Patient Electronic Health Research Articles

Development and validation of electronic health record-based, machine learning algorithms to predict quality of life among family practice patients

Health-related quality of life (HRQol) is a crucial dimension of care outcomes. Many HRQoL measures exist, but methodological and implementation challenges impede primary care (PC) use. We aim to develop and evaluate a novel machine learning (ML) algorithm that predicts binary risk levels among PC patients by combining validated elements from existing measures with demographic data from patient electronic health records (eHR) to increase predictive accuracy while reducing prospectively-collected data required to generate valid risk estimates. Self-report questions from previously validated QoL surveys were collected from PC patients and combined with their demographic and social determinant (SD) data to form a 53-question item bank from which ML chose the most predictive elements. For algorithm development, 375 observations were allocated to training (n = 301, 80%) or test partitions (n = 74, 20%). Questions that asked participants to rate how happy or satisfied they have been with their lives and how easy or hard their emotional health makes work/school showed a good ability to classify participants’ mental QoL (98% max balanced accuracy). Questions that asked participants to rate how easy or hard it is to do activities such as walking or climbing stairs and how much pain limits their everyday activities showed ability to classify physical QoL (94% max balanced accuracy). No demographic or SD factors were significantly predictive. Supervised machine learning can inform QoL measurements to reduce data collection, simplify scoring, and allow for meaningful use by clinicians. Results from the current study show that a reduced 4-question model may predict QoL almost as well as a full-length 40-question measure.

Association of Benzodiazepine Prescription With Short-Term Prognosis in Elderly Patients Attended in Emergency Department: Results From the EDEN PROJECT.

Benzodiazepine prescription is a growing phenomenon among the elderly population. However, information related to the frequency of these drugs among the elderly population attending in emergency departments (ED) and its impact over prognosis is scarce. The aim of this study is to assess the prevalence of benzodiazepine prescription and to analyze its association with short-term prognosis in elderly patients attended in ED. A retrospective analysis of the EDEN (Emergency Department Elderly in Need) cohort was conducted. This registry included all elderly patients attending in 52 Spanish EDs for any condition, between April 1st and 7th in 2019. Socio-demographic data, comorbidities, and medication were recorded by consulting the patient's electronic health records. The assessed outcomes consisted on new ED visit, hospitalization, and mortality at 30 days after the first ED visit, associated with the use of benzodiazepines at baseline in comparison with no prescription of benzodiazepines. Crude and adjusted logistic regression analyses including patient's comorbidities were performed. Two sensitivity analyses were performed considering concomitant prescription of other central nervous system depressants as well as direct discharge from the ED. 25 557 patients were evaluated (mean age 78 [IQR: 71-84]). 7865 (30.8%) patients were taken benzodiazepines at admission. After adjustment for comorbidities and other central nervous system drugs, benzodiazepine prescription was associated with ED revisit [OR: 1.10 (95%CI: 1.03-1.18)]. Similar results were found in the sensitivity analysis, eliminating patients with central nervous depressors [OR: 1.11 (1.03-1.25)] and patients discharged to home [OR: 1.13 (1.04-1.23)]. No association was found between the use of these drugs and new hospitalizations [OR: 0.90 (0.77-1.05)] or mortality 30 days after discharge [OR: 1.01 (0.88-1.18)]. The results held for all three outcomes in the sensitivity analyses. The use of benzodiazepines is a frequent phenomenon among the elderly population attended in the ED, being associated with an increased risk of new visits to the emergency room, but not with an increased risk of 30-day hospitalization or mortality.

Increased frailty is associated with risk of hospital-acquired infections and death during heart failure hospital admissions

