in ICUs. DESIGN: A retrospective analysis using data extracted from the national database Health Facts. SETTING: One hundred sixty-one ICUs in the United States with pediatric admissions. PATIENTS: Children in ICUs receiving medications from 2009 to 2016. EXPOSURE/INTERVENTION: Frequency and duration of administration of sedation, analgesia, and neuromuscular blockade medications. MEASUREMENTS AND MAIN RESULTS: Of 66,443 patients with a median age of 1.3 years (interquartile range, 0-14.5), 63.3% (n = 42,070) received nonopioid analgesic, opioid analgesic, sedative, and/or neuromuscular blockade medications consisting of 83 different agents. Opioid and nonopioid analgesics were dispensed to 58.4% (n = 38,776), of which nonopioid analgesics were prescribed to 67.4% (n = 26,149). Median duration of opioid analgesic administration was 32 hours (interquartile range, 7-92). Sedatives were dispensed to 39.8% (n = 26,441) for a median duration of 23 hours (interquartile range, 3-84), of which benzodiazepines were most common (73.4%; n = 19,426). Neuromuscular-blocking agents were dispensed to 17.3% (n = 11,517) for a median duration of 2 hours (interquartile range, 1-15). Younger age was associated with longer durations in all medication classes. A greater proportion of operative patients received these medication classes for a longer duration than nonoperative patients. A greater proportion of patients with musculoskeletal and hematologic/oncologic diseases received these medication classes. CONCLUSIONS: Analgesic, sedative, and neuromuscular-blocking medications were prescribed to 63.3% of children in ICUs. The durations of opioid analgesic and sedative medication administration found in this study can be associated with known complications, including tolerance and withdrawal. Several medications dispensed to pediatric patients in this analysis are in conflict with Food and Drug Administration warnings, suggesting that there is potential risk in current sedation and analgesia practice that could be reduced with practice changes to improve efficacy and minimize risks.

A Proficient Spelling Analysis Method Applied to a Pharmacovigilance Task.

Misspellings in clinical free text present potential challenges to pharmacovigilance tasks, such as monitoring for potential ineffective treatment of drug-resistant infections. We developed a novel method using Word2Vec, Levenshtein edit distance constraints, and a customized lexicon to identify correct and misspelled pharmaceutical word forms. We processed a large corpus of clinical notes in a real-world pharmacovigilance task, achieving positive predictive values of 0.929 and 0.909 in identifying valid misspellings and correct spellings, respectively, and negative predictive values of 0.994 and 0.333 as assessments where the program did not produce output. In a specific Methicillin-Resistant Staphylococcus Aureus use case, the method identified 9,815 additional instances in the corpus for potential inaffective drug administration inspection. The findings suggest that this method could potentially achieve satisfactory results for other pharmacovigilance tasks.

An efficient prototype method to identify and correct misspellings in clinical text.

OBJECTIVE: Misspellings in clinical free text present challenges to natural language processing. With an objective to identify misspellings and their corrections, we developed a prototype spelling analysis method that implements Word2Vec, Levenshtein edit distance constraints, a lexical resource, and corpus term frequencies. We used the prototype method to process two different corpora, surgical pathology reports, and emergency department progress and visit notes, extracted from Veterans Health Administration resources. We evaluated performance by measuring positive predictive value and performing an error analysis of false positive output, using four classifications. We also performed an analysis of spelling errors in each corpus, using common error classifications. RESULTS: In this small-scale study utilizing a total of 76,786 clinical notes, the prototype method achieved positive predictive values of 0.9057 and 0.8979, respectively, for the surgical pathology reports, and emergency department progress and visit notes, in identifying and correcting misspelled words. False positives varied by corpus. Spelling error types were similar among the two corpora, however, the authors of emergency department progress and visit notes made over four times as many errors. Overall, the results of this study suggest that this method could also perform sufficiently in identifying misspellings in other clinical document types.

Understanding cancer survivors' information needs and information-seeking behaviors for complementary and alternative medicine from short- to long-term survival: a mixed-methods study.

OBJECTIVE: The research examined complementary and alternative medicine (CAM) information-seeking behaviors and preferences from short- to long-term cancer survival, including goals, motivations, and information sources. METHODS: A mixed-methods approach was used with cancer survivors from the "Assessment of Patients' Experience with Cancer Care" 2004 cohort. Data collection included a mail survey and phone interviews using the critical incident technique (CIT). RESULTS: Seventy survivors from the 2004 study responded to the survey, and eight participated in the CIT interviews. Quantitative results showed that CAM usage did not change significantly between 2004 and 2015. The following themes emerged from the CIT: families' and friends' provision of the initial introduction to a CAM, use of CAM to manage the emotional and psychological impact of cancer, utilization of trained CAM practitioners, and online resources as a prominent source for CAM information. The majority of participants expressed an interest in an online information-sharing portal for CAM. CONCLUSION: Patients continue to use CAM well into long-term cancer survivorship. Finding trustworthy sources for information on CAM presents many challenges such as reliability of source, conflicting information on efficacy, and unknown interactions with conventional medications. Study participants expressed interest in an online portal to meet these needs through patient testimonials and linkage of claims to the scientific literature. Such a portal could also aid medical librarians and clinicians in locating and evaluating CAM information on behalf of patients.

