Electronic Medical Record Database Research Articles

Missed diagnosis—a major barrier to patient access to obesity healthcare in the primary care setting

ObjectiveTo investigate whether individuals with an elevated BMI measurement, for whom a diagnosis of overweight or obesity (OW/OB) is not recorded, are less likely to be offered clinical care for obesity compared to those with a recorded diagnosis.SubjectsA retrospective cohort study using the electronic medical record database of Maccabi Healthcare Services (MHS) in Israel. Included were 200,000 adults with BMI ≥ 25 kg/m2 measurement recorded during a primary care visit between 2014 and 2020, and no prior diagnosis of OW/OB or related co-morbidities.MethodsThe relationships between a recorded diagnosis of OW/OB and two composite outcomes: 1. A composite of referrals to screening tests for metabolic complications; 2. A composite of weight loss intervention and follow up, were analyzed using multivariate logistic regression models.ResultsIn only 18% of individuals, a diagnosis of OW/OB was recorded. After adjusting for multiple potential confounding factors, individuals who received a recorded diagnosis were 18% more likely to be offered an evaluation for obesity-related metabolic complication, (OR 1.18, 95% CI 1.15–1.21, p < 0.001), and almost twice as likely to be offered intervention and follow up for their excess body weight (OR 1.84, 95% CI 1.76–1.94, p < 0.001) compared to individuals with missed diagnosis. These results persisted after adjusting for inter-physician variability. In addition, male sex, older age, and Arab sector were all associated with lower rates of weight loss intervention and follow up, while young individuals were less likely to be screened for metabolic complications.ConclusionBeyond BMI measurement, a recorded diagnosis of OW/OB is associated with statistically and clinically significant higher rates of performance of obesity care and intervention. Undiagnosed OW/OB presents a significant clinical opportunity, as recording a diagnosis of OW/OB would predict improved patient access to obesity healthcare and improved clinical outcomes.

Validation of an Inhaled Therapy Beliefs Questionnaire in Patients with Chronic Obstructive Pulmonary Disease.

Background: To carry out a validation questionnaire that assesses beliefs about inhaled treatments in patients with chronic obstructive pulmonary disease (COPD), as knowing patients' beliefs could help to improve medication adherence and health outcomes. Methods: We evaluated data from 260 COPD patients from electronic medical record databases from five primary healthcare centers, in a descriptive, cross-sectional study with a sample size calculated for a 10-item questionnaire, with an estimated Cronbach's alpha of 0.70 and a 95% confidence level. Study participants were selected via systematic random sampling. Variables: Ten-item Inhaled Therapy Beliefs Questionnaire, CCTI-Questionnaire v.2.0, time for completion, age, sex, educational level, spirometry severity (GOLD criteria), exacerbations (previous year), characteristics of inhaled treatment, and smoking habit. A two-year follow-up in a subsample of 77 patients from one health center was utilized. The Morisky-Green test, pharmacy dispensing data, test-retest (kappa coefficient), and an exploratory analysis of the adherence-belief relationship (ji-squared) were measured. Results: The 10-item questionnaire showed good viability (3 min completion time) when performed face-to-face or telephonically; its psychometric properties were acceptable, with an internal consistency (Cronbach's alpha) score of 0.613. Three factors explained 47.58% of the total variance (p < 0.0001): use (factor 1), effects (factor 2), and objectives (factor 3) of inhalers. The two-year follow-up ultimately considered 58 out of the 77 patients (10 deceased, 4 unlocated, 2 mistakes, 2 no inhaled treatment, and 1 withdrawal). Non-adherence was 48.3% in terms of the Morisky-Green test; 31% in terms of pharmacy dispensing data; and 40.4% considering both methods. There was low test-retest reliability, indicated by items 4, 8, and 9 of the CCTI-Questionnaire (Kappa = 0.4, 0.26, and 0.34; p-value < 0.0001, 0.008, and 0.001, respectively). There was mild correlation between beliefs and adherence. Conclusions: The ten-item CCTI-Questionnaire v.2.0 demonstrated acceptable psychometric properties regarding feasibility, reliability, and content validity.

Prevalence, T2 Biomarkers, and Cost of Severe Asthma in the Era of Biologics: The BRAVO-1 Study.