Abstract Background Frailty is common in patients admitted with acute decompensation of heart failure (ADHF) and it is unclear how the frailty syndrome affects heart failure (HF) patients during their acute hospital admission. Understanding the interplay between frailty, impaired renal function, hospital-acquired infections (HAI), and death may be useful in optimising patient care and prognosis. Purpose The aim of this analysis was to explore the relationship between admission frailty levels and the incidence of HAI, acute kidney injury (AKI), and mortality in acute HF hospital admissions. Methods This is a retrospective analysis of data submitted from a single centre to the National HF audit between January and June 2023. Additional frailty and baseline renal function data was retrospectively collected from electronic patient health records. Patients were categorised into three groups according to their Rockwood Clinical Frailty Scale (CSF) (1): no/mild frailty (CFS 1-3), moderate frailty (CFS 4-6) and severe frailty (CFS 7-9). AKI was defined based on baseline and admission creatinine, as per KDIGO guidelines (2). Chi-squared and t-tests were used to compare baseline characteristics between frailty groups. Chi-squared tests were used to compare frailty groups in terms of: HAI, AKI, mortality. Results Between 1st January and 22nd June 2023, there were 344 acute HF hospital admissions in our centre. Most patients had moderate frailty on admission (n=256, 74.4%). Frail patients were significantly older, had a lower admission mean haemoglobin and more were female, compared to non-frail patients (Figure 1A). Patients who were more frail were more likely to experience HAI during their admission; p=0.035 (figure 1B). Furthermore, patients with HAI had a greater risk of mortality, at a median follow-up of 171 (IQR 100.25) days after admission, compared to those without HAI (p>0.001). 41.9% of acute HF admissions were associated with an AKI, and although numerically more frail patients had an AKI, this did not achieve statistical significance (p=0.084). Patients who were more frail were more likely to die, both during hospital admission and at 3 months following discharge; p<0.001 (Figure 1B). Furthermore, this trend persisted after follow-up; p<0.001 (Figure 2). Conclusions Patients with moderate or severe frailty are at greater risk of adverse events (HAI, death) during an acute HF hospital admission, than patients with no or mild frailty. It is imperative to consider the frailty status of all patients with ADHF, to be aware of the frailty-associated risks, and act proactively with targeted interventions (screen for infections, daily bloods, early conversations with patients and their families about treatment escalation plans and patient priorities) to improve outcomes and enhance the quality of care for this vulnerable patient population. Further research is warranted to explore potential interventions to address frailty-related risks in HF management.

PRISM: Patient Records Interpretation for Semantic clinical trial Matching system using large language models

Clinical trial matching is the task of identifying trials for which patients may be eligible. Typically, this task is labor-intensive and requires detailed verification of patient electronic health records (EHRs) against the stringent inclusion and exclusion criteria of clinical trials. This process also results in many patients missing out on potential therapeutic options. Recent advancements in Large Language Models (LLMs) have made automating patient-trial matching possible, as shown in multiple concurrent research studies. However, the current approaches are confined to constrained, often synthetic, datasets that do not adequately mirror the complexities encountered in real-world medical data. In this study, we present an end-to-end large-scale empirical evaluation of a clinical trial matching system and validate it using real-world EHRs. We perform comprehensive experiments with proprietary LLMs and our custom fine-tuned model called OncoLLM and show that OncoLLM outperforms GPT-3.5 and matches the performance of qualified medical doctors for clinical trial matching.

Explainable machine learning and multi-modal data for predicting readmission in patients with heart failure

Abstract Background Cardiovascular diseases (CVDs) are among the leading causes of mortality and morbidity globally, with an estimated 1 million hospital admissions for CVD in England in 2019/20, leading to 5.5 million bed days. In addition, patients with CVDs have a notably high rate of hospital readmission, with approximately one in four patients experiencing heart failure (HF) being readmitted within one month of discharge. We explored whether machine learning models trained on electrocardiogram (ECG) and patient electronic health records (EHRs) can support decision-making and risk management and ultimately improve health outcomes for HF patients. Purpose The project aims to develop a machine learning model to predict health outcomes for HF patients. To test the hypothesis, we trained and tested an explainable machine learning model to predict the risk of mortality and hospital readmission rate for HF patients. Methods We used EHR data and tabulated ECG records to train an XGBoost model to predict the risk of 30-day mortality and 30-day hospital readmission for HF patients. EHR data from a cohort of 2,868 patients with their ECG records were used for model development. We extracted a total of 78 features from the ECG data, such as heart rate variability (HRV), P-wave durations, and QT intervals. These ECG features were combined with coded and free-text EHR data (e.g., demographic information, diagnosis, medication history, and lab results) and used as tabular inputs to the XGBoost model. Of the 2,868 patients, an internal set of 574 patients stratified by age, sex, and health outcomes (30-day mortality and 30-day hospital readmission) were held out for testing. Results Among 2,868 HF patients with 10-year follow-up, 1065 patients were reported for mortality or hospital readmission within 30 days of discharge. When the machine learning model was trained on ECG data in addition to EHR data, the model was able to predict patient mortality and hospital readmission rate in the test set with an area under the curve (AUC) of 0.91 and 0.70, respectively. For mortality prediction, the model achieved a sensitivity and specificity of 0.81 and 0.82, respectively, when employing an operating point with minimum difference between the sensitivity and specificity. When the threshold was adjusted to favor sensitivity at 0.91, the model specificity decreased to 0.74. The machine learning model also identified high sensitive troponin levels, QT dispersion, and lactate dehydrogenase levels as features of high importance for predicting the outcomes for HF patients. Conclusions Our developed machine learning model was able to predict mortality and readmission risks in HF patients using multi-modal data. Upon further validation, the machine learning model has the potential to be tested for supporting clinical decision-making and risk management for HF patients.ROC CurveFeature Importance

A network-based systems genetics framework identifies pathobiology and drug repurposing in Parkinson's disease.

