Large Electronic Medical Record Research Articles

Impact of polygenic risk score for triglyceride trajectory and diabetic complications in subjects with type 2 diabetes based on large electronic medical record data from Taiwan: a case control study.

The prevalence of diabetic dyslipidemia has gradually increased worldwide and individuals with hypertriglyceridemia often have a high polygenic burden of triglyceride (TG)-increasing variants. However, the contribution of genetic variants to dyslipidemia in patients with type 2 diabetes (T2D) remains limited. Therefore, in this study, we aimed to investigate the genetic characteristics of longitudinal changes in TG levels among patients with T2D and summarize the genetic effects of polygenic risk score (PRS) on TG trajectory and risk of diabetic complications. We conducted a case-control study. A total of 11,312 patients with T2D with longitudinal TG and genetic data were identified from a large hospital database in Taiwan. We then performed a genome-wide association study and calculated the relative PRS. In total, 21 single-nucleotide polymorphisms (SNPs) related to TG trajectory were identified and yielded an area under the receiver operating characteristic curve (ROC) of 0.712 for high TG trajectory risk among Taiwanese patients with T2D. A cumulative genetic effect was observed for high TG trajectory, even when considering the adherence of a lipid-lowering agent in stratified analysis. An increased PRS increases high TG trajectory risk in a logistic regression model (odds ratio = 1.55; 95% confidence interval [CI] = 1.31-1.83 in the validation cohort). The TG-specific PRS was associated with the risk of diabetic microvascular complications, including diabetic retinopathy and nephropathy (with hazard ratios of 1.11 [95% CI = 1.01-1.21, P = 0.027] and 1.05 [95% CI = 1.01-1.1, P = 0.018], respectively). This study may contribute to the identification of patients with T2D who are at risk of abnormal TG levels and diabetic microvascular complications using polygenic information.

Reducing unwarranted chest x-rays in bronchiolitis: Importance of a robust analysis.

Bronchiolitis is the commonest reason for hospitalisation amongst infants and is often a target for low-value care (LVC) reduction. We aimed to assess the impact of a multifaceted intervention (clinician education, parent engagement, audit-feedback) on rates of chest x-rays (CXR) in bronchiolitis. Longitudinal study of CXRs ordered in infants (1-12 months) diagnosed with bronchiolitis in the Emergency Department (ED) of an Australian paediatric hospital between May 2016 and February 2023. We used logistic regression to measure the impact of the intervention on unwarranted CXR orders, controlling for other potential impacting variables such as time, patient characteristics (age/sex), clinical variables (fever, hypoxia, tachypnoea), seasonal factors (month, day of the week, business hours) and time passed since intervention. Ten thousand one hundred and nine infants were diagnosed with bronchiolitis in the ED over the study period, with 939 (9.3%) receiving a CXR, of which 69% (n = 651) were considered unwarranted. Rates of unwarranted CXRs reduced from 7.9% to 5.4% post-intervention (P < 0.0001). Logistic regression showed the intervention had no significant effect (OR 0.89, 95% CI 0.65-1.23) once other variables and underlying time-based trends were accounted for. Although pre-post rates appeared significantly improved, a robust analysis demonstrated that our multi-faceted intervention was not effective in reducing CXRs in bronchiolitis. The decision to order CXR was associated with clinical features that overlap with pneumonia suggesting ongoing misconceptions regarding the role of CXR for this indication. Our study highlights the value of large electronic medical record datasets and robust methodology to avoid falsely attributing underlying trends to the LVC intervention.

Establishing the Reportable Interval for Routine Clinical Laboratory Tests: A Data-Driven Strategy Leveraging Retrospective Electronic Medical Record Data.