Background: The last decade has seen new classifications of the pathophysiology of asthma that have changed the treatment options available. Objectives: To update data on the prevalence of T2 asthma, comorbidities, biomarker characterization, and costs of severe asthma in patients aged ≥12 years, taking into account new classifications and treatment options. Retrospective, observational, nationwide study using a top-down approach. Data were obtained from BIG-PAC®, an electronic medical record database of 1.7 million patients in Spain. The study population comprised patients aged ≥12 years who had received medical care during the period 2016-2017 and been diagnosed with asthma at least 1 year prior to the index date. Patients were followed for 1 year. The prevalence of asthma was 5.5%. Asthma was severe in 3031 of these patients (7.7%), 81.2% of whom presented T2 asthma. Among patients with severe asthma, 64.1% had uncontrolled disease, 31.2% were oral corticosteroid-dependent (37% in the uncontrolled severe asthma group), and only 3.8% were receiving biologics. The most common T2 comorbidities were allergic rhinitis (66.1%), atopic dermatitis (29.1%), and chronic rhinosinusitis with nasal polyps (14.6%). Mortality rates in the total population and uncontrolled severe asthma groups were 4.2% and 5.5%, respectively. The total annual costs per patient with severe asthma were €5890 (uncontrolled) and €2841 (controlled). In the era of biologics, most severe asthma patients present T2 asthma. Despite the availability of new treatments, rates of oral corticosteroid-dependent patients with uncontrolled severe asthma remain high, although biologics continue to be underused. The costs of uncontrolled severe asthma are twice as high as those of controlled severe asthma.

Characterization of the Epidemiologic Profile and Non-pharmacologic Therapy Utilization Profile of Functional Neurologic Disorder in Patients with Migraine from a Multicenter Electronic Medical Record Database (P1-12.008)

Characterization of the Epidemiologic Profile and Non-pharmacologic Therapy Utilization Profile of Functional Neurologic Disorder in Patients with Migraine from a Multicenter Electronic Medical Record Database (P1-12.008)

Abstract 6464: Real-world evidence (RWE) presented at AACR annual meetings: A 10-year trend analysis

Abstract Introduction: RWE is increasingly used in the oncology community, particularly following the 2016 21st Century Cures Act. A recent analysis of RWE presented at ASCO reported a significant increase in relative frequency of RWE abstracts from 2011-2021[1]. We conducted a similar trend analysis of RWE presented at AACR and predictive factors associated with publication. Methods: EMBASE was queried for abstracts presented at AACR 2013-2022, using terms related to study designs and methodologies that employ use of RWE. Details on study design, location, data source, patient number, cancer site, analytical approach, endpoints, and presentation type (oral vs. poster) were extracted and analyzed. Searches in Google Scholar identified published manuscripts associated with each abstract. The Cochran-Armitage trend test and Monte-Carlo estimation of the exact test were used to detect trends over time; p-values are provided for descriptive purposes. Relative risk ratios were calculated to determine statistically significant predictors of publication. Results: Of 56,920 total abstracts presented at AACR from 2013-2022, 611 (1.1%) RWE abstracts were identified. The proportion of presented abstracts with RWE increased from 0.6% in 2013 to 2.0% in 2022 (p&amp;lt;0.001). Among the RWE abstracts, use of electronic medical record (2013: 0% vs. 2022: 9.3%; p=0.002) and genomics databases (i.e., GENIE) (2013: 0% vs. 2022: 8.3%; p=0.006) increased over time. While majority of RWE abstracts were presented as posters (95.9%), the proportion of oral presentations increased in the last decade (2013: 0.0% vs. 2022: 3.7%; p=0.005). Abstracts with sample sizes &amp;lt;500 decreased significantly over time (p=0.002); those with sample sizes 5,000-&amp;lt;50,000 increased significantly over time (p=0.001). The proportion of abstracts including multiple cancer sites (vs. single-site) increased from 5.1% in 2013 to 20.4% in 2022 (p&amp;lt;0.001). The most common cancer sites were lung, breast, and colorectal. Statistically significant factors associated with publishing an abstract into a manuscript included study sample size 500-&amp;lt;5,000 or ≥100,000, inferential statistics (vs. descriptive), propensity score matching, and effectiveness or disease burden endpoints.Conclusion: RWE studies presented at AACR increased in the recent decade, with more robust study designs, incorporation of novel clinico-genomic databases such as GENIE, and advanced analytical approaches. These findings suggest the value of RWE has been increasingly recognized by the oncology research community and that RWE is being used to provide critical insights and inform patient care. Given FDA’s recent guidance on RWE to support regulatory decision-making, this trend will likely continue. [1] Azad Fet al. Longitudinal trends of real-world evidence (RWE) reporting in oncology conferences: An 11-year ASCO Annual Meeting analysis. Citation Format: Jihong Zong, Lauren C. Bylsma, Orsolya Lunacsek, Jon Fryzek. Real-world evidence (RWE) presented at AACR annual meetings: A 10-year trend analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6464.