Parkinson's disease (PD) is the second most prevalent neurodegenerative disorder. However, current treatments are directed at symptoms and lack ability to slow or prevent disease progression. Large-scale genome-wide association studies (GWAS) have identified numerous genomic loci associated with PD, which may guide the development of disease-modifying treatments. We presented a systems genetics approach to identify potential risk genes and repurposable drugs for PD. First, we leveraged non-coding GWAS loci effects on multiple human brain-specific quantitative trait loci (xQTLs) under the protein-protein interactome (PPI) network. We then prioritized a set of PD likely risk genes (pdRGs) by integrating five types of molecular xQTLs: expression (eQTLs), protein (pQTLs), splicing (sQTLs), methylation (meQTLs), and histone acetylation (haQTLs). We also integrated network proximity-based drug repurposing and patient electronic health record (EHR) data observations to propose potential drug candidates for PD treatments. We identified 175 pdRGs from QTL-regulated GWAS findings, such as SNCA, CTSB, LRRK2, DGKQ, CD38 and CD44. Multi-omics data validation revealed that the identified pdRGs are likely to be druggable targets, differentially expressed in multiple cell types and impact both the parkin ubiquitin-proteasome and alpha-synuclein (a-syn) pathways. Based on the network proximity-based drug repurposing followed by EHR data validation, we identified usage of simvastatin as being significantly associated with reduced incidence of PD (fall outcome: hazard ratio (HR) = 0.91, 95% confidence interval (CI): 0.87-0.94; for dementia outcome: HR = 0.88, 95% CI: 0.86-0.89), after adjusting for 267 covariates. Our network-based systems genetics framework identifies potential risk genes and repurposable drugs for PD and other neurodegenerative diseases if broadly applied.

Risk of Suicidal Ideation or Attempts in Adolescents With Obesity Treated With GLP1 Receptor Agonists

Glucagon-like peptide 1 receptor agonists (GLP1R) are increasingly being used for the treatment of obesity in adolescents. It is currently unknown whether GLP1R treatment is associated with suicidal ideation or attempts in this population. To investigate the association between GLP1R initiation and suicidal ideation or attempts in adolescents with obesity. Retrospective propensity score-matched cohort study using electronic health records from the TriNetX global federated network between December 2019 and June 2024. The analysis included data from 120 health care organizations, mainly from the USA. Participants were adolescents aged 12 to 18 years with a diagnosis of obesity and evidence of an antiobesity GLP1R prescription or lifestyle intervention without GLP1R within the following year. Cohorts were balanced for baseline demographic characteristics, psychiatric medications and comorbidities, and diagnoses associated with socioeconomic status and health care access using propensity score matching. Initial prescription of GLP1R (study cohort) or lifestyle intervention without GLP1R (control cohort). Incidence of suicidal ideation or attempts based on International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes recorded in patient electronic health records during 12 months of follow-up. Diagnoses of upper respiratory tract infections (URTI) were used as negative control outcomes, and gastrointestinal symptoms (GI) were used as positive control outcomes. A total of 4052 adolescents with obesity and a concomitant antiobesity intervention were identified for the GLP1R cohort and 50 112 were identified for the control cohort. Propensity score matching resulted in 3456 participants in each balanced cohort. Prescription of GLP1R was associated with a 33% reduced risk for suicidal ideation or attempts over 12 months of follow-up (1.45% vs 2.26%; hazard ratio [HR], 0.67; 95% CI, 0.47-0.95; P = .02) and a higher rate of GI symptoms (6.9% vs 5.4%; HR, 1.41; 95% CI, 1.12-1.78; P = .003) but no difference in rates of URTI diagnoses. In this study, adolescents with obesity prescribed a GLP1R had a lower incidence of suicidal ideation or attempts compared with matched patients not prescribed GLP1R who were treated with lifestyle intervention. These results suggest a favorable psychiatric safety profile of GLP1R in adolescents. The detected reduction in HRs for suicidal ideation among adolescents with obesity prescribed GLP1R suggests potential avenues for future research.