This paper presents a data-driven strategy for establishing the reportable interval in clinical laboratory testing. The reportable interval defines the range of laboratory result values beyond which reporting should be withheld. The lack of clear guidelines and methodology for determining the reportable interval has led to potential errors in reporting and patient risk. To address this gap, the study developed an integrated strategy that combines statistical analysis, expert review, and hypothetical outlier calculations. A large data set from an accredited clinical laboratory was utilized, analyzing over 124 million laboratory test records from 916 distinct tests. The Dixon test was applied to identify outliers and establish the highest and lowest non-outlier result values for each test, which were validated by clinical pathology experts. The methodology also included matching the reportable intervals with relevant Logical Observation Identifiers Names and Codes (LOINC) and Unified Code for Units of Measure (UCUM)-valid units for broader applicability. Upon establishing the reportable interval for 135 routine laboratory tests (493 LOINC codes), we applied these to a primary care laboratory data set of 23 million records, demonstrating their efficacy with over 1% of result records identified as implausible. We developed and tested a data-driven strategy for establishing reportable intervals utilizing large electronic medical record (EMR) data sets. Implementing the established interval in clinical laboratory settings can improve autoverification systems, enhance data reliability, and reduce errors in patient care. Ongoing refinement and reporting of cases exceeding the reportable limits will contribute to continuous improvement in laboratory result management and patient safety.

Parkinson Disease Genetics Extended to African and Hispanic Ancestries in the VA Million Veteran Program.

Nearly all genetic analyses of Parkinson disease (PD) have been in populations of European ancestry. We sought to test the ability of a machine learning method to extract accurate PD diagnoses from an electronic medical record (EMR) system, to see whether genetic variants identified in European populations generalize to individuals of African and Hispanic ancestries, and to compare the rates of PD across ancestries. A machine learning method using natural language processing was applied to EMRs of US veterans participating in the VA Million Veteran Program (MVP) to identify individuals with PD. These putative cases were vetted via blind chart review by a movement disorder specialist. A polygenic risk score (PRS) of 90 established genetic variants whose genotypes were imputed from a customized Axiom Biobank Array was evaluated in different case groups. The EMR prediction scores had a distinct trimodal distribution, with 97% of the high group and only 30% of the middle group having a credible diagnosis of PD. Using the 3,542 cases from the high group matched 4:1 to controls, the PRS was highly predictive in individuals of European ancestry (n = 3,137 cases; OR = 1.82; p = 8.01E-48), and nearly identical effect sizes were seen in individuals of African (n = 184; OR = 2.07; p = 3.4E-4) and Hispanic ancestries (n = 221; OR = 2.13; p = 3.9E-6). The PRS was much less predictive for the 2,757 European ancestry cases who had an ICD code for PD but for whom the machine learning method had a lower confidence in their diagnosis. No novel ancestry-specific genetic variants were identified. Individuals with African ancestry had one-quarter the rate of PD compared with European or Hispanic ancestries aged 60-70 years and one half the rate in the 70-80 years age range. African American cases had a higher proportion of their DNA originating in Europe compared with African American controls. Machine learning can reliably classify PD using data from a large EMR. Larger studies of non-European populations are required to confirm the generalizability of PD risk variants identified in populations of European ancestry and the increased risk coming from a higher proportion of European DNA in African Americans.

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.

Back to Babies: Reducing Documentation Time in the NICU

Abstract Background There is no accepted best practice for generation and content of daily progress notes in the neonatal intensive care unit (NICU). Objectives This study aimed to implement a consistent documentation standard process for a neonatology provider group at a level IV tertiary care NICU. The primary aim was to improve timeliness of daily progress note completion. Secondary aims were to maintain or improve clinician satisfaction, reduce variability, and reduce attending neonatologist electronic medical record (EMR) documentation tasks. Methods We formed a work group including advanced practice providers (APPs) and physicians from the NICU that met over 6 months to define the ideal NICU documentation content, map the workflow for documentation, identify gaps in EMR content, and create solutions for each gap. Baseline assessment included a change readiness survey to identify barriers to workflow change and a review of neonatologist signature timestamp to determine time to note completion. Twenty random progress notes were sampled weekly for 6 months prior to implementation of new workflow as well as 6 months postimplementation. Average time to note completion was compared in the pre- and postintervention groups. Results In total, 962 notes were sampled, 481 each in the pre- and postintervention states. Twenty neonatologists were captured in the preintervention state, 24 in the postintervention state, 18 were captured in both samples. Final note completion time mean improved from 10 hours and 32 minutes (from starting note to final sign) to 8 hours and 40 minutes (p &lt; 0.01). Those sampled in both epochs improved from 10 hours and 6 minutes to 8 hours and 30 minutes (p &lt; 0.05). Conclusion Progress notes generated by neonatologists are completed earlier than those generated by an APP with a Neonatologist addendum. Specialty-specific education and training are critical to high satisfaction in large EMR workflow transitions.