The correlation between serum total bile acid and adverse perinatal outcomes in pregnant women with intrahepatic cholestasis of pregnancy (ICP) and non-ICP hypercholanemia of pregnancy.

The association between excessive serum total bile acid (TBA) and adverse perinatal outcomes in individuals with non-intrahepatic cholestasis of pregnancy (non-ICP) hypercholanemia has not been determined, and it is unclear if this link is similar to that observed in patients with ICP. To examine the adverse perinatal outcomes in two specific subcategories: those with ICP and those with non-ICP, including individuals with liver disease and asymptomatic hypercholanemia of pregnancy (AHP), at different levels of TBA. Investigate the correlation between TBA levels and adverse perinatal outcomes of ICP, liver disease, and AHP. From 2013 to 2021, pregnant women with excessive TBA levels were taken from the electronic medical record database of our hospital and categorized into three groups: ICP (n = 160), liver disease (n = 164), and AHP (n = 650). This was done as part of a retrospective cohort research project. Multivariable regression and subgroup analyses were performed to examine the association between TBA levels and adverse perinatal outcomes in each group. The study found no significant differences in adverse perinatal outcomes between the ICP and liver disease groups at different TBA levels. However, at moderate TBA levels, both groups had a higher risk of adverse perinatal outcomes than the AHP group (p < 0.017). Among liver disease cases with TBA ≥ 100µmol/L, three cases of perinatal deaths (6.67%) associated with moderate-to-severe acute hepatitis occurred between 27 and 33 weeks of gestation. A 59% higher chance of perinatal death was found for every 10 µmol/L rise in TBA, even after significant variables and confounders were taken into account (adjusted odds ratio (aOR) = 1.59; 95% confidence interval (CI): 1.06-2.40; p = 0.03). If a pregnant woman has moderate-to-severe liver disease and TBA ≥ 100µmol/L, preterm termination of pregnancy (before 34 weeks) may be considered.

Use of a machine learning algorithm with a focus on spinopelvic parameters to predict development of symptomatic tethered cord after initial untethering surgery.

Among patients with a history of prior lipomyelomeningocele repair, an association between increased lumbosacral angle (LSA) and cord retethering has been described. The authors sought to build a predictive algorithm to determine which complex tethered cord patients will develop the symptoms of spinal cord retethering after initial surgical repair with a focus on spinopelvic parameters. An electronic medical record database was reviewed to identify patients with complex tethered cord (e.g., lipomyelomeningocele, lipomyeloschisis, myelocystocele) who underwent detethering before 12 months of age between January 1, 2008, and June 30, 2022. Descriptive statistics were used to characterize the patient population. The Caret package in R was used to develop a machine learning model that predicted symptom development by using spinopelvic parameters. A total of 72 patients were identified (28/72 [38.9%] were male). The most commonly observed dysraphism was lipomyelomeningocele (41/72 [56.9%]). The mean ± SD age at index MRI was 2.1 ± 2.2 months, at which time 87.5% of patients (63/72) were asymptomatic. The mean ± SD lumbar lordosis at the time of index MRI was 23.8° ± 11.1°, LSA was 36.5° ± 12.3°, sacral inclination was 30.4° ± 11.3°, and sacral slope was 23.0° ± 10.5°. Overall, 39.6% (25/63) of previously asymptomatic patients developed new symptoms during the mean ± SD follow-up period of 44.9 ± 47.2 months. In the recursive partitioning model, patients whose LSA increased at a rate ≥ 5.84°/year remained asymptomatic, whereas those with slower rates of LSA change experienced neurological decline (sensitivity 77.5%, specificity 84.9%, positive predictive value 88.9%, and negative predictive value 70.9%). This is the first study to build a machine learning algorithm to predict symptom development of spinal cord retethering after initial surgical repair. The authors found that, after initial surgery, patients who demonstrate a slower rate of LSA change per year may be at risk of developing neurological symptoms.

