Comparing penalization methods for linear models on large observational health data.

OBJECTIVE: This study evaluates regularization variants in logistic regression (L1, L2, ElasticNet, Adaptive L1, Adaptive ElasticNet, Broken adaptive ridge [BAR], and Iterative hard thresholding [IHT]) for discrimination and calibration performance, focusing on both internal and external validation. MATERIALS AND METHODS: We use data from 5 US claims and electronic health record databases and develop models for various outcomes in a major depressive disorder patient population. We externally validate all models in the other databases. We use a train-test split of 75%/25% and evaluate performance with discrimination and calibration. Statistical analysis for difference in performance uses Friedman's test and critical difference diagrams. RESULTS: Of the 840 models we develop, L1 and ElasticNet emerge as superior in both internal and external discrimination, with a notable AUC difference. BAR and IHT show the best internal calibration, without a clear external calibration leader. ElasticNet typically has larger model sizes than L1. Methods like IHT and BAR, while slightly less discriminative, significantly reduce model complexity. CONCLUSION: L1 and ElasticNet offer the best discriminative performance in logistic regression for healthcare predictions, maintaining robustness across validations. For simpler, more interpretable models, L0-based methods (IHT and BAR) are advantageous, providing greater parsimony and calibration with fewer features. This study aids in selecting suitable regularization techniques for healthcare prediction models, balancing performance, complexity, and interpretability.

Prediction of 30-day, 90-day, and 1-year mortality after colorectal cancer surgery using a data-driven approach.

PURPOSE: To develop prediction models for short-term mortality risk assessment following colorectal cancer surgery. METHODS: Data was harmonized from four Danish observational health databases into the Observational Medical Outcomes Partnership Common Data Model. With a data-driven approach using the Least Absolute Shrinkage and Selection Operator logistic regression on preoperative data, we developed 30-day, 90-day, and 1-year mortality prediction models. We assessed discriminative performance using the area under the receiver operating characteristic and precision-recall curve and calibration using calibration slope, intercept, and calibration-in-the-large. We additionally assessed model performance in subgroups of curative, palliative, elective, and emergency surgery. RESULTS: A total of 57,521 patients were included in the study population, 51.1% male and with a median age of 72 years. The model showed good discrimination with an area under the receiver operating characteristic curve of 0.88, 0.878, and 0.861 for 30-day, 90-day, and 1-year mortality, respectively, and a calibration-in-the-large of 1.01, 0.99, and 0.99. The overall incidence of mortality were 4.48% for 30-day mortality, 6.64% for 90-day mortality, and 12.8% for 1-year mortality, respectively. Subgroup analysis showed no improvement of discrimination or calibration when separating the cohort into cohorts of elective surgery, emergency surgery, curative surgery, and palliative surgery. CONCLUSION: We were able to train prediction models for the risk of short-term mortality on a data set of four combined national health databases with good discrimination and calibration. We found that one cohort including all operated patients resulted in better performing models than cohorts based on several subgroups.

Real-world treatment trajectories of adults with newly diagnosed asthma or COPD.

BACKGROUND: There is a lack of knowledge on how patients with asthma or chronic obstructive pulmonary disease (COPD) are globally treated in the real world, especially with regard to the initial pharmacological treatment of newly diagnosed patients and the different treatment trajectories. This knowledge is important to monitor and improve clinical practice. METHODS: This retrospective cohort study aims to characterise treatments using data from four claims (drug dispensing) and four electronic health record (EHR; drug prescriptions) databases across six countries and three continents, encompassing 1.3 million patients with asthma or COPD. We analysed treatment trajectories at drug class level from first diagnosis and visualised these in sunburst plots. RESULTS: In four countries (USA, UK, Spain and the Netherlands), most adults with asthma initiate treatment with short-acting ß2 agonists monotherapy (20.8%-47.4% of first-line treatments). For COPD, the most frequent first-line treatment varies by country. The largest percentages of untreated patients (for asthma and COPD) were found in claims databases (14.5%-33.2% for asthma and 27.0%-52.2% for COPD) from the USA as compared with EHR databases (6.9%-15.2% for asthma and 4.4%-17.5% for COPD) from European countries. The treatment trajectories showed step-up as well as step-down in treatments. CONCLUSION: Real-world data from claims and EHRs indicate that first-line treatments of asthma and COPD vary widely across countries. We found evidence of a stepwise approach in the pharmacological treatment of asthma and COPD, suggesting that treatments may be tailored to patients' needs.