Abstract 17671: Comparative Analysis of Warfarin and Direct Oral Anticoagulants for the Treatment of Left Ventricular Thrombus

Introduction: This study aims to evaluate the efficacy and safety of direct oral anticoagulants (DOACs) in the treatment of left ventricular thrombus (LVT). Warfarin has been shown to be an effective treatment in mitigating thrombotic events; however, numerous limitations exist, including the need for frequent monitoring and increased risk of side effects. Despite this, insight into DOAC use for the treatment of LVT remains limited. Thus, this study seeks to investigate the safety and efficacy of DOACs as a potential alternative to warfarin for the treatment of LVT. Hypothesis: DOACs decreases the risk of adverse events compared to Warfarin in patients with LVT. Methods: A retrospective cohort study was conducted using a federated realtime database called TriNetX which is comprised of over 100 million patient records. ICD-10 diagnostic codes and national drug codes were used to stratify adults by warfarin or DOAC treatment within 3 months of an initial LVT diagnosis. A 1:1 matched propensity score analysis was conducted, adjusting for comorbidities and demographics, to calculate adjusted Risk Ratios (aRR) with 95% CI. 90-day risk for mortality, stroke, need for blood transfusion, hemorrhage, and arterial embolism &amp; thrombosis were examined. Results: In a matched sample of 11,206 patients in each cohort, patients with LVT who took warfarin within 3 months of initial diagnosis is at higher risk for stroke (aRR[95%CI]=1.44(1.34,1.55)], need for blood transfusion [1.29(1.11,1.51)], hemorrhage [1.56(1.35,1.8)], and arterial thromboembolism [1.28(1.13,1.45)] than LVT patients who took a DOAC. Risk for death was non-significant between cohorts. Conclusions: The findings from this large electronic medical record patient cohort establish higher rates of adverse events for patients with LVT treated with warfarin versus DOAC, with no statistically significant difference in therapeutic efficacy.

Comparative Effectiveness Analysis of Lumpectomy and Mastectomy for Elderly Female Breast Cancer Patients: A Deep Learning-based Big Data Analysis.

Objectives: To evaluate the comparative effectiveness of treatments, a randomized clinical trial remains the gold standard but can be challenged by a high cost, a limited sample size, an inability to fully reflect the real world, and feasibility concerns. The objective is to showcase a big data approach that takes advantage of large electronic medical record (EMR) data to emulate clinical trials. To overcome the limitations of regression analysis, a deep learning-based analysis pipeline was developed. Study Design and Setting: Lumpectomy (breast-conserving surgery) and mastectomy are the two most commonly used surgical procedures for early-stage female breast cancer patients. An emulation trial was designed using the Surveillance, Epidemiology, and End Results (SEER)-Medicare data to evaluate their relative effectiveness in overall survival. The analysis pipeline consisted of a propensity score step, a weighted survival analysis step, and a bootstrap inference step. Results: A total of 65,997 subjects were enrolled in the emulated trial, with 50,704 and 15,293 in the lumpectomy and mastectomy arms, respectively. The two surgery procedures had comparable effects in terms of overall survival (survival year change = 0.08, 95% confidence interval (CI): -0.08, 0.25) for the elderly SEER-Medicare early-stage female breast cancer patients. Conclusion: This study demonstrated the power of "mining large EMR data + deep learning-based analysis," and the proposed analysis strategy and technique can be potentially broadly applicable. It provided convincing evidence of the comparative effectiveness of lumpectomy and mastectomy.