A single-center real-world review of 10 kHz high-frequency spinal cord stimulation outcomes for treatment of chronic pain

ObjectiveTo compare pragmatic real-world 10-kHz high-frequency spinal cord stimulation (HF-SCS) outcomes at a single academic center to the industry-sponsored SENZA-RCT and Stauss et al. study. MethodsThis single-center retrospective study included patients with refractory back or limb pain trialed and/or permanently implanted with the Nevro HF-SCS system from 2016 to 2021. Demographic and outcome data were obtained from the electronic medical record (EMR) and real-world global database maintained by Nevro Corp. Data obtained from the global database were confirmed using the EMR. Main outcome measures included positive responder status (≥50% patient-reported percentage pain reduction (PRPPR)), improvement in function, improvement in sleep, and reduction in pain medication usage. Comparison groups included patient outcomes from the SENZA-RCT and Stauss et al. study. ResultsPatients (N = 147) trialed with HF-SCS were reviewed, with data available for 137. Positive trialed patient responder rate (≥50% PRPPR) was 77% (106/137, 95CI 70–84%) vs. 87% (1393/1607, 95CI 85–89%) Stauss et al. vs. 93% (90/97, 95CI 88–98%) SENZA-RCT HF-SCS. At the last available follow-up, positive implanted patient responder rate was 73% (58/80, 95CI 63–82%) vs. 78% (254/326, 95CI 73–82%) Stauss et al. vs. 79% (71/90, 95CI 70–87%) SENZA-RCT HF-SCS. Sixty-seven percent (59/88, 95CI 57–77%) reported improved function vs. 72% (787/1088, 95CI 70–75%) Stauss et al.; 45% (31/69, 95CI 33–57%) reported improved sleep vs. 68% (693/1020, 95CI 65–71%) Stauss et al. and 16% (9/56, 95CI 6–26%) reported decrease in medication use vs. 32% (342/1070, 95CI 29–35%) Stauss et al. ConclusionPatient responder rates in this retrospective pragmatic real-world study of HF-SCS are consistent with previous industry-sponsored studies. However, improvements in quality-of-life measures and reduction in medication usage were not as robust as reported in industry-sponsored studies. The findings of this non-industry-sponsored, independent study of HF-SCS complement those of previously published studies by reporting patient outcomes collected in the absence of industry sponsorship.

Safety risks of interstitial lung disease upon real-world usage of Janus kinase inhibitors and biologics for patients with autoimmune diseases: epidemiological study using nationwide electronic medical record database in Japan

Although Janus kinase inhibitor (JAKi) therapy is used for patients with autoimmune diseases (AD), one safety concern, interstitial lung disease (ILD), is life-threatening. We evaluated actual usage of JAKi and safety upon JAKi treatment, in an epidemiological retrospective cohort study utilizing the electronic medical record database in Japan. Among 391,565 AD patients, we analyzed data of new-users receiving JAKi or tumor necrosis factor alpha inhibitor (TNFi)/biologics during the period July 2013–May 2022. ILD (ICD10: J70.2, J70.3, J70.4 and J84) criteria were defined: new-ILD (1) and new-ILD (2) which differed in the latter’s prompter therapeutics cessation upon ILD development. We analyzed ILD occurrence and death, ILD cumulative incidence by the Kaplan–Meier method, and hazard ratio (HR) by the Cox model, for 957 JAKi and 3931 TNFi users. JAKi use has become widespread amidst additional drug-development. Among JAKi users, two-year new-ILD (2) incidence, at 1.4%, was higher than for TNFi users (risk ratio: new-ILD (2) 1.75, death 2.31). Cumulative incidence (2.9% in 20.48 days) was also significantly higher (log-rank test p = .013, HR 2.23 (95% CI 1.16–4.27)); risk factors estimated by HR included JAKi (2.14), rheumatoid arthritis (4.94), diabetes mellitus (2.67) and cerebrovascular disease (2.86). ILD screening is essential.

Clinical and Economic Burden of Antibiotic Use Among Pediatric Patients With Varicella Infection in the Outpatient Setting: A Retrospective Cohort Analysis of Real-world Data in France.

Varicella infects 90% of children before age 9. Though varicella is self-limiting, its complications may require antibiotics, though how antibiotics are utilized for varicella in France is not well known. This study assessed antibiotic use and costs associated with varicella and its complications in pediatric patients managed in the outpatient setting in France. A retrospective cohort study using the Cegedim Strategic Data-Longitudinal Patient Database, an electronic medical record database from general practitioners and office-based specialists in France, was conducted. Children <18 years old diagnosed with varicella between January 2014 and December 2018 with 3-month follow-up available were included. We used descriptive analysis to assess varicella-related complications, medication use, healthcare resource utilization and costs. Overall, 48,027 patients were diagnosed with varicella; 15.3% (n = 7369) had ≥1 varicella-related complication. Antibiotics were prescribed in up to 25.1% (n = 12,045/48,027) of cases with greater use in patients with complications (68.1%, n = 5018/7369) compared with those without (17.3%, n = 7027/40,658). Mean medication and outpatient varicella-related costs were €32.82 per patient with medications costing a mean of €5.84 per patient; antibiotics contributed ~23% to total costs annually. This study showed high antibiotic use for the management of varicella and its complications. A universal varicella vaccination program could be considered to alleviate complications and associated costs in France.