Risk of Arterial and Venous Thrombotic Events Among Patients with COVID-19: A Multi-National Collaboration of Regulatory Agencies from Canada, Europe, and United States.

Purpose: Few studies have examined how the absolute risk of thromboembolism with COVID-19 has evolved over time across different countries. Researchers from the European Medicines Agency, Health Canada, and the United States (US) Food and Drug Administration established a collaboration to evaluate the absolute risk of arterial (ATE) and venous thromboembolism (VTE) in the 90 days after diagnosis of COVID-19 in the ambulatory (eg, outpatient, emergency department, nursing facility) setting from seven countries across North America (Canada, US) and Europe (England, Germany, Italy, Netherlands, and Spain) within periods before and during COVID-19 vaccine availability. Patients and Methods: We conducted cohort studies of patients initially diagnosed with COVID-19 in the ambulatory setting from the seven specified countries. Patients were followed for 90 days after COVID-19 diagnosis. The primary outcomes were ATE and VTE over 90 days from diagnosis date. We measured country-level estimates of 90-day absolute risk (with 95% confidence intervals) of ATE and VTE. Results: The seven cohorts included 1,061,565 patients initially diagnosed with COVID-19 in the ambulatory setting before COVID-19 vaccines were available (through November 2020). The 90-day absolute risk of ATE during this period ranged from 0.11% (0.09-0.13%) in Canada to 1.01% (0.97-1.05%) in the US, and the 90-day absolute risk of VTE ranged from 0.23% (0.21-0.26%) in Canada to 0.84% (0.80-0.89%) in England. The seven cohorts included 3,544,062 patients with COVID-19 during vaccine availability (beginning December 2020). The 90-day absolute risk of ATE during this period ranged from 0.06% (0.06-0.07%) in England to 1.04% (1.01-1.06%) in the US, and the 90-day absolute risk of VTE ranged from 0.25% (0.24-0.26%) in England to 1.02% (0.99-1.04%) in the US. Conclusion: There was heterogeneity by country in 90-day absolute risk of ATE and VTE after ambulatory COVID-19 diagnosis both before and during COVID-19 vaccine availability.

Blood pressure measurements for diagnosing hypertension in primary care: room for improvement.

BACKGROUND: In the adult population, about 50% have hypertension, a risk factor for cardiovascular disease and subsequent premature death. Little is known about the quality of the methods used to diagnose hypertension in primary care. OBJECTIVES: The objective was to assess the frequency of use of recognized methods to establish a diagnosis of hypertension, and specifically for OBPM, whether three distinct measurements were taken, and how correctly the blood pressure levels were interpreted. METHODS: A retrospective population-based cohort study using electronic medical records of patients aged between 40 and 70 years, who visited their general practitioner (GP) with a new-onset of hypertension in the years 2012, 2016, 2019, and 2020. A visual chart review of the electronic medical records was used to assess the methods employed to diagnose hypertension in a random sample of 500 patients. The blood pressure measurement method was considered complete if three or more valid office blood pressure measurements (OBPM) were performed, or home-based blood pressure measurements (HBPM), the office- based 30-minute method (OBP30), or 24-hour ambulatory blood pressure measurements (24 H-ABPM) were used. RESULTS: In all study years, OBPM was the most frequently used method to diagnose new-onset hypertension in patients. The OBP-30 method was used in 0.4% (2012), 4.2% (2016), 10.6% (2019), and 9.8% (2020) of patients respectively, 24 H-ABPM in 16.0%, 22.2%, 17.2%, and 19.0% of patients and HBPM measurements in 5.4%, 8.4%, 7.6%, and 7.8% of patients, respectively. A diagnosis of hypertension based on only one or two office measurements occurred in 85.2% (2012), 87.9% (2016), 94.4% (2019), and 96.8% (2020) of all patients with OBPM. In cases of incomplete measurement and incorrect interpretation, medication was still started in 64% of cases in 2012, 56% (2016), 60% (2019), and 73% (2020). CONCLUSION: OBPM is still the most often used method to diagnose hypertension in primary care. The diagnosis was often incomplete or misinterpreted using incorrect cut-off levels. A small improvement occurred between 2012 and 2016 but no further progress was seen in 2019 or 2020. If hypertension is inappropriately diagnosed, it may result in under treatment or in prolonged, unnecessary treatment of patients. There is room for improvement in the general practice setting.

