Electronic Health Record Data Research Articles

The Association Between Rheumatic Disease Therapies and Cardiovascular Outcomes in People with HIV-A Retrospective Cohort Study.

Introduction: Inflammation is a significant contributor to cardiovascular disease (CVD) in people with HIV (PWH), who face twice the risk of CVD compared to the general population. The presence of co-existing rheumatic disease (RD) may further exacerbate inflammation and increase the incidence of CVD events in this population. Methods: We conducted a retrospective cohort study using electronic health record (EHR) data from the Veterans Affairs Medical Center in Atlanta, covering the period from 2000 to 2019. A total of 5000 patients aged 20-87 years who were diagnosed with HIV and receiving care at the Atlanta VAMC between 2000 and 2019 were eligible for this analysis. This study included 3930 veterans with HIV and assessed the impact of rheumatic disease therapies (RDTs) on CVD outcomes. The primary outcome was the first occurrence of a CVD event. Results: Rheumatic disease was significantly associated with an increased risk of CVD events (OR = 2.67; p < 0.001). Additionally, exposure to multiple RDTs (aHR = 2.121, p = 0.047), NSAIDs (aHR = 1.694, p = 0.003), glucocorticoids (aHR = 2.332, p < 0.0001), and hypouricemic agents and colchicine (aHR = 3.445, p < 0.0001) were all significantly associated with increased CVD events. Conclusions: The co-existence of HIV infection and rheumatic disease, along with the use of RDTs, may amplify the risk of CVD events in PWH. These findings underscore the need for further investigation into the relationship between RD, RDTs, and CVD risk in larger, controlled studies, given the potential implications for treatment decisions in this patient population. A limitation of our study is that due to its retrospective design, we could not examine the impact of the sequential use of RDT groups and RD severity on CVD events.

Identification of Severe Acute Exacerbations of Chronic Obstructive Pulmonary Disease Subgroups by Machine Learning Implementation in Electronic Health Records.

Acute exacerbations of COPD (AECOPD) are heterogeneous. Machine learning (ML) has previously been used to dissect some of the heterogeneity in COPD. The widespread adoption of electronic health records (EHRs) has led to the rapid accumulation of large amounts of patient data as part of routine clinical care. However, it is unclear whether the implementation of ML in EHR-derived data has the potential to identify subgroups of AECOPD. Determine whether ML implementation using EHR data from severe AECOPD requiring hospitalization identifies relevant subgroups. This study used two retrospective cohorts of patients with AECOPD (non-COVID-19 and COVID-19) treated at Yale-New Haven Hospital (YNHHS). K-means clustering was used to identify patient subgroups. We identified three subgroups in the non-COVID cohort (n=1,736). Each subgroup had distinct clinical characteristics. The reference subgroup was the largest (n=904), followed by cardio-renal (n = 548) and eosinophilic (n=284). The eosinophilic subgroup had milder severity of AECOPD, including a shorter hospital stay (p<0.01). The cardio-renal subgroup had the highest mortality during (5%) and in the year after hospitalization (30%). Validation of the severe AECOPD classifier in the COVID-19 cohort recapitulated the characteristics seen in the non-COVID cohort. AECOPD subgroups in the COVID-19 cohort had different IL-1 beta, IL-2R, and IL-8 levels (FDR ≤ 0.05. These specific leukocyte and cytokine profiles resulted in inflammatory differences between the AECOPD subgroups based on C-reactive protein levels. Incorporating ML with EHR-data allows the identification of specific clinical and biological subgroups for severe AECOPD.

Population Performance and Individual Agreement of Coronary Artery Disease Polygenic Risk Scores.