Temporal Trends in Practice Patterns After Introduction of Pediatric Hypertension Guidelines in Canada

In 2016 and 2017, respectively, new Canadian and US guidelines for diagnosis and management of pediatric hypertension (HTN) were published. Six years after their publication, it is unknown whether the recommendations have led to changes in primary care practice patterns. To determine whether HTN guidelines are associated with changes in practice patterns among primary care clinicians. This retrospective, multicenter, population-based cohort study was conducted across 3 phases: January 1, 2011, to December 31, 2015 (era 1), January 1, 2016, to December 31, 2017 (washout period), and January 1, 2018, to December 31, 2019 (era 2). Data were collected from 7 Canadian provinces using the Canadian Primary Care Sentinel Surveillance Network Electronic Medical Record database. Eligible participants included children and adolescents (aged ≥3 to <18 years) with 1 or more encounters in the database. Data analysis was conducted from February 2022 to February 2023. Implementation of the 2016 Hypertension Canada and 2017 American Academy of Pediatrics guidelines. The primary outcomes were annual BP screening documentation, high BP follow-up documentation at 6 months and 1-year, HTN prevalence, laboratory testing rates, and medication prescription rates. Interrupted time series analysis was used to assess the association of the introduction of the Canadian and US guidelines with outcomes. The study included 343 191 children and adolescents (mean [SD] age at first encounter, 6.7 (4.6) years; 173 290 female [50.5%]; 169 901 male [49.5%]), including 235 094 patients in era 1 and 193 473 patients in era 2. In era 1, 55 550 patients (23.6%) had at least 1 BP measurement, and in era 2, 45 006 patients (23.3%) had at least 1 BP measurement. There was a significant increase in BP screening in era 2 from 26 876 of 148 554 screenings (18.1%) to 28 556 of 141 192 screenings (20.2%; β = 0.202; 95% CI, 0.009 to 0.390; P = .04), and the increasing trend was sustained. There was a significant decrease in the trend of follow-up of high BP measurement at 6 months (1265 of 4941 patients with BP measurements [25.6%] to 1718 of 7321 patients with BP measurements [23.5%]; β = -0.490; 95% CI, -0.758 to -0.223; P = .001) and 1 year (1974 of 4941 measurements [40.0%] to 2314 of 7321 measurements [31.6%]; β = -1.392; 95% CI, -1.573 to -1.212; P < .001) in era 2. The proportion of patients meeting HTN criteria significantly increased from 2540 of 55 550 patients (4.6%) in era 1 to 5690 of 45 006 patients (12.6%) in era 2 (β = 0.0210; 95% CI, 0.0021 to 0.0410; P = .03). There was no significant change in the trend of laboratory testing rates in era 2 (949 of 4941 patients tested [19.2%] to 1149 of 7321 patients tested [15.7%]; β = -0.159; 95% CI, -0.364 to 0.046; P = .12). The trend in prescribing of medications to patients with HTN also decreased in era 2 (1305 of 4941 patients prescribed medication [26.4%] to 1415 of 7321 patients prescribed medication [19.3%]; β = -0.605; 95% CI, -0.830 to -0.358; P < .001). The findings of this cohort study within the Canadian primary care setting suggest that there was a significant increase in BP screening and HTN prevalence after the publication of national and international HTN guidelines; however, the follow-up of high BP was still suboptimal. Increasing rates of pediatric HTN emphasize the need for better adherence to pediatric HTN guidelines to improve care and outcomes.

Transmastoid Approach to Repair Iatrogenic Cerebrospinal Fluid Otorhinorrhea After Cerebellopontine Angle Tumor Resection.

The precise treatment of iatrogenic cerebrospinal fluid (CSF) otorhinorrhea has been poorly studied. The purpose of the study was to investigate the clinical manifestation, surgical results, and management of CSF leak. Electronic medical record database of iatrogenic CSF leaks after erebellopontine angle(CPA) surgery from 2019 to 2022 was retrospectively analyzed. Three patients returned to the hospital with the complication of CSF leak. After failed attempts of conservative strategies or reverse surgical repair, adipose tissue was applied to the mastoid cracks repair. With the techniques described above, the CSF leaks were successfully settled. The identified patients were observed for at least 10 months. and there was no recurrence or other complications. Conservative treatment and initial surgical methods for occult postoperative CSF leaks are prone to delay effective results, particularly in patients with well-evaporated temporal bone. This complication can be minimized with transmastoid closure utilizing autologous fat.