Intravitreal therapy in neovascular age-related macular degeneration-adapting to increasing demand and changing times.

To report the outcomes of patients with neovascular age-related macular degeneration (nAMD) at Palmerston North Eye Clinic (PNEC) during 2020 and 2021, comparing time to treatment initiation based on nurse-injector availability and during COVID-19 restrictions. Data were recorded from a prospective database for patients with nAMD at PNEC. Each patient's electronic health record was reviewed to ensure the accuracy of the database and to fill in missing data points. Statistics were done using Microsoft Excel and R. One hundred and fifty-six eyes were diagnosed with nAMD during the study. Mean time from referral triage to first injection was 13.08 days across the study period. Time to treatment initiation was not statistically different by level of COVID-19 restriction but there was a significant difference in first specialist appointment to injection interval when three nurse-injectors were available compared to four. The effect seemed most evident in subsequent months after reduced nurse-injector availability began. Despite an increase in nAMD diagnoses each year, PNEC continues to meet national guidelines for interval from referral to treatment initiation through innovations in practice. As demand for intravitreal injections continues to increase, further resourcing and research into newer agents will be required to keep wait times compliant with guidelines.

Abstract A130: Examining breast cancer screening completion: Comparing patient self-report and electronic health record data at a Federally Qualified Health Center during the SARS-CoV-2 pandemic

Abstract Intro: There is evidence to suggest that preventive services, such as breast cancer screening, were impacted during the SARS-CoV-2 pandemic with interruption in screening mammogram services during March-July 2020. Federally Qualified Health Centers (FQHCs) are a critical health system to help better understand pandemic sequelae among diverse patient populations so teams can better prepare for a future pandemic. US Preventive Services Task Force (USPSTF) recommends average-risk patients aged 50-74 screen for breast cancer biyearly. Based on USPSTF recommendations in 2021, 76% of non-Hispanic White women and 74 % of Latina women were up-to-date on mammography screening. Our study, RESTORE, focuses on understanding patient barriers and facilitators to completing breast cancer screening at one large urban FQHC during the pandemic. Here, we compare FQHC patient self-reported completion of breast cancer screening to documented breast cancer screening found in the electric health record (EHR). Methods: Patients (n=20) were interviewed between July 2023 and February 2024 about their healthcare utilization, breast cancer screening (including medical site), and health literacy throughout the COVID period (2018-2022). When a screening date was documented in the EHR, interviewers mentioned the screening date to participants during the interview. Interviews were audio recorded and transcribed. A preliminary review of these interviews was conducted to collect self-reported breast cancer screening date and compare it to the date in the EHR. Results: Interviewees’ average age was 65.1 years; 70 % were Spanish-speaking; 60% were Latino, and 95% were female. Of the Spanish-speakers, 8 out of 14 patients self-reported completing a mammogram between 2018-2022, with 2 of 8 reporting having completed it with an onsite vendor. Only 3 of these 8 mammograms were captured in the EHR. Of these three, 2 patients were prompted with the EHR date during the interview: one patient agreed with the date while the other rejected the date. Among the English-speaking patients, 4 out of 6 patients self-reported completing a mammogram between years 2018-2022, all of these 4 self-reported dates were captured in the EHR. Of these 4 patients, all were prompted with the EHR date, 3 agreed to the date and one rejected the date, with 3 of 4 reporting having completed it with an onsite vendor.Conclusion: Throughout COVID, FQHC patients sought mammogram screening; however, gaps in data systems limited completeness of the health system’s mammogram EHR data. Time delays between patient experience and interview dates, complicated our understanding of patients’ COVID experiences. The inconsistency between self-reported dates and EHR dates also creates challenges educating and convincing patients to stay up-to-date on screenings based on recommended intervals. Missing EHR data could have also impacted staff not accurately following-up for mammograms or referrals. Additional work is needed to understand the opportunities to better integrate FQHC EHR with mammogram vendor data systems. Citation Format: Anne Escaron, Crystal Gomez, Jennifer Schneider, Jennifer Rivelli, Fabian Hidalgo, Gloria Coronado. Examining breast cancer screening completion: Comparing patient self-report and electronic health record data at a Federally Qualified Health Center during the SARS-CoV-2 pandemic [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr A130.

Implementing TeleSleep at Veterans Healthcare Administration: an organizational case study of adaptation and sustainment.