General Symptom Extraction from VA Electronic Medical Notes.

There is need for cataloging signs and symptoms, but not all are documented in structured data. The text from clinical records are an additional source of signs and symptoms. We describe a Natural Language Processing (NLP) technique to identify symptoms from text. Using a human-annotated reference corpus from VA electronic medical notes we trained and tested an NLP pipeline to identify and categorize symptoms. The technique includes a model created from an automatic machine learning model selection tool. Tested on a hold-out set, its precision at the mention level was 0.80, recall 0.74 and an overall f-score of 0.80. The tool was scaled-up to process a large corpus of 964,105 patient records.

PlateRunner: A Search Engine to Identify EMR Boilerplates.

Medical text contains boilerplated content, an artifact of pull-down forms from EMRs. Boilerplated content is the source of challenges for concept extraction on clinical text. This paper introduces PlateRunner, a search engine on boilerplates from the US Department of Veterans Affairs (VA) EMR. Boilerplates containing concepts should be identified and reviewed to recognize challenging formats, identify high yield document titles, and fine tune section zoning. This search engine has the capability to filter negated and asserted concepts, save and search query results. This tool can save queries, search results, and documents found for later analysis.

Spark, an application based on Serendipitous Knowledge Discovery.

Findings from information-seeking behavior research can inform application development. In this report we provide a system description of Spark, an application based on findings from Serendipitous Knowledge Discovery studies and data structures known as semantic predications. Background information and the previously published IF-SKD model (outlining Serendipitous Knowledge Discovery in online environments) illustrate the potential use of information-seeking behavior in application design. A detailed overview of the Spark system illustrates how methodologies in design and retrieval functionality enable production of semantic predication graphs tailored to evoke Serendipitous Knowledge Discovery in users.

Clinical questions raised by clinicians at the point of care: a systematic review.

IMPORTANCE: In making decisions about patient care, clinicians raise questions and are unable to pursue or find answers to most of them. Unanswered questions may lead to suboptimal patient care decisions. OBJECTIVE: To systematically review studies that examined the questions clinicians raise in the context of patient care decision making. DATA SOURCES: MEDLINE (from 1966), CINAHL (from 1982), and Scopus (from 1947), all through May 26, 2011. STUDY SELECTION Studies that examined questions raised and observed by clinicians (physicians, medical residents, physician assistants, nurse practitioners, nurses, dentists, and care managers) in the context of patient care were independently screened and abstracted by 2 investigators. Of 21,710 citations, 72 met the selection criteria. DATA EXTRACTION AND SYNTHESIS: Question frequency was estimated by pooling data from studies with similar methods. MAIN OUTCOMES AND MEASURES: Frequency of questions raised, pursued, and answered and questions by type according to a taxonomy of clinical questions. Thematic analysis of barriers to information seeking and the effects of information seeking on decision making. RESULTS In 11 studies, 7012 questions were elicited through short interviews with clinicians after each patient visit. The mean frequency of questions raised was 0.57 (95% CI, 0.38-0.77) per patient seen, and clinicians pursued 51% (36%-66%) of questions and found answers to 78% (67%-88%) of those they pursued. Overall, 34% of questions concerned drug treatment, and 24% concerned potential causes of a symptom, physical finding, or diagnostic test finding. Clinicians' lack of time and doubt that a useful answer exists were the main barriers to information seeking. CONCLUSIONS AND RELEVANCE: Clinicians frequently raise questions about patient care in their practice. Although they are effective at finding answers to questions they pursue, roughly half of the questions are never pursued. This picture has been fairly stable over time despite the broad availability of online evidence resources that can answer these questions. Technology-based solutions should enable clinicians to track their questions and provide just-in-time access to high-quality evidence in the context of patient care decision making. Opportunities for improvement include the recent adoption of electronic health record systems and maintenance of certification requirements.

Semantic MEDLINE for discovery browsing: using semantic predications and the literature-based discovery paradigm to elucidate a mechanism for the obesity paradox.

Applying the principles of literature-based discovery (LBD), we elucidate the paradox that obesity is beneficial in critical care despite contributing to disease generally. Our approach enhances a previous extension to LBD, called "discovery browsing," and is implemented using Semantic MEDLINE, which summarizes the results of a PubMed search into an interactive graph of semantic predications. The methodology allows a user to construct argumentation underpinning an answer to a biomedical question by engaging the user in an iterative process between system output and user knowledge. Components of the Semantic MEDLINE output graph identified as "interesting" by the user both contribute to subsequent searches and are constructed into a logical chain of relationships constituting an explanatory network in answer to the initial question. Based on this methodology we suggest that phthalates leached from plastic in critical care interventions activate PPAR gamma, which is anti-inflammatory and abundant in obese patients.

A literature-based assessment of concept pairs as a measure of semantic relatedness.