POSTER: MDS-567 Real-World Analysis of a Large Electronic Medical Record Database of Patients With Higher-Risk Myelodysplastic Neoplasms (HR MDS): Treatment Profiles, Clinical Effectiveness, and Key Adverse Event Rates

POSTER: MDS-567 Real-World Analysis of a Large Electronic Medical Record Database of Patients With Higher-Risk Myelodysplastic Neoplasms (HR MDS): Treatment Profiles, Clinical Effectiveness, and Key Adverse Event Rates

MDS-567 Real-World Analysis of a Large Electronic Medical Record Database of Patients With Higher-Risk Myelodysplastic Neoplasms (HR MDS): Treatment Profiles, Clinical Effectiveness, and Key Adverse Event Rates

MDS-567 Real-World Analysis of a Large Electronic Medical Record Database of Patients With Higher-Risk Myelodysplastic Neoplasms (HR MDS): Treatment Profiles, Clinical Effectiveness, and Key Adverse Event Rates

The risk of osteoporosis in patients with ankylosing spondylitis—A large retrospective matched cohort study

BackgroundAnkylosing spondylitis (AS) is associated with increased bone turnover and systemic inflammation. Osteoporosis is common but frequently underappreciated in AS, studies regarding the incidence of osteoporosis in AS are limited and based on small cohorts. The aim of this study is to assess the risk of osteoporosis in patients with AS compared to matched controls. MethodsA population based retrospective cohort study using data retrieved from a large electronic medical record in Israel, the Clalit health services. Included patients that were diagnosed with AS from January 2002 to December 2018 were followed for development of osteoporosis. The incidence of osteoporosis was compared between AS and controls and a logistic regression model was used to assess the interaction between AS and osteoporosis. ResultsThe study included 5476 AS patients, and 27,657 age- and sex-frequency matched controls. The incidence of osteoporosis in AS patients was significantly higher than controls (4.7% vs 2.8%, p<0.001) in the whole cohort as well as when stratified by sex. Osteoporosis developed earlier in patients with AS versus controls (4.1 vs 5.2 years, p<0.001). In multivariate analysis and after adjustment to several potential confounders, AS was found to independently associated with osteoporosis (HR 1.83, 95%CI 1.58–2.11, p<0.0001). ConclusionsOur study confirms the higher incidence and earlier development of osteoporosis in patients with AS. Such finding highlights the increased need of awareness and earlier detection of such comorbidity allowing prompt treatment to prevent undesired sequalae including increased risk of fractures.

The risk of osteoporosis in patients with ankylosing spondylitis—A large retrospective matched cohort study

BackgroundAnkylosing spondylitis (AS) is associated with increased bone turnover and systemic inflammation. Osteoporosis is common but frequently underappreciated in AS, studies regarding the incidence of osteoporosis in AS are limited and based on small cohorts. The aim of this study is to assess the risk of osteoporosis in patients with AS compared to matched controls. MethodsA population based retrospective cohort study using data retrieved from a large electronic medical record in Israel, the Clalit health services. Included patients that were diagnosed with AS from January 2002 to December 2018 were followed for development of osteoporosis. The incidence of osteoporosis was compared between AS and controls and a logistic regression model was used to assess the interaction between AS and osteoporosis. ResultsThe study included 5476 AS patients, and 27,657 age- and sex-frequency matched controls. The incidence of osteoporosis in AS patients was significantly higher than controls (4.7% vs 2.8%, p<0.001) in the whole cohort as well as when stratified by sex. Osteoporosis developed earlier in patients with AS versus controls (4.1 vs 5.2 years, p<0.001). In multivariate analysis and after adjustment to several potential confounders, AS was found to independently associated with osteoporosis (HR 1.83, 95%CI 1.58–2.11, p<0.0001). ConclusionsOur study confirms the higher incidence and earlier development of osteoporosis in patients with AS. Such finding highlights the increased need of awareness and earlier detection of such comorbidity allowing prompt treatment to prevent undesired sequalae including increased risk of fractures.

Construction of Cohorts of Similar Patients From Automatic Extraction of Medical Concepts: Phenotype Extraction Study.