Polygenic risk scores (PRSs) for coronary artery disease (CAD) are a growing clinical and commercial reality. Whether existing scores provide similar individual-level assessments of disease liability is a critical consideration for clinical implementation that remains uncharacterized. Characterize the reliability of CAD PRSs that perform equivalently at the population level at predicting individual-level risk. Cross-sectional Study. All of Us Research Program (AOU), Penn Medicine Biobank (PMBB), and UCLA ATLAS Precision Health Biobank. Volunteers of diverse genetic backgrounds enrolled in AOU, PMBB, and UCLA with available electronic health record and genotyping data. Polygenic risk for CAD from previously published PRSs and new PRSs developed separately from the testing cohorts. Sets of CAD PRSs that perform population prediction equivalently were identified by comparing calibration and discrimination (Brier score and AUROC) of generalized linear models of prevalent CAD using Bayesian analysis of variance. Among equivalently performing scores, individual-level agreement between risk estimates was tested with intraclass correlation (ICC) and Light's Kappa, measures of inter-rater reliability. 50 PRSs were calculated for 171,095 AOU participants. When included in a model of prevalent CAD, 48 scores had practically equivalent Brier scores and AUROCs (region of practical equivalence = 0.02). Across these scores, 84% of participants had at least one score in both the top and bottom risk quintile. Continuous agreement of individual risk predictions from the 48 scores was poor, with an ICC of 0.351 (95% CI; 0.349, 0.352). Agreement between two statistically equivalent scores was moderate, with an ICC of 0.649 (95% CI; 0.646, 0.652). Light's Kappa, used to evaluate consistency of assignment to high-risk thresholds, did not exceed 0.56 (interpreted as 'fair') across statistically and practically equivalent scores. Repeating the analysis among 41,193 PMBB and 50,748 UCLA participants yielded different sets of statistically and practically equivalent scores which also lacked strong individual agreement. Across three diverse biobanks, CAD PRSs that performed equivalently at the population level produced unreliable individual risk estimates. Approaches to clinical implementation of CAD PRSs must consider the potential for discordant individual risk estimates from otherwise indistinguishable scores.

Understanding hospital rehabilitation using electronic health records in patients with and without COVID-19

BackgroundMany hospitalised patients require rehabilitation during recovery from acute illness. We use routine data from Electronic Health Records (EHR) to report the quantity and intensity of rehabilitation required to achieve hospital discharge, comparing patients with and without COVID-19.MethodsWe performed a retrospective cohort study of consecutive adults in whom COVID-19 testing was undertaken between March 2020 and August 2021 across three acute hospitals in Scotland. We defined rehabilitation contacts (physiotherapy, occupational therapy, dietetics and speech and language therapy) from timestamped EHR data and determined contact time from a linked workforce planning dataset. Our aim was to clarify rehabilitation required to achieve hospital discharge and so we excluded patients who died during their admission, and those who did not require rehabilitation (fewer than two specialist contacts). The primary outcome was total rehabilitation time. Secondary outcomes included the number of contacts, admission to first contact, and rehabilitation minutes per day. A multivariate regression analysis for identifying patient characteristics associated with rehabilitation time included age, sex, comorbidities, and socioeconomic status.ResultsWe included 11,591 consecutive unique patient admissions (76 [63,85] years old, 56% female), of which 651 (6%) were with COVID-19, and 10,940 (94%) were admissions with negative testing. There were 128,646 rehabilitation contacts. Patients with COVID-19 received more than double the rehabilitation time compared to those without (365 [165, 772] vs 170 [95, 350] mins, p<0.001), and this was delivered over more specialist contacts (12 [6, 25] vs 6 [3, 11], p<0.001). Admission to first rehabilitation contact was later in patients with COVID-19 (3 [1, 5] vs 2 [1, 4] days from admission). Overall, patients with COVID-19 received fewer minutes of rehabilitation per day of admission (14.1 [9.8, 18.7] vs 15.6 [10.6, 21.3], p<0.001). In our regression analyses, older age and COVID-19 were associated with increased rehabilitation time.ConclusionsPatients with COVID received more rehabilitation contact time than those without COVID, but this was delivered less intensively and was commenced later in an admission. Rehabilitation data derived from the EHR represents a novel measure of delivered hospital care.

Predicting 30-Day and 1-Year Mortality in Heart Failure with Preserved Ejection Fraction (HFpEF).