Predicting preterm birth using auto-ML frameworks: a large observational study using electronic inpatient discharge data.

To develop and compare different AutoML frameworks and machine learning models to predict premature birth. The study used a large electronic medical record database to include 715,962 participants who had the principal diagnosis code of childbirth. Three Automatic Machine Learning (AutoML) were used to construct machine learning models including tree-based models, ensembled models, and deep neural networks on the training sample (N = 536,971). The area under the curve (AUC) and training times were used to assess the performance of the prediction models, and feature importance was computed via permutation-shuffling. The H2O AutoML framework had the highest median AUC of 0.846, followed by AutoGluon (median AUC: 0.840) and Auto-sklearn (median AUC: 0.820), and the median training time was the lowest for H2O AutoML (0.14 min), followed by AutoGluon (0.16 min) and Auto-sklearn (4.33 min). Among different types of machine learning models, the Gradient Boosting Machines (GBM) or Extreme Gradient Boosting (XGBoost), stacked ensemble, and random forrest models had better predictive performance, with median AUC scores being 0.846, 0.846, and 0.842, respectively. Important features related to preterm birth included premature rupture of membrane (PROM), incompetent cervix, occupation, and preeclampsia. Our study highlights the potential of machine learning models in predicting the risk of preterm birth using readily available electronic medical record data, which have significant implications for improving prenatal care and outcomes.

Real-world prescription of anti-COVID-19 drugs in hospitalized patients with COVID-19 in Japan.

Prescription trends and patterns of anti-COVID-19 drugs in hospitalized patients were examined based on real world data to understand the use of anti-COVID-19 drugs in clinical practice in Japan. The longitudinal and cross-sectional study was conducted utilizing data from January 1, 2019 to December 31, 2021 of the MID-NET® medical information database, which stored the electronic medical records, administrative claim data, and diagnosis procedure combination data of patients in Japan. Hospitalized patients with a COVID-19-related diagnosis who received at least one anti-COVID-19 drug between April 1, 2020 and December 31, 2021. The following 14 drugs were included in this study: remdesivir, baricitinib, combination product of casirivimab and imdevimab, favipiravir, dexamethasone, ivermectin, azithromycin, nafamostat mesylate, camostat mesylate, ciclesonide, tocilizumab, sarilumab, combination product of lopinavir and ritonavir, and hydroxychloroquine. We identified 5,717 patients hospitalized with COVID-19 and prescribed at least one anti-COVID-19 drug. The entire cohort generally included patients over 41-50 years and more males. The most common prescription pattern was dexamethasone monotherapy (22.9%), followed by the concomitant use of remdesivir and dexamethasone (15.0%), azithromycin monotherapy (15.0%), remdesivir monotherapy (10.2%), and nafamostat mesylate monotherapy (8.5%). However, an often prescribed anti-COVID-19 drug differed depending on the period. This study revealed the real-world situation of anti-COVID-19 drug prescriptions in hospitalized COVID-19 patients in Japan. A prescribed drug would depend on the latest scientific evidence, such as efficacy, safety, and approval status, at the time of prescription. Understanding the prescription of anti-COVID-19 drugs will be important for providing the most up-to-date treatments to patients and evaluating the benefit and/or risk of anti-COVID-19 drugs based on the utilization of an electronic medical record database.

P971 The relapse rates and dynamics of disease severity of inflammatory bowel disease after forced biologic treatment discontinuation: Taiwanese multicentre retrospective study