Reliable and interpretable automatic extraction of clinical phenotypes from large electronic medical record databases remains a challenge, especially in a language other than English. We aimed to provide an automated end-to-end extraction of cohorts of similar patients from electronic health records for systemic diseases. Our multistep algorithm includes a named-entity recognition step, a multilabel classification using medical subject headings ontology, and the computation of patient similarity. A selection of cohorts of similar patients on a priori annotated phenotypes was performed. Six phenotypes were selected for their clinical significance: P1, osteoporosis; P2, nephritis in systemic erythematosus lupus; P3, interstitial lung disease in systemic sclerosis; P4, lung infection; P5, obstetric antiphospholipid syndrome; and P6, Takayasu arteritis. We used a training set of 151 clinical notes and an independent validation set of 256 clinical notes, with annotated phenotypes, both extracted from the Assistance Publique-Hôpitaux de Paris data warehouse. We evaluated the precision of the 3 patients closest to the index patient for each phenotype with precision-at-3 and recall and average precision. For P1-P4, the precision-at-3 ranged from 0.85 (95% CI 0.75-0.95) to 0.99 (95% CI 0.98-1), the recall ranged from 0.53 (95% CI 0.50-0.55) to 0.83 (95% CI 0.81-0.84), and the average precision ranged from 0.58 (95% CI 0.54-0.62) to 0.88 (95% CI 0.85-0.90). P5-P6 phenotypes could not be analyzed due to the limited number of phenotypes. Using a method close to clinical reasoning, we built a scalable and interpretable end-to-end algorithm for extracting cohorts of similar patients.

Inflammatory Bowel Diseases Are Associated with Polymyositis and Dermatomyositis-A Retrospective Cohort Analysis.

Background and Objectives: Polymyositis and dermatomyositis (PM/DM) are classified as polygenic autoimmune diseases, whereas inflammatory bowel disease (IBD) is considered a polygenic autoinflammatory disease. In the literature, several cases exist reporting the co-occurrence of both conditions. At the molecular level, PM/DM and IBD share common genetic determinants including interferon regulatory factor and vitamin D receptor susceptibility loci. Accumulating evidence underline several indicators that confer poor prognosis in IBD, including antinuclear antibody positivity and the presence of other autoimmune diseases, therefore the aim of this study is to assess the association between these entities. Materials and Methods: This is a population-based retrospective study using data retrieved from a large electronic medical record in Israel, the Clalit health registry. The sample included PM/DM patients and age- and sex-frequency matched controls. The prevalence of IBD in PM/DM was compared between the two groups and logistic regression was applied to control for confounding variables. Predictors of IBD in patients with PM/DM were also explored. Results: Our study included 12,278 subjects with 2085 PM/DM patients and 10,193 age- and sex- frequency-matched controls. The incidence of IBD in patients with PM/DM was significantly higher even after controlling for various confounding variables (OR of 1.73, 95% CI 1.05-2.86, p-value = 0.033). Anti-nuclear antibodies (ANA) positivity was found to be an independent predictor for IBD diagnosis in patients with PM/DM (OR 3.67, 95% CI 1.01-13.36, p = 0.048). Conclusion: Our analysis reports an association between IBD and PM/DM. Such association could point towards a common pathophysiological background. Further research is needed to further describe the clinical courses and whether a unique therapeutic approach is warranted.

Genome-wide association study of hyperthyroidism based on electronic medical record from Taiwan.

Excess thyroid hormones have complex metabolic effects, particularly hyperthyroidism, and are associated with various cardiovascular risk factors. Previous candidate gene studies have indicated that genetic variants may contribute to this variable response. Electronic medical record (EMR) biobanks containing clinical and genomic data on large numbers of individuals have great potential to inform the disease comorbidity development. In this study, we combined electronic medical record (EMR) -derived phenotypes and genotype information to conduct a genome-wide analysis of hyperthyroidism in a 35,009-patient cohort in Taiwan. Diagnostic codes were used to identify 2,767 patients with hyperthyroidism. Our genome-wide association study (GWAS) identified 44 novel genomic risk markers in 10 loci on chromosomes 2, 6, and 14 (P < 5 × 10–14), including CTLA4, HCP5, HLA-B, POU5F1, CCHCR1, HLA-DRA, HLA-DRB9, TSHR, RPL17P3, and CEP128. We further conducted a comorbidity analysis of our results, and the data revealed a strong correlation between hyperthyroidism patients with thyroid storm and stroke. In this study, we demonstrated application of the PheWAS using large EMR biobanks to inform the comorbidity development in hyperthyroidism patients. Our data suggest significant common genetic risk factors in patients with hyperthyroidism. Additionally, our results show that sex, body mass index (BMI), and thyroid storm are associated with an increased risk of stroke in subjects with hyperthyroidism.