Health-Analytics Data to Evidence Suite (HADES): Open-Source Software for Observational Research.

The Health-Analytics Data to Evidence Suite (HADES) is an open-source software collection developed by Observational Health Data Sciences and Informatics (OHDSI). It executes directly against healthcare data such as electronic health records and administrative claims, that have been converted to the Observational Medical Outcomes Partnership (OMOP) Common Data Model. Using advanced analytics, HADES performs characterization, population-level causal effect estimation, and patient-level prediction, potentially across a federated data network, allowing patient-level data to remain locally while only aggregated statistics are shared. Designed to run across a wide array of technical environments, including different operating systems and database platforms, HADES uses continuous integration with a large set of unit tests to maintain reliability. HADES implements OHDSI best practices, and is used in almost all published OHDSI studies, including some that have directly informed regulatory decisions.

OHDSI Standardized Vocabularies-a large-scale centralized reference ontology for international data harmonization.

IMPORTANCE: The Observational Health Data Sciences and Informatics (OHDSI) is the largest distributed data network in the world encompassing more than 331 data sources with 2.1 billion patient records across 34 countries. It enables large-scale observational research through standardizing the data into a common data model (CDM) (Observational Medical Outcomes Partnership [OMOP] CDM) and requires a comprehensive, efficient, and reliable ontology system to support data harmonization. MATERIALS AND METHODS: We created the OHDSI Standardized Vocabularies-a common reference ontology mandatory to all data sites in the network. It comprises imported and de novo-generated ontologies containing concepts and relationships between them, and the praxis of converting the source data to the OMOP CDM based on these. It enables harmonization through assigned domains according to clinical categories, comprehensive coverage of entities within each domain, support for commonly used international coding schemes, and standardization of semantically equivalent concepts. RESULTS: The OHDSI Standardized Vocabularies comprise over 10 million concepts from 136 vocabularies. They are used by hundreds of groups and several large data networks. More than 8600 users have performed 50 000 downloads of the system. This open-source resource has proven to address an impediment of large-scale observational research-the dependence on the context of source data representation. With that, it has enabled efficient phenotyping, covariate construction, patient-level prediction, population-level estimation, and standard reporting. DISCUSSION AND CONCLUSION: OHDSI has made available a comprehensive, open vocabulary system that is unmatched in its ability to support global observational research. We encourage researchers to exploit it and contribute their use cases to this dynamic resource.

A High-Fidelity Combined ATC-Rxnorm Drug Hierarchy for Large-Scale Observational Research.

Observational research utilizes patient information from many disparate databases worldwide. To be able to systematically analyze data and compare the results of such research studies, information about exposure to drugs or classes of drugs needs to be harmonized across these data. The NLM's RxNorm drug terminology and WHO's ATC classification serve these needs but are currently not satisfactorily combined into a common system. Creating such system is hampered by a number of challenges, resulting from different approaches to representing attributes of drugs and ontological rules. Here, we present a combined ATC-RxNorm drug hierarchy, allowing to use ATC classes for retrieval of drug information in large scale observational data. We present the heuristic for maintaining this resource and evaluate it in a real world database containing drug and drug classification information.

IncidencePrevalence: An R package to calculate population-level incidence rates and prevalence using the OMOP common data model.

PURPOSE: Real-world data (RWD) offers a valuable resource for generating population-level disease epidemiology metrics. We aimed to develop a well-tested and user-friendly R package to compute incidence rates and prevalence in data mapped to the observational medical outcomes partnership (OMOP) common data model (CDM). MATERIALS AND METHODS: We created IncidencePrevalence, an R package to support the analysis of population-level incidence rates and point- and period-prevalence in OMOP-formatted data. On top of unit testing, we assessed the face validity of the package. To do so, we calculated incidence rates of COVID-19 using RWD from Spain (SIDIAP) and the United Kingdom (CPRD Aurum), and replicated two previously published studies using data from the Netherlands (IPCI) and the United Kingdom (CPRD Gold). We compared the obtained results to those previously published, and measured execution times by running a benchmark analysis across databases. RESULTS: IncidencePrevalence achieved high agreement to previously published data in CPRD Gold and IPCI, and showed good performance across databases. For COVID-19, incidence calculated by the package was similar to public data after the first-wave of the pandemic. CONCLUSION: For data mapped to the OMOP CDM, the IncidencePrevalence R package can support descriptive epidemiological research. It enables reliable estimation of incidence and prevalence from large real-world data sets. It represents a simple, but extendable, analytical framework to generate estimates in a reproducible and timely manner.