Abstract Background The reimbursement regulations in 2011 for inflammatory bowel disease (IBD) in Taiwan require patients (pts) to discontinue biologic treatment after using the medication for one year and pts can be retreated after ≥ 6 months of forced discontinue. This study aimed to evaluate relapse rates, predictors of relapse and dynamics of disease severity under Taiwan reimbursement policy. Methods This was a multicentre retrospective study using electronic medical record (EMR) database to identify the association between clinical variables and relapse after forced discontinuation in the real-world setting. Pts aged ≥ 20 years old with at least 6 months of biologic treatment were further analyzed to identify the disease severity prior to biologic treatment (closest endpoint before receiving 1st biologic), 1st forced discontinuation, 2nd biologic and 2nd forced discontinuation. EU PAS register number: EUPAS48289. Results We identified 455 newly diagnosed IBD pts from February 2008 to March 2020 in Taiwan. The percentage of relapse after biologic discontinuation was 47.0% (N=353), 41.4% (N=87), and 40.0%(N=15) for 1st, 2nd and 3rd biologic treatment cycle (TxC), respectively. Among pts in the 1st biologic ST, 353 pts were identified who received at least 6 months of biological treatment. Only 87 (24.6%) pts were able to be re-treated for the 2nd biologic TxC under reimbursement criteria after forced biologic discontinuation. Prior to biologic treatment, the use of anti-diarrheal symptom control treatment (HR: 1.452, p = 0.0186) and IBD-related surgery (HR: 1.434, p = 0.0216) were factors associated with an increased risk of relapse on 1st biological TxC. Pts who received the 1st biologic TxC and 2nd biologic TxC with a long biologic treatment duration at the end of treatment (HR: 0.998, p &amp;lt;0.0001); (HR: 0.998, p =0.0003) were less likely to experience relapse. Consistent results were found in the 3rd biologic TxC (HR: 0.963; p =0.0292) long biologic treatment duration was a protective factor against relapse. The percentage of severe pts prior to biologic treatment was 43.7% and dropped to 11.4% after receiving the 1st biologic treatment. However, the percentage of severe pts increased to 32.0% at 1st forced discontinuation and further rose to 35.7% at 2nd forced discontinuation. Conclusion The predictors of relapse could assist physicians in clinical decision-making especially when pts faced with discontinuation. Proportion of moderate to severe disease severity decreased after receiving each TxC and due to forced discontinuation, half of the pts exhibited moderate to severe disease severity after forced discontinuation, suggesting that one year of treatment time was insufficient to reach the desired therapeutic goals.

End-of-Life Practices in an Intensive Care Unit of a Private Hospital in Mexico.

Many factors, such as religion, geography, and customs, influence end-of-life practices. This variability exists even between different physicians. To observe and describe the end-of-life actions of patients in the intensive care unit (ICU) and document the variables that might influence decision-making at the end of life. This is a cross-sectional study performed in the ICU patients of a private hospital from March 2017 to March 2022. We used the Philips Tasy Electronic Medical Record database of clinical records; 298 patients were included in the study during these five years (2017-2022). The data analysis was done with the statistical package SPSS version 23 for Windows. A total of 297 patients were included in this study, of which more than half were men. About 60% of our sample had private health insurance, whereas the remaining paid out of pocket. Most patients had withholding treatment, followed by failed cardiopulmonary resuscitation, withdrawal treatment, and brain death, and none of the patients had acceleration of the dying process. The main cause of admission to the ICU in our center was respiratory complications. Most of our samples were Catholics. Decision-making at the end of life is a complex process. Active participation of the patient, when possible, the patient's family, doctors, and nurses, can give different perspectives and a more compassionate and individualized approach to end-of-life care.

A new electronic medical record database linked to claims data and discharge abstract data (the RWD database) in Japan: Study design and profile.

This article aims to introduce the Real World Database-a new clinical database in Japan. The Health, Clinic, and Education Information Evaluation Institute and Real World Data Co., Ltd. began developing the Real World Database in 2015. This is an electronic medical record database linked to claims data and discharge abstract data from medical institutions in Japan. The institutions agreed to collect data from 218 medical institutions as of June 2021. In 2019, 82 medical institutions provided data, which showed that 2,184,666 patients received treatment at medical institutions. There were also 334,437 inpatients with at least one hospital stay and 2,011,628 outpatients with at least one visit. More than 200 laboratory test results were available. This database is a potential data source for producing descriptive studies, comparative effectiveness studies, studies of adverse effects, and prediction studies. The Real World Database provides an opportunity and strategy to produce real-world evidence for Japan.

Performance of ChatGPT on prehospital acute ischemic stroke and large vessel occlusion (LVO) stroke screening.