Clinical Characteristics and Healthcare Resource Utilization among Patients with Obstructive Hypertrophic Cardiomyopathy Treated in a Range of Settings in the United States.

Obstructive hypertrophic cardiomyopathy (oHCM) has been studied primarily in comprehensive centers of excellence. Broadening the understanding of patients with oHCM in the general population may improve identification and treatment in other settings. This retrospective cohort study identified adults with oHCM from a large electronic medical record database comprising data from 39 integrated delivery networks (IBM Explorys; observational period: January 2009–July 2019). Clinical characteristics, healthcare resource utilization (HCRU), and outcomes were reported. Of 8791 patients, 53.0% were female and the mean index age was 61.8 years. Cardiovascular drugs prescribed included beta-blockers (80.5%), calcium channel blockers (46.0%), and disopyramide (2.4%). Over time, heart failure, atrial fibrillation, and ventricular arrhythmias increased. Surgical procedures included septal myectomy (22.0%), alcohol septal ablation (0.6%), and heart transplantation (0.3%). Implantable cardioverter defibrillators were present in 11.2% of patients. After initial septal reduction therapy (SRT), HCRU increased and 550 patients (27.7%) required a reintervention. Of the overall group, 2.7% experienced sudden cardiac arrest by end of study. In conclusion, this cohort of patients with oHCM had guideline-recommended drug therapy and procedures. Despite this, heart failure, atrial fibrillation, and ventricular arrhythmias increased, and more than a quarter of patients undergoing SRT required reintervention. These unresolved issues emphasize the unmet need for new, effective therapies for patients with oHCM.

Leveraging unstructured electronic medical record notes to derive population-specific suicide risk models

Electronic medical record (EMR)-based suicide risk prediction methods typically rely on analysis of structured variables such as demographics, visit history, and prescription data. Leveraging unstructured EMR notes may improve predictive accuracy by allowing access to nuanced clinical information. We utilized natural language processing (NLP) to analyze a large EMR note corpus to develop a data-driven suicide risk prediction model. We developed a matched case-control sample of U.S. Department of Veterans Affairs (VA) patients in 2015 and 2016. We randomly matched each case (all patients that died by suicide in that interval, n = 5029) with five controls (patients that remained alive). We processed note corpus using NLP methods and applied machine-learning classification algorithms to output. We calculated area under the curve (AUC) and risk tiers to determine predictive accuracy. NLP-derived models demonstrated strong predictive accuracy. Patients that scored within top 10% of risk model accounted for up to 29% of suicide decedents. NLP-derived model compares positively to other leading prediction methods. Our approach is highly implementable, only requiring access to text data and open-source software. Additional studies should evaluate ensemble models incorporating NLP-derived information alongside more typical structured variables.

Association of Celiac Serology Normalization With the Risk of Hypothyroidism: A Cohort Study.

We evaluated whether persistent-positive celiac serology is associated with the risk of hypothyroidism. We extracted a cohort of subjects aged 1-80 years with a positive IgA anti-tissue transglutaminase between January 1, 2008, and December 31, 2012, and a repeat anti-tissue transglutaminase test within 6-36 months from a large population-based electronic medical record database. Based on serology tests, we categorized the pediatric (age &lt;21 years) and adult cohorts into normalized or persistent-positive serology groups. All subjects were followed up for incident diagnosis of hypothyroidism from the last serology date up to December 31, 2017. Hazard ratio (HR) along 95% confidence intervals (CIs) were prepared to evaluate the association of celiac serology group with a diagnosis of hypothyroidism, crude, and adjusted for age, sex, and diagnosis of type 1 diabetes mellitus. Among the pediatric cohort (n = 2,687), during a median follow-up of 64 months (interquartile range 48-80), 2.3% (16/681) of the persistent-positive serology group and 1.0% (20/2,006) of the normalized serology group developed hypothyroidism (HR 2.07 [95% CI 1.07-4.44], adjHR 1.77 [95% CI 0.91-3.46]). The rate among the pediatric cohort with an established diagnosis of celiac disease was 3.4% (10/486) vs 1.0% (5/481), HR 2.83 (0.96-8.32). In the adult cohort (n = 1,286), 4.5% (20/442) of the persistent-positive group and 3.9% (33/811) of the normalized serology group developed hypothyroidism (HR 1.13 [95% CI 0.65-1.97]). In this retrospective, age-stratified analysis, we report that persistent-positive serology may be associated with the risk of hypothyroidism among the pediatric population. Prospective cohorts are needed to validate our findings.