The semantic relatedness between two concepts, according to human perception, is domain-rooted and reflects prior knowledge. We developed a new method for semantic relatedness assessment that reflects human judgment, utilizing semantic predications extracted from PubMed citations by SemRep. We compared the new method to other approaches utilizing path-based, statistical, and context vector methods, using a gold standard for evaluation. The new method outperformed all others, except one variation of the context vector technique. These findings have implications in several natural language processing applications, such as serendipitous knowledge discovery.

Text summarization as a decision support aid.

BACKGROUND: PubMed data potentially can provide decision support information, but PubMed was not exclusively designed to be a point-of-care tool. Natural language processing applications that summarize PubMed citations hold promise for extracting decision support information. The objective of this study was to evaluate the efficiency of a text summarization application called Semantic MEDLINE, enhanced with a novel dynamic summarization method, in identifying decision support data. METHODS: We downloaded PubMed citations addressing the prevention and drug treatment of four disease topics. We then processed the citations with Semantic MEDLINE, enhanced with the dynamic summarization method. We also processed the citations with a conventional summarization method, as well as with a baseline procedure. We evaluated the results using clinician-vetted reference standards built from recommendations in a commercial decision support product, DynaMed. RESULTS: For the drug treatment data, Semantic MEDLINE enhanced with dynamic summarization achieved average recall and precision scores of 0.848 and 0.377, while conventional summarization produced 0.583 average recall and 0.712 average precision, and the baseline method yielded average recall and precision values of 0.252 and 0.277. For the prevention data, Semantic MEDLINE enhanced with dynamic summarization achieved average recall and precision scores of 0.655 and 0.329. The baseline technique resulted in recall and precision scores of 0.269 and 0.247. No conventional Semantic MEDLINE method accommodating summarization for prevention exists. CONCLUSION: Semantic MEDLINE with dynamic summarization outperformed conventional summarization in terms of recall, and outperformed the baseline method in both recall and precision. This new approach to text summarization demonstrates potential in identifying decision support data for multiple needs.

Rethinking information delivery: using a natural language processing application for point-of-care data discovery.

OBJECTIVE: This paper examines the use of Semantic MEDLINE, a natural language processing application enhanced with a statistical algorithm known as Combo, as a potential decision support tool for clinicians. Semantic MEDLINE summarizes text in PubMed citations, transforming it into compact declarations that are filtered according to a user's information need that can be displayed in a graphic interface. Integration of the Combo algorithm enables Semantic MEDLINE to deliver information salient to many diverse needs. METHODS: The authors selected three disease topics and crafted PubMed search queries to retrieve citations addressing the prevention of these diseases. They then processed the citations with Semantic MEDLINE, with the Combo algorithm enhancement. To evaluate the results, they constructed a reference standard for each disease topic consisting of preventive interventions recommended by a commercial decision support tool. RESULTS: Semantic MEDLINE with Combo produced an average recall of 79% in primary and secondary analyses, an average precision of 45%, and a final average F-score of 0.57. CONCLUSION: This new approach to point-of-care information delivery holds promise as a decision support tool for clinicians. Health sciences libraries could implement such technologies to deliver tailored information to their users.

Dynamic summarization of bibliographic-based data.

BACKGROUND: Traditional information retrieval techniques typically return excessive output when directed at large bibliographic databases. Natural Language Processing applications strive to extract salient content from the excessive data. Semantic MEDLINE, a National Library of Medicine (NLM) natural language processing application, highlights relevant information in PubMed data. However, Semantic MEDLINE implements manually coded schemas, accommodating few information needs. Currently, there are only five such schemas, while many more would be needed to realistically accommodate all potential users. The aim of this project was to develop and evaluate a statistical algorithm that automatically identifies relevant bibliographic data; the new algorithm could be incorporated into a dynamic schema to accommodate various information needs in Semantic MEDLINE, and eliminate the need for multiple schemas. METHODS: We developed a flexible algorithm named Combo that combines three statistical metrics, the Kullback-Leibler Divergence (KLD), Riloff's RlogF metric (RlogF), and a new metric called PredScal, to automatically identify salient data in bibliographic text. We downloaded citations from a PubMed search query addressing the genetic etiology of bladder cancer. The citations were processed with SemRep, an NLM rule-based application that produces semantic predications. SemRep output was processed by Combo, in addition to the standard Semantic MEDLINE genetics schema and independently by the two individual KLD and RlogF metrics. We evaluated each summarization method using an existing reference standard within the task-based context of genetic database curation. RESULTS: Combo asserted 74 genetic entities implicated in bladder cancer development, whereas the traditional schema asserted 10 genetic entities; the KLD and RlogF metrics individually asserted 77 and 69 genetic entities, respectively. Combo achieved 61% recall and 81% precision, with an F-score of 0.69. The traditional schema achieved 23% recall and 100% precision, with an F-score of 0.37. The KLD metric achieved 61% recall, 70% precision, with an F-score of 0.65. The RlogF metric achieved 61% recall, 72% precision, with an F-score of 0.66. CONCLUSIONS: Semantic MEDLINE summarization using the new Combo algorithm outperformed a conventional summarization schema in a genetic database curation task. It potentially could streamline information acquisition for other needs without having to hand-build multiple saliency schemas.