Veteran access to sleep medicine is of paramount importance to the Veterans Health Administration (VA). To increase access, VA has created community referral policies and programs, as well as telehealth programs. In 2017, the Office of Rural Health (ORH) funded a TeleSleep initiative focused on reaching rural Veterans with unmet sleep needs. ORH provided 3-6 years of funding to help 19 hubs support 98 spoke sites serving rural Veterans. As ORH funding concluded, each hub identified its path to sustainment. This case study follows one TeleSleep hub in VA's western geographic region as it transitioned from ORH funding sustainment as a regional Sleep Clinical Resource Hub. This case study describes the real-world process of adaptation in care delivery strategies. One key area of adaptation revolved around whether to deliver care via the patient's home facility or the provider's home facility. In early 2021, the TeleSleep team implemented an innovative provider transfer model, where temporary reinforcements from the TeleSleep hub increased the workforce capacity of spoke sites, similar to the concept of locum tenens. In this provider transfer model, TeleSleep clinicians scheduled, documented, and billed for each encounter at the Veteran's home facility. Positioning TeleSleep clinicians as local providers facilitated communication and referrals and promoted continuity and quality of care for Veterans in their home facility. This provider transfer model reduced the administrative burden of providers and schedulers and supported patient-side-only documentation of care. While this mirrors current locum tenens practice, transferring providers did not fit VA's financial model as implemented by the western region's Sleep Clinical Resource Hub. Therefore, in December 2021, VA aligned TeleSleep with VA's preferred practice of patient rather than provider transfers. In the patient transfer model, providers schedule and document in both the provider and patient electronic health records, and bill in the provider's facility. However, reflecting on this period of innovation, TeleSleep team members concluded that the provider transfer model could improve patient safety and care coordination while reducing the administrative burden of frontline clinicians. Further research and development are needed to align the provider transfer model with VA's financial model.

A Machine Learning Method for Allocating Scarce COVID-19 Monoclonal Antibodies

During the COVID-19 pandemic, the effective distribution of limited treatments became a crucial policy goal. Yet, limited research exists using electronic health record data and machine learning techniques, such as policy learning trees (PLTs), to optimize the distribution of scarce therapeutics. To evaluate whether a machine learning PLT-based method of scarce resource allocation optimizes the treatment benefit of COVID-19 neutralizing monoclonal antibodies (mAbs) during periods of resource constraint. This retrospective cohort study used electronic health record data from October 1, 2021, to December 11, 2021, for the training cohort and data from June 1, 2021, to October 1, 2021, for the testing cohort. The cohorts included patients who had positive test results for SARS-CoV-2 and qualified for COVID-19 mAb therapy based on the US Food and Drug Administration's emergency use authorization criteria, ascertained from the patient electronic health record. Only some of the qualifying candidates received treatment with mAbs. Data were analyzed between from January 2023 to May 2024. The primary outcome was overall expected hospitalization, assessed as the potential reduction in overall expected hospitalization if the PLT-based allocation system was used. This was compared to observed allocation using risk differences. Among 9542 eligible patients in the training cohort (5418 female [56.8%]; age distribution: 18-44 years, 4151 [43.5%]; 45-64 years, 3146 [33.0%]; and ≥65 years, 2245 [23.5%]), a total of 3862 (40.5%) received mAbs. Among 6248 eligible patients in the testing cohort (3416 female [54.7%]; age distribution: 18-44 years, 2827 [45.2%]; 45-64 years, 1927 [30.8%]; and ≥65 years, 1494 [23.9%]), a total of 1329 (21.3%) received mAbs. Treatment allocation using the trained PLT model led to an estimated 1.6% reduction (95% CI, -2.0% to -1.2%) in overall expected hospitalization compared to observed treatment allocation in the testing cohort. The visual assessment showed that the PLT-based point system had a larger reduction in 28-day hospitalization compared with the Monoclonal Antibody Screening Score (maximum overall hospitalization difference, -1.0% [95% CI, -1.3% to -0.7%]) in the testing cohort. This retrospective cohort study proposes and tests a PLT method, which can be linked to a electronic health record data platform to improve real-time allocation of scarce treatments. Use of this PLT-based allocation method would have likely resulted in fewer hospitalizations across a population than were observed in usual care, with greater expected reductions than a commonly used point system.

Embracing Digital: The Transition to Electronic Patient Health Education.

Embracing Digital: The Transition to Electronic Patient Health Education.

Improving Clinical Decision Making With a Two-Stage Recommender System.