To develop and compare prediction models for 30-day and 1-year mortality in Heart failure with preserved ejection fraction (HFpEF) using EHR data, utilizing both traditional and machine learning (ML) techniques. HFpEF represents 1 in 2 heart failure patients. Predictive models in HFpEF, specifically those derived from electronic health record (EHR) data, are less established. Using MIMIC-IV EHR data from 2008-2019, patients aged ≥ 18 years admitted with a primary diagnosis of HFpEF were identified using ICD-9 and 10 codes. Demographics, vital signs, prior diagnoses, and lab data were extracted. Data was partitioned into 80% training, 20% test sets. Prediction models from seven model classes (Support Vector Classifier (SVC), Logistic Regression, Lasso Regression, Elastic Net, Random Forest, Histogram-based Gradient Boosting Classifier (HGBC), and XGBoost) were developed using various imputation and oversampling techniques with 5-fold cross-validation. Model performance was compared using several metrics, and individual feature importance assessed using SHapley Additive exPlanations (SHAP) analysis. Among 3910 hospitalizations for HFpEF, 30-day mortality was 6.3%, and 1-year mortality was 29.2%. Logistic regression performed well for 30-day mortality (Area Under the Receiver operating characteristic curve (AUC) 0.83), whereas Random Forest (AUC 0.79) and HGBC (AUC 0.78) for 1-year mortality. Age and NT-proBNP were the strongest predictors in SHAP analyses for both outcomes. Models derived from EHR data can predict mortality after HFpEF hospitalization with comparable performance to models derived from registry or trial data, highlighting the potential for clinical implementation.

Validation of electronic health record data to identify hospital-associated Clostridioides difficile infections for retrospective research.

Clostridioides difficile infection (CDI) research relies upon accurate identification of cases when using electronic health record (EHR) data. We developed and validated a multi-component algorithm to identify hospital-associated CDI using EHR data and determined that the tandem of CDI-specific treatment and laboratory testing has 97% accuracy in identifying HA-CDI cases.

Developing a suicide risk model for use in the Indian Health Service

We developed and evaluated an electronic health record (EHR)-based model for suicide risk specific to an American Indian patient population. Using EHR data for all patients over 18 with a visit between 1/1/2017 and 10/2/2021, we developed a model for the risk of a suicide attempt or death in the 90 days following a visit. Features included demographics, medications, diagnoses, and scores from relevant screening tools. We compared the predictive performance of logistic regression and random forest models against existing suicide screening, which was augmented to include the history of previous attempts or ideation. During the study, 16,835 patients had 331,588 visits, with 490 attempts and 37 deaths by suicide. The logistic regression and random forest models (area under the ROC (AUROC) 0.83 [0.80–0.86]; both models) performed better than enhanced screening (AUROC 0.64 [0.61–0.67]). These results suggest that an EHR-based suicide risk model can add value to existing practices at Indian Health Service clinics.

Familial Renal Glucosuria and Potential Pharmacogenetic Impact on SGLT2 Inhibitors.

Renal glucosuria is a rare inheritable trait caused by loss-of-function variants in the gene that encodes SGLT2 (i.e., SLC5A2). The genetics of renal glucosuria is poorly understood and even less is known on how loss-of-function variants in SLC5A2 may affect response to SGLT2 inhibitors, a new class of medication gaining popularity to treat diabetes by artificially inducing glucosuria. We used two biobanks that link genomic with electronic health record data to study the genetics of renal glucosuria. This included 245,394 participants enrolled in the All of Us (AoU) Research Program and 11,011 enrolled in Marshfield Clinic's Personalized Research Project (PMRP). Association studies in AoU and PMRP identified 10 variants that reached an experiment-wise Bonferroni threshold in either cohort, nine were novel. PMRP was further used as a recruitment source for a prospective SGLT2 pharmacogenetic trial. During a glucose tolerance test, the trial measured urine glucose concentrations in 15 SLC5A2 variant-positive individuals and 15 matched wild types with and without an SGLT2 inhibitor. This trial demonstrated that carriers of SLC5A2 risk variants may be more sensitive to SGLT2 inhibitors compared to wild types (P=0.075). Based on population data, 2% of an ethnically diverse population carry rare variants in SLC5A2 and are at risk for renal glucosuria. As a result, 2% of individuals being treated with SGLT2 inhibitors may respond differently to this new class of medication compared to the general population suggesting a larger investigation into SLC5A2 variants and SGLT2 inhibitors is needed.

The association between glucose-dependent insulinotropic polypeptide and/or glucagon-like peptide-1 receptor agonist prescriptions and substance-related outcomes in patients with opioid and alcohol use disorders: A real-world data analysis.