Validation of diagnostic coding for interstitial lung diseases in an electronic health record system in Hong Kong.

Large electronic medical record (EMR) databases can facilitate epidemiology research into uncommon diseases such as interstitial lung disease (ILD). Given the rarity and diagnostic difficulty of ILD, the validity of the coding in EMR requires clarification. We aimed to assess the validity of International Classification of Diseases, 9th Revision (ICD-9) code algorithms for identifying ILD in the territory-wide electronic medical health record system of Clinical Data Analysis and Reporting System (CDARS) in Hong Kong. Patients who visited the Queen Mary Hospital in 2005-2018 with ILD were identified using the following ICD-9 codes: post-inflammatory pulmonary fibrosis (PPF; ICD-9: 515), idiopathic fibrosing alveolitis (IFA; ICD-9: 516.3), connective tissue disease-associated interstitial lung disease (CTD-ILD; ICD-9: 517.2, 517.8, 714.81), sarcoidosis (ICD-9: 135) and extrinsic allergic alveolitis (EAA; ICD-9: 495). A random selection was conducted in cases with diagnostic code of PPF and IFA, where a relative higher case number was identified. All the cases of CTD-ILD, sarcoidosis and EAA were included in validation for relatively small case number. Two hundred and sixty nine cases were validated using medical record review by a respiratory specialist. The overall positive predictive value (PPV) was 79% (95% CI, 74%-84%). In subgroup analysis, true positive case numbers of PPF, IFA, CTD-ILD, sarcoidosis and EAA were 74/100 (74%), 95/100 (95%), 11/15 (73%), 27/32 (84%) and 6/22 (27%), respectively. This was the first ICD-9 coding validation for ILD in Hong Kong CDARS. Our study demonstrated that using ICD-9 algorithms 515, 516.3, 517.2, 517.8, 714.81 and 135 enhanced identifications of ILDs with PPV that was reliable to support utility of CDARS database for further clinical research on ILDs. The validity is particularly high with 516.3.

Using Japanese big data to investigate novel factors and their high-risk combinations that affect vancomycin-induced nephrotoxicity.

Several factors related to vancomycin-induced nephrotoxicity (VIN) have not yet been clarified. In the present study, we used Japanese big data to investigate novel factors and their high-risk combinations that influence VIN. We employed a large Japanese electronic medical record database and included patients who had been administered intravenous vancomycin between June 2000 and December 2020. VIN was defined as an increase in serum creatinine ≥0.5mg/dL or 1.5-fold higher than the baseline. The outcomes were: (1) factors affecting VIN that were identified using multiple logistic regression analysis, and (2) combinations of factors that affect the risk of VIN according to a decision tree analysis, which is a typical machine learning method. Of the 7306 patients that were enrolled, VIN occurred in 14.2% of them (1035). A multivariate analysis extracted 22 variables as independent factors. Concomitant ramelteon use (odds ratio 0.701, 95% confidence interval 0.512-0.959), ward pharmacy service (0.741, 0.638-0.861), duration of VCM < 7days (0.748, 0.623-0.899) and trough concentrations 10-15 mg/L (0.668, 0.556-0.802) reduce the risk of VIN. Meanwhile, concomitant piperacillin-tazobactam use (2.056, 1.754-2.409) and piperacillin use (2.868, 1.298-6.338) increase the risk. The decision tree analysis showed that a combination of vancomycin trough concentrations ≥20 mg/L and concomitant piperacillin-tazobactam use was associated with the highest risk. We revealed that the concomitant ramelteon use and ward pharmacy service may decrease the risk of VIN, while the concomitant use of not only piperacillin-tazobactam but also piperacillin may increase the risk.