Clinical decision-making is complex and time-intensive. To help in this effort, clinical recommender systems (RS) have been designed to facilitate healthcare practitioners with personalized advice. However, designing an effective clinical RS poses challenges due to the multifaceted nature of clinical data and the demand for tailored recommendations. In this article, we introduce a 2-Stage Recommendation framework for clinical decision-making, which leverages a publicly accessible dataset of electronic health records. In the first stage, a deep neural network-based model is employed to extract a set of candidate items, such as diagnoses, medications, and prescriptions, from a patient's electronic health records. Subsequently, the second stage utilizes a deep learning model to rank and pinpoint the most relevant items for healthcare providers. Both retriever and ranker are based on pre-trained transformer models that are stacked together as a pipeline. To validate our model, we compared its performance against several baseline models using different evaluation metrics. The results reveal that our proposed model attains a performance gain of approximately 12.3% macro-average F1 compared to the second best performing baseline. Qualitative analysis across various dimensions also confirms the model's high performance. Furthermore, we discuss challenges like data availability, privacy concerns, and shed light on future exploration in this domain.

262. EOSINOPHILIC ESOPHAGITIS: A SEVEN YEAR ANALYSIS OF EPIDEMIOLOGY IN A UNITED KINGDOM COUNTY

Abstract Background Eosinophilic Esophagitis (EoE) is growing in incidence internationally. It is an immune-mediated disease that is characterised by esophageal mucosal eosinophilia and esophageal dysfunction. Awareness of EoE is also crucial for Esophagogastric surgeons as symptoms may mimic reflux disease or neoplasia, key considerations for an accurate differential diagnosis. Presently, there are no incidence estimates of EoE in the adult population in the UK. This study seeks to establish a cohort of EoE patients in Norfolk, a county in the East of England region, and to describe the incidence, characteristics, and diagnostic features of affected patients in both the adult and paediatric populations. Methods A retrospective analysis was performed of a prospectively maintained pathology database servicing a population of 761,000 persons in Norfolk. All newly diagnosed cases of EoE across a seven-year period between January 2017 and December 2023 were identified. Patients included for analysis had a histological diagnosis of EoE with >15 eosinophils/hpf and symptoms of oesophageal dysfunction. Patient electronic health care records were reviewed for histology reports, endoscopic findings, and clinical history. The primary outcome was incidence of EoE within the county. Secondary outcomes assessed included: patient characteristics; allergy and atopy history; endoscopic features, biopsy protocol and referral indication. Results 187 (median age 43 (4-85) years, Male: Female (126:61)) new cases of EoE were identified, of whom eight were in the paediatric age group of ≤16 years. The average annual incidence of EoE was 4.29 [95% CI 2.92, 5.66] and 0.89 [95% CI 0, 1.39] per 100,000 person/years in adult and paediatric groups respectively, with a significantly higher incidence in adults (RR 4.86 (95% CI [2.39 – 9.85] P<0.005) (Figure 1). The most common referral indications for endoscopy were dysphagia (80.7%) and food bolus obstruction episodes (25.1%). The most common endoscopic feature of EoE was circular rings (49.2%). Conclusion This is the first study reporting incidence of EoE in the UK adult population. There was a significantly greater incidence of EoE in adults compared to the paediatric population. However, there was no significant change in incidence of EoE in either group over the study period. Incidence estimates identified are similar to those reported in other European populations. Importantly 18.2% of patients had normal appearance of oesophagus on endoscopy, which reinforces the need to take oesophageal biopsies where there is clinical suspicion. Ideally, prospective registry data is needed for a comprehensive characterisation of incidence and prevalence of EoE.

Teledentistry utilization by oral health professionals and policy considerations: A mixed methods case study.

To examine variations in the delivery of teledentistry after the COVID-19 pandemic and to explore oral health professionals' utilization of various teledentistry modes of delivery. In this mixed methods case study, we collected teledentistry claims data from patient electronic health records (n = 78,756) encompassing various types of teledentistry utilization in clinical settings and through community outreach from a Dental Support Organization (DSO) in Oregon from January 2021 to November 2022. We analyzed the patient demographic and claims data using descriptive statistics and logistic regression analyses to identify patterns of teledentistry service delivery. Qualitatively, we conducted virtual interviews (n = 13) through Microsoft Teams with oral health professionals about their experiences with teledentistry utilization. We used inductive and deductive coding to code individual transcripts and identify common themes among provider experiences. Out of the 78,756 electronic health record claims for teledentistry, 75.7% used synchronous audio, 13.4% used synchronous video, and 10.9% used asynchronous teledentistry methods. We observed a 8.6% increase in synchronous audio teledentistry utilization at the end of the study period, compared with a 4.2% increase in synchronous video and a 4.4% decrease in asynchronous teledentistry. Oral health professionals interviewed reported choosing the type of teledentistry delivery based on patient and provider access to the virtual teledentistry platform. Oral health professionals' knowledge of and experiences with teledentistry need to be considered when developing policy and best practices for the use of teledentistry for patient care.