This study aimed to estimate the strength of association between prescriptions of glucose-dependent insulinotropic polypeptide (GIP) and/or glucagon-like peptide-1 receptor agonists (GLP-1 RA) and the incidence of opioid overdose and alcohol intoxication in patients with opioid use disorder (OUD) and alcohol use disorder (AUD), respectively. This study also aimed to compare the strength of the GIP/GLP-1 RA and substance use-outcome association among patients with comorbid type 2 diabetes and obesity. A retrospective cohort study analyzing de-identified electronic health record data from the Oracle Cerner Real-World Data. About 136 United States of America health systems, covering over 100 million patients, spanning January 2014 to September 2022. The study included 503 747 patients with a history of OUD and 817 309 patients with a history of AUD, aged 18 years or older. The exposure indicated the presence (one or more) or absence of GIP/GLP-1 RA prescriptions. The outcomes were the incidence rates of opioid overdose in the OUD cohort and alcohol intoxication in the AUD cohort. Potential confounders included comorbidities and demographic factors. Patients with GIP/GLP-1 RA prescriptions demonstrated statistically significantly lower rates of opioid overdose [adjusted incidence rate ratio (aIRR) in OUD patients: 0.60; 95% confidence interval (CI) = 0.43-0.83] and alcohol intoxication (aIRR in AUD patients: 0.50; 95% CI = 0.40-0.63) compared to those without such prescriptions. When stratified by comorbid conditions, the rate of incident opioid overdose and alcohol intoxication remained similarly protective for those prescribed GIP/GLP-1 RA among patients with OUD and AUD. Prescriptions of glucose-dependent insulinotropic polypeptide and/or glucagon-like peptide-1 receptor agonists appear to be associated with lower rates of opioid overdose and alcohol intoxication in patients with opioid use disorder and alcohol use disorder. The protective effects are consistent across various subgroups, including patients with comorbid type 2 diabetes and obesity.

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.

Characteristics of men and women with medically diagnosed cluster headache in a national integrated healthcare system: A Veterans Health Administration cohort study.

Describe the epidemiology of cluster headache (CH) using Veterans Health Administration (VHA) Electronic Health Record (EHR) data. Epidemiologic studies of CH at the population level are difficult because it has a prevalence of ~0.1%. Hospital system-wide studies are an attractive alternative as they have large numbers of patients and broader populations than headache or neurology clinic-based studies. The VHA is an ideal hospital-based system in which to study CH because it is nationwide, predominantly male, has a strong focus on mood disorders and suicidality, and has accessible individual medical records. Here, we report the first headache study based on an ongoing longitudinal cohort of patients with CH using VHA EHR data. The VHA EHR data were accessed from Fiscal Year 2008 to 2019. Patients with CH consisted of all patients with at least one outpatient visit containing a CH diagnosis code from the International Classification of Diseases (ICD)-9 or -10. We extracted data on demographic features, incidence, and prevalence, as well as pain and psychiatric comorbidities. Of the 1,524,960 distinct patients who presented for headache treatment in the VHA between Fiscal Year 2008-2019, 24,131 had at least one visit with a CH diagnosis. The 1-year period prevalence of a CH diagnosis in the VHA ranges from 0.08% to 0.10% for women and 0.10% to 0.18% for men. A larger proportion of women versus men received a diagnosis of unspecified CH (59.6% [1412/2368] vs. 53.6% [11,663/21,763], p < 0.001). Most patients with CH had both comorbid headache and non-headache pain diagnoses. Headache not-otherwise-specified was the most common comorbid headache disorder at 70.0% (16,885/24,131) and was more common in women (76.1%, 1801/2368) compared to men (69.3%, 15,084/21,763). Other common comorbidities included migraine, depression, tobacco use, and obstructive sleep apnea. Rates of suicidal ideation or attempt were almost 50% higher in women (5-year proportion 9.4%, 222/2368) with CH compared to men (6.6%, 1433/21,763). To our knowledge this is the largest hospital system study of CH to date and reinforces several previous studies. Pain, mental health, and sleep disorders comorbidities are particularly prevalent in this group and were often more common in women compared to men with CH. Future work should examine gender and race stratified prevalence estimates within the VHA and other healthcare systems.

Health disparities in the risk of severe acidosis: real-world evidence from the All of Us cohort.