The management of acute ischemic stroke (AIS) is time-sensitive, yet prehospital delays remain prevalent. The application of large language models (LLMs) for medical text analysis may play a potential role in clinical decision support. We assess the performance of LLMs on prehospital AIS and large vessel occlusion (LVO) stroke screening. This retrospective study sourced cases from the electronic medical record database of the emergency department (ED) at Maoming People's Hospital, encompassing patients who presented to the ED between June and November 2023. We evaluate the diagnostic accuracy of GPT-3.5 and GPT-4 for the detection of AIS and LVO stroke by comparing the sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and positive likelihood ratio and AUC of both LLMs. The neurological reasoning of LLMs was rated on a five-point Likert scale for factual correctness and the occurrence of errors. On 400 records from 400 patients (mean age, 70.0 years ± 12.5 [SD]; 273 male), GPT-4 outperformed GPT-3.5 in AIS screening (AUC 0.75 (0.65-0.84) vs 0.59 (0.50-0.69), P = 0.015) and LVO identification (AUC 0.71 (0.65-0.77) vs 0.60 (0.53-0.66), P < 0.001). GPT-4 achieved higher accuracy than GPT-3.5 in screening of AIS (89.3% [95% CI: 85.8, 91.9] vs 86.5% [95% CI: 82.8, 89.5]) and LVO stroke identification (67.0% [95% CI: 62.3%, 71.4%] vs 47.3% [95% CI: 42.4%, 52.2%]). In neurological reasoning, GPT-4 had higher Likert scale scores for factual correctness (4.24 vs 3.62), with a lower rate of error (6.8% vs 24.8%) than GPT-3.5 (all P < 0.001). The result demonstrates that LLMs possess diagnostic capability in the prehospital identification of ischemic stroke, with the ability to exhibit neurologically informed reasoning processes. Notably, GPT-4 outperforms GPT-3.5 in the recognition of AIS and LVO stroke, achieving results comparable to prehospital scales. LLMs are supposed to become a promising supportive decision-making tool for EMS practitioners in screening prehospital stroke.

The Relationship between Cognitive Impairment in Psychiatric Patients and Readmission Rate to an Inpatient Facility

The primary intention of this study was to further understand the impact of assessing cognitive impairment in psychiatric patients, as a mediating factor on readmission rates. Mild cognitive dysfunction impacts a patient’s functional outcomes [1,2]. Little information exists to guide best practices in the treatment of adults with cognitive impairment who are hospitalized for acute conditions [2]. A cognitive impairment may impact patient prognosis and ability to function outside of a setting focused on stabilization. Neuropsychological testing is a valuable tool in predicting a patient’s cognitive potential. However, creating a battery of tests that must be short enough to fit the needs of patients within an inpatient setting, allowing them to demonstrate their ability, while still providing enough information for accurate diagnosis and prognosis is a challenge. Overall, this study aimed to investigate the utilization of brief cognitive screeners in identifying patients that may be more vulnerable to relapse and readmission. Cognitive impairment was assessed utilizing the Montreal Cognitive Assessment (MoCA). Readmission rate is defined by total admissions per individual patient. The use of community resources, such as medication management, case management, housing programs, and their relationship with readmission rates was examined. Archival data was taken from an inpatient facility’s HIPAA-compliant electronic medical record (EMR) database. All personal identifying information was de-identified to ensure minimal risk of breaching confidentiality. Although all objectives demonstrated significant correlations among the variables, there were no significant predictive models. This study found a positive linear relationship between admission rate and a higher MoCA score. This may partially be explained by Age (M = 69.09, SD = 8.25) having a negative relationship with inpatient readmission rate and a positive relationship with community resources. Within the inpatient setting, patients with higher cognitive functioning may be more aware of their mental health symptoms and are more likely to seek help. This may lead to more frequent admissions as they are able to seek help to address their mental health concerns. The relationship between patient demographics, community resource usage, and readmission rate was also examined. Limitations of this study included the use of archival data, a small sample size, and a lack of diversity within the sample. Future directions for research include norming the MoCA for psychiatric populations and examining deficit patterns within the cognitive domains.

Exercised-based cardiac rehabilitation associates with lower all-cause mortality in patients with primary pulmonary hypertension.

Despite pharmacological therapies to improve outcomes of pulmonary hypertension (PH), poor long-term survival remains. Exercised-based cardiac rehabilitation (ExCR) may be an alternative strategy to improve prognosis. Therefore, using an electronic medical record (EMR) database, the objective of this study was to compare mortality between patients with primary PH with ExCR vs. propensity-matched PH patients without ExCR. The retrospective analysis was conducted on February 15, 2023 using anonymized data within TriNetX, a global federated health research network. All patients were aged ≥18 years with primary PH recorded in EMRs with at least 1-year follow-up from ExCR. Using logistic regression models, patients with PH with an EMR of ExCR were 1:1 propensity score-matched with PH patients without ExCR for age, sex, race, and comorbidities, and cardiovascular care. In total, 109,736 patients with primary PH met the inclusion criteria for the control group and 784 patients with primary PH met the inclusion criteria for the ExCR cohort. Using the propensity score-matched cohorts, 1-year mortality from ExCR was proportionally lower with 13.6% (n = 101 of 744 patients) in the ExCR cohort compared to 23.3% (n = 174 of 747 patients) in the controls (OR 0.52, 95% CI 0.40-0.68). The present study of 1,514 patients with primary PH suggests that ExCR is associated with 48% lower odds of 1-year mortality, when compared to propensity score-matched patients without ExCR.