Clinical presentation and factors associated with gluten exposure in children with celiac disease.

The prevalence of celiac disease (CeD) is increasing, yet it is still underdiagnosed, in part because of its heterogeneous presentation. Diagnostic criteria are evolving and management with strict adherence to a gluten-free diet is challenging for many. We aimed to characterize the clinical presentation of CeD among a large multicenter cohort of pediatric patients and to identify factors associated with gluten-free diet adherence. Patients with CeD aged 0-18 years were recruited from 11 United States health centers. Parents completed surveys about gluten-free diet adherence and patient electronic health records were reviewed. Logistic regression analyses were performed to identify risk factors associated with gluten exposure. Charts were reviewed for 460 children with a median age of 6.4 years. Abdominal pain was reported in 57% of the cohort, but diverse symptoms were identified. Parent surveys were completed for 455 participants. Sixty-five (14%) participants were at high risk for gluten exposure based on parental reports of weekly or daily gluten exposure or eating gluten by choice in the past year. Participants under the age of 5 years had a lower risk of gluten exposure, while participants without repeat serology testing 18 months after initial diagnosis were at higher risk of gluten exposure. In a large, multicenter cohort of pediatric CeD patients, clinical presentation is highly variable, necessitating a high index of suspicion to make a diagnosis. Parent surveys indicate that 14% of patients are at high risk of gluten exposure, with patient age and lack of close follow-up associated with gluten-free diet adherence.

Otolaryngic Allergy Patient Journey Mapping: A Framework for Allergy Immunotherapy Adherence.

Allergen-specific immunotherapy (AIT) is an effective treatment for allergic disease but requires long treatment duration and premature cessation is of significant concern. Drivers of premature cessation remain poorly understood and no predictive models currently exist. We hypothesized that a novel patient journey map and de novo real-time patient electronic health status instruments (eHSIs) could effectively capture patient perceived cost, commitment, and treatment benefit to identify individual patients at risk for premature AIT cessation. Cross-Sectional Observational Study. A single Otolaryngology allergy immunotherapy (AIT) program was studied over 5 years. Instances of premature cessation were classified. An Otolaryngic Allergy Patient Journey Map was developed to identify and target automated, real-time, patient-reported, electronic health status instrument responses. Data capture was robust, with 61,406 data points collected and an eHSI survey completion rate of 81.3%. However, based on correlation analysis and logistic regression alone, real-time eHSI responses were not predictive of individual patient premature AIT cessation. A total of 597 AIT patients discontinued treatment prematurely: 64.4% stopping within the first year. Specifically, 74.0%-76.3% of subcutaneous AIT patients and 88.5%-100% of sublingual AIT patients did not complete the minimum recommended treatment duration of 3 years. Patient journey mapping can aid in the design of longitudinal care models and patient engagement strategies. Yet, eHSI patient responses of perception of AIT cost, benefit, and convenience did not correlate with the likelihood of premature treatment cessation. Our imperfect clinical intuition may not account for the dynamic drivers of premature AIT discontinuation. Future development of predictive tools feed by large patient-centric data sets may be incorporated into routine practice resulting in delivery of a more streamlined and personalized approach with reduced premature AIT cessation, improved outcomes, and reduced health care expenditures. NA Laryngoscope, 134:S1-S14, 2024.

Association of Semaglutide With Tobacco Use Disorder in Patients With Type 2 Diabetes : Target Trial Emulation Using Real-World Data.