To assess the health disparities across social determinants of health (SDoH) domains for the risk of severe acidosis independent of demographical and clinical factors. A retrospective case-control study (n = 13310, 1:4 matching) is performed using electronic health records (EHRs), SDoH surveys, and genomics data from the All of Us participants. The propensity score matching controls confounding effects due to EHR data availability. Conditional logistic regressions are used to estimate odds ratios describing associations between SDoHs and the risk of acidosis events, adjusted for demographic features, and clinical conditions. Those with employer-provided insurance and those with Medicaid plans show dramatically different risks [adjusted odds ratio (AOR): 0.761 vs 1.41]. Low-income groups demonstrate higher risk (household income less than $25k, AOR: 1.3-1.57) than high-income groups ($100-$200k, AOR: 0.597-0.867). Other high-risk factors include impaired mobility (AOR: 1.32), unemployment (AOR: 1.32), renters (AOR: 1.41), other non-house-owners (AOR: 1.7), and house instability (AOR: 1.25). Education was negatively associated with acidosis risk. Our work provides real-world evidence of the comprehensive health disparities due to socioeconomic and behavioral contributors in a cohort enriched in minority groups or underrepresented populations. SDoHs are strongly associated with systematic health disparities in the risk of severe metabolic acidosis. Types of health insurance, household income levels, housing status and stability, employment status, educational level, and mobility disability play significant roles after being adjusted for demographic features and clinical conditions. Comprehensive solutions are needed to improve equity in healthcare and reduce the risk of severe acidosis.

Performance of Machine Learning Suicide Risk Models in an American Indian Population

Few suicide risk identification tools have been developed specifically for American Indian and Alaska Native populations, even though these populations face the starkest suicide-related inequities. To examine the accuracy of existing machine learning models in a majority American Indian population. This prognostic study used secondary data analysis of electronic health record data collected from January 1, 2017, to December 31, 2021. Existing models from the Mental Health Research Network (MHRN) and Vanderbilt University (VU) were fitted. Models were compared with an augmented screening indicator that included any previous attempt, recent suicidal ideation, or a recent positive suicide risk screen result. The comparison was based on the area under the receiver operating characteristic curve (AUROC). The study was performed in partnership with a tribe and local Indian Health Service (IHS) in the Southwest. All patients were 18 years or older with at least 1 encounter with the IHS unit during the study period. Data were analyzed between October 6, 2022, and July 29, 2024. Suicide attempts or deaths within 90 days. Model performance was compared based on the ability to distinguish between those with a suicide attempt or death within 90 days of their last IHS visit with those without this outcome. Of 16 835 patients (mean [SD] age, 40.0 [17.5] years; 8660 [51.4%] female; 14 251 [84.7%] American Indian), 324 patients (1.9%) had at least 1 suicide attempt, and 37 patients (0.2%) died by suicide. The MHRN model had an AUROC value of 0.81 (95% CI, 0.77-0.85) for 90-day suicide attempts, whereas the VU model had an AUROC value of 0.68 (95% CI, 0.64-0.72), and the augmented screening indicator had an AUROC value of 0.66 (95% CI, 0.63-0.70). Calibration was poor for both models but improved after recalibration. This prognostic study found that existing risk identification models for suicide prevention held promise when applied to new contexts and performed better than relying on a combined indictor of a positive suicide risk screen result, history of attempt, and recent suicidal ideation.

Cybersecurity lessons from the Vastaamo psychotherapy data breach for psychiatrists and other mental healthcare providers.

The Vastaamo psychotherapy data breach in Finland is perhaps the largest cybersecurity incident in mental healthcare to date, resulting in significant patient harm. There are specific lessons for mental healthcare providers from an analysis of the incident. Case study of this specific electronic health record data breach, based on detailed media reporting. The issues raised include: the importance of governance of the cybersecurity of sensitive personal patient data, such as compliance with legislative requirements on privacy and data security; specific security measures such as de-identification of data, data protection via passwords, multi-factor authentication, firewalls and encryption; and timely and effective communication, and support of those who have been affected. The implications for mental healthcare providers, including psychiatrists and trainees, are that, within their capability, providers need to assess the efficacy and robustness of cybersecurity of electronic health record systems they use, and carefully consider the information that is recorded to minimise exposures such as in the Vastaamo breach.

Rethinking Measures and Mortality Attribution in Health Care: The ENT Example.

National performance metrics ultimately enhance patient decision-making and promote meaningful improvements in health care delivery, which makes having valid and reliable measures essential. This study examined US News and World Report metrics from 2019 to 2012 and used electronic health record data, combined with detailed chart review across 3 in-system hospitals, to assess the provision of care compared to the attribution of patients assigned to the ear, nose, and throat (ENT) mortality group. Of the initial 47 ENT-attributed deaths, 23 of those were verified, dimensioning the mortality rate from 1.7% to just 0.8%. These results underscore the necessity of rethinking measures and mortality attribution methodologies to be more accurate. Current methods use Medicare Severity Diagnosis Related Group billing coding to map the attribution. We suggest transitioning away from specialty ranking approaches and towards a procedure and condition "rating" approach to ensure that these ranking types capture data about the provision of care within a given encounter.