Descriptive analysis on disproportionate medication errors and associated patient characteristics in the Food and Drug Administration's Adverse Event Reporting System.

BACKGROUND: Medication errors (MEs) are a major public health concern which can cause harm and financial burden within the healthcare system. Characterizing MEs is crucial to develop strategies to mitigate MEs in the future. OBJECTIVES: To characterize ME-associated reports, and investigate signals of disproportionate reporting (SDRs) on MEs in the Food and Drug Administration's Adverse Event Reporting System (FAERS). METHODS: FAERS data from 2004 to 2020 was used. ME reports were identified with the narrow Standardised Medical Dictionary for Regulatory Activities® (MedDRA®) Query (SMQ) for MEs. Drug names were converted to the Anatomical Therapeutic Chemical (ATC) classification. SDRs were investigated using the reporting odds ratio (ROR). RESULTS: In total 488 470 ME reports were identified, mostly (59%) submitted by consumers and mainly (55%) associated with females. Median age at time of ME was 57 years (interquartile range: 37-70 years). Approximately 1 out of 3 reports stated a serious health outcome. The most prevalent reported drug class was "antineoplastic and immunomodulating agents" (25%). The most common ME type was "incorrect dose administered" (9%). Of the 1659 SDRs obtained, adalimumab was the most common drug associated with MEs, noting a ROR of 1.22 (95% confidence interval: 1.21-1.24). CONCLUSION: This study offers a first of its kind characterization of MEs as reported to FAERS. Reported MEs are frequent and may be associated with serious health outcomes. This FAERS data provides insights on ME prevention and offers possibilities for additional in-depth analyses.

Attention-based neural networks for clinical prediction modelling on electronic health records.

BACKGROUND: Deep learning models have had a lot of success in various fields. However, on structured data they have struggled. Here we apply four state-of-the-art supervised deep learning models using the attention mechanism and compare against logistic regression and XGBoost using discrimination, calibration and clinical utility. METHODS: We develop the models using a general practitioners database. We implement a recurrent neural network, a transformer with and without reverse distillation and a graph neural network. We measure discrimination using the area under the receiver operating characteristic curve (AUC) and the area under the precision recall curve (AUPRC). We assess smooth calibration using restricted cubic splines and clinical utility with decision curve analysis. RESULTS: Our results show that deep learning approaches can improve discrimination up to 2.5% points AUC and 7.4% points AUPRC. However, on average the baselines are competitive. Most models are similarly calibrated as the baselines except for the graph neural network. The transformer using reverse distillation shows the best performance in clinical utility on two out of three prediction problems over most of the prediction thresholds. CONCLUSION: In this study, we evaluated various approaches in supervised learning using neural networks and attention. Here we do a rigorous comparison, not only looking at discrimination but also calibration and clinical utility. There is value in using deep learning models on electronic health record data since it can improve discrimination and clinical utility while providing good calibration. However, good baseline methods are still competitive.

Evaluating the impact of alternative phenotype definitions on incidence rates across a global data network.

Objective: Developing accurate phenotype definitions is critical in obtaining reliable and reproducible background rates in safety research. This study aims to illustrate the differences in background incidence rates by comparing definitions for a given outcome. Materials and Methods: We used 16 data sources to systematically generate and evaluate outcomes for 13 adverse events and their overall background rates. We examined the effect of different modifications (inpatient setting, standardization of code set, and code set changes) to the computable phenotype on background incidence rates. Results: Rate ratios (RRs) of the incidence rates from each computable phenotype definition varied across outcomes, with inpatient restriction showing the highest variation from 1 to 11.93. Standardization of code set RRs ranges from 1 to 1.64, and code set changes range from 1 to 2.52. Discussion: The modification that has the highest impact is requiring inpatient place of service, leading to at least a 2-fold higher incidence rate in the base definition. Standardization showed almost no change when using source code variations. The strength of the effect in the inpatient restriction is highly dependent on the outcome. Changing definitions from broad to narrow showed the most variability by age/gender/database across phenotypes and less than a 2-fold increase in rate compared to the base definition. Conclusion: Characterization of outcomes across a network of