Reports of reduced desire to smoke in patients treated with semaglutide, a glucagon-like peptide receptor agonist (GLP-1RA) medication for type 2 diabetes mellitus (T2DM) and obesity, have raised interest about its potential benefit for tobacco use disorders (TUDs). To examine the association of semaglutide with TUD-related health care measures in patients with comorbid T2DM and TUD. Emulation target trial based on a nationwide population-based database of patient electronic health records. United States, 1 December 2017 to 31 March 2023. Seven target trials were emulated among eligible patients with comorbid T2DM and TUD by comparing the new use of semaglutide versus 7 other antidiabetes medications (insulins, metformin, dipeptidyl-peptidase-4 inhibitors, sodium-glucose cotransporter-2 inhibitors, sulfonylureas, thiazolidinediones, and other GLP-1RAs). The TUD-related health care measures (medical encounter for diagnosis of TUD, smoking cessation medication prescriptions, and smoking cessation counseling) that occurred within a 12-month follow-up were examined using Cox proportional hazards and Kaplan-Meier survival analyses. The study compared 222 942 new users of antidiabetes medications including 5967 of semaglutide. Semaglutide was associated with a significantly lower risk for medical encounters for TUD diagnosis compared with other antidiabetes medications, and was strongest compared with insulins (hazard ratio [HR], 0.68 [95% CI, 0.63 to 0.74]) and weakest but statistically significant compared with other GLP-1RAs (HR, 0.88 [CI, 0.81 to 0.96]). Semaglutide was associated with reduced smoking cessation medication prescriptions and counseling. Similar findings were observed in patients with and without a diagnosis of obesity. For most of the group comparisons, the differences occurred within 30 days of prescription initiation. Documentation bias, residual confounding, missing data on current smoking behavior, body mass index, and medication adherence. Semaglutide was associated with lower risks for TUD-related health care measures in patients with comorbid T2DM and TUD compared with other antidiabetes medications including other GLP-1Ras, primarily within 30 days of prescription. These findings suggest the need for clinical trials to evaluate semaglutide's potential for TUD treatment. National Institutes of Health.

Associations between patient portal use and electronic health record (EHR) data timeliness in type 2 diabetes mellitus care

Abstract Objective Patient data is subject to missingness and errors. Patient portals enable patients managing type 2 diabetes mellitus (T2DM) to review and correct data to avoid retesting, medication errors, and diagnostic mistakes. We examined whether patient portal use was associated with electronic health record (EHR) data timeliness in T2DM care. Research Design and methods We analyzed EHR data from a panel of adult patients to determine whether portal use improved data timeliness. EHR data timeliness is measured as the number of days between patient encounters, accounting for mean attribute update periods, where available EHR attribute updates for T2DM measurements were present, including body mass, weight, glycated hemoglobin A1c, cholesterol, blood pressure, serum creatinine, and smoking status. We performed negative binomial regressions with fixed effects to estimate the association between patient portal use and EHR data timeliness. Sensitivity analyses were conducted using Poisson regressions. Results Nearly a third (31.3%) of patients in our sample actively used the health portal. There were fewer days (111.9 days vs. 136.7 days; p < 0.001) between EHR attribute updates for patients who used health portals compared to patients who did not. Data timeliness was lower among female, non-Hispanic White Medicare beneficiaries. Based on regression analyses, portal use was associated with an expected 3.6 (p < 0.001) percentage point decrease in days between attribute updates, indicating improved EHR timeliness. Conclusion Improving the quality of health information may streamline decision-making in partnership with patients who produce data points across clinical settings. Active use of patient portals and digital health tools in chronic disease care are critical for care management and clinical decision-making, especially for patients managing type 2 diabetes across clinical settings.

Ginkgo biloba extract prescriptions are associated with less frequent repeat visits to ENT doctors due to tinnitus: a retrospective cohort study.

We aimed to evaluate the drug utilization of Ginkgo biloba extract (Gbe), systemic corticosteroids (CSs), and pentoxifylline (PTXF) for the treatment of acute tinnitus by analyzing electronic patient health record data. In addition, we assessed whether the different drug treatments were associated with different frequencies of repeat visits to ear, nose, and throat (ENT) doctors. This retrospective cohort study used data from the IQVIA Disease Analyzer (DA) database. It included patients with an initial diagnosis of tinnitus between January 2005 and December 2021, treated by ENT specialists in Germany. Of 111,629 patients meeting all selection criteria, 51,205 received prescriptions of Gbe, 34,817 of systemic CSs, and 25,607 of PTXF. Gbe prescription was associated with significantly lower odds of a repeat consultation due to tinnitus compared to systemic CSs [odds ratio (OR) 0.91; 95% confidence interval (CI): 0.88-0.95] as well as PTXF (OR 0.74; 95% CI: 0.72-0.77). This association was significant in both men and women and in some age groups. Gbe is the most frequently ENT specialist-prescribed drug for the treatment of acute tinnitus. It is associated with a reduced likelihood of patients consulting their ENT specialist for tinnitus again compared to systemic CSs and PTXF.