Identification of Novel Genetic Risk Variants Associated with Hidradenitis Suppurativa in an Exome Sequencing Cohort of 92,455 Individuals

Introduction: Hidradenitis suppurativa (HS) is a prevalent and persistent inflammatory skin disorder, lacking a known cure or effective biomarkers for early diagnosis at present. The genetic determinants of HS have not been fully documented, but it is believed to result from a combination of genetic and environmental factors. Methods: To identify relevant HS gene variants in sporadic HS patients, this study utilized longitudinal electronic health records (EHRs) and whole-exome sequencing. DNA exome sequencing data from 92,455 participant samples in the MyCode biobank, linked to Geisinger’s EHR, were analyzed. This cohort included 1,092 HS cases and 91,363 healthy controls. The MyCode EHR has a median longitudinal follow-up of 15 years per participant, with an average of 87 clinical encounters, 687 laboratory tests, and 7 procedures. Results: There were 1,092 (901 females and 191 males) participants aged 14–89 years (median 47 years) with HS (L73.2), indicating a 1.18% prevalence and accounting for a 4.7:1 female-to-male ratio among the individuals presenting for clinical care. γ-secretase complex, syndromic, and autoinflammatory gene variants were assessed. Potential pathogenic variants were identified among 66 individuals in the HS genes studied. Molecularly, the estimated HS variant prevalence was 1:1,400 in the cohort, 12.3% of variant carriers had HS diagnosis in EHR. Conclusions: Using longitudinal EHR data, genomic screening identified HS-associated gene variants in a defined group of sporadic HS patients to augment the clinical diagnosis, particularly in cases of ambiguity. Based on this study, the field of skin disorders can benefit from a personalized approach to HS diagnosis using large-scale sequencing.

Evaluation of an Outpatient Cervical Ripening Program Using Osmotic Dilators and Foley Balloon Catheters.

This study aimed to describe patient characteristics, satisfaction, and outcome measures for patients undergoing outpatient cervical ripening. A retrospective cohort study using electronic health record data from March 2020 to March 2022 from a large health system. The sample included patients with a low-risk singleton pregnancy undergoing outpatient cervical ripening with either an osmotic dilator or Foley balloon catheter. A subset of patients completed satisfaction surveys. Frequencies and means were used to describe the population and conduct comparisons by device type. Inverse probability of treatment weighted estimates were generated to address baseline differences between patients in the two device groups. Outpatient cervical ripening was completed by 120 patients (80 osmotic dilators and 40 Foley balloon catheters). The mean time from insertion to inpatient admission was 16.2 ± 4.8 hours. The mean change in simplified Bishop score (SBS) was 1.8 ± 1.4 and the mean change in dilation was 1.8 ± 1.1 cm. There were no differences in the amount of cervical change by device type. Patients returned earlier than planned 16.7% of the time, primarily for contractions or rupture of membranes. Following outpatient cervical ripening, the time from admission to delivery was 19.9 ± 10.3 hours, with no difference by device type. Vaginal delivery occurred for 74.8% of patients. Patients reported overall satisfaction with the outpatient cervical ripening experience, with the highest satisfaction among those with osmotic dilators. Patients with both device types stated they would recommend outpatient cervical ripening to others, and experienced low levels of stress and discomfort at home prior to hospital admission. Patients participating in outpatient cervical ripening with osmotic dilators or Foley balloon catheters experienced clinically meaningful changes in dilation and SBSs while at home and reported general satisfaction with the outpatient program experience. · Outpatient use of osmotic dilators or Foley balloon catheters improved Bishop scores.. · Patient and device complications were comparable to other research findings.. · Patients reported overall satisfaction with outpatient cervical ripening..

Equity-Focused Evaluation of a Medicaid-Funded Statewide Diabetes Quality Improvement Project Collaborative.

To evaluate the Ohio Diabetes Quality Improvement Project (QIP) equity aim to reduce the percentage of Non-Hispanic Black (NHB) and Hispanic patients with A1C >9% by ≥20% over 2 years. The Ohio Department of Medicaid, Ohio Colleges of Medicine Government Resource Center, Ohio Medicaid managed care plans, and seven medical schools in Ohio formed the Diabetes QIP collaborative using the collective impact model to improve diabetes outcomes and equity in 20 practices across 11 health systems. The quality improvement (QI) strategies included data audit and feedback, peer-to-peer learning, QI coaching/practice facilitation, and subject matter expert consultation through coaching calls, monthly webinars, and annual virtual learning sessions. Electronic health record data were collected for preintervention (2019-2020) and intervention (2020-2022) periods. Assessments of improvements in A1C were based on prevalence of A1C >9% from preintervention, year 1, and year 2 with stratification by race and ethnicity. The Diabetes QIP included 7,689 (54% female) sociodemographically diverse patients, self-identifying as non-Hispanic White (NHW) (42%), NHB (43%), Hispanic (8%), non-Hispanic Asian (4%), or other (3%). In year 2 compared with baseline, there were decreases in the proportion of patients with A1C >9% among NHW, NHB, and Hispanic patients (NHW from 19% to 12% [37% reduction], NHB 23% to 18% [22% reduction], and Hispanic 29% to 23% [20% reduction]). The Ohio Diabetes QIP, focused on multisector collaborative approaches, reduced the percentage of patients with A1C >9% by ≥20% among NHW, NHB, and Hispanic populations. Given the persistence of disparities, further equity-focused refinements are warranted to address disparities in diabetes control.

Tree-based classification model for Long-COVID infection prediction with age stratification using data from the National COVID Cohort Collaborative.

We propose and validate a domain knowledge-driven classification model for diagnosing post-acute sequelae of SARS-CoV-2 infection (PASC), also known as Long COVID, using Electronic Health Records (EHRs) data. We developed a robust model that incorporates features strongly indicative of PASC or associated with the severity of COVID-19 symptoms as identified in our literature review. The XGBoost tree-based architecture was chosen for its ability to handle class-imbalanced data and its potential for high interpretability. Using the training data provided by the Long COVID Computation Challenge (L3C), which was a sample of the National COVID Cohort Collaborative (N3C), our models were fine-tuned and calibrated to optimize Area Under the Receiver Operating characteristic curve (AUROC) and the F1 score, following best practices for the class-imbalanced N3C data. Our age-stratified classification model demonstrated strong performance with an average 5-fold cross-validated AUROC of 0.844 and F1 score of 0.539 across the young adult, mid-aged, and older-aged populations in the training data. In an independent testing dataset, which was made available after the challenge was over, we achieved an overall AUROC score of 0.814 and F1 score of 0.545. The results demonstrated the utility of knowledge-driven feature engineering in a sparse EHR data and demographic stratification in model development to diagnose a complex and heterogeneously presenting condition like PASC. The model's architecture, mirroring natural clinician decision-making processes, contributed to its robustness and interpretability, which are crucial for clinical translatability. Further, the model's generalizability was evaluated over a new cross-sectional data as provided in the later stages of the L3C challenge. The study proposed and validated the effectiveness of age-stratified, tree-based classification models to diagnose PASC. Our approach highlights the potential of machine learning in addressing the diagnostic challenges posed by the heterogeneity of Long-COVID symptoms.

Digital Health Readiness: Making Digital Health Care More Inclusive.

This paper proposes an approach to assess digital health readiness in clinical settings to understand how prepared, experienced, and equipped individual people are to participate in digital health activities. Existing digital health literacy and telehealth prediction tools exist but do not assess technological aptitude for particular tasks or incorporate available electronic health record data to improve efficiency and efficacy. As such, we propose a multidomain digital health readiness assessment that incorporates a person's stated goals and motivations for use of digital health, a focused digital health literacy assessment, passively collected data from the electronic health record, and a focused aptitude assessment for critical skills needed to achieve a person's goals. This combination of elements should allow for easy integration into clinical workflows and make the assessment as actionable as possible for health care providers and in-clinic digital health navigators. Digital health readiness profiles could be used to match individuals with support interventions to promote the use of digital tools like telehealth, mobile apps, and remote monitoring, especially for those who are motivated but do not have adequate experience. Moreover, while effective and holistic digital health readiness assessments could contribute to increased use and greater equity in digital health engagement, they must also be designed with inclusivity in mind to avoid worsening known disparities in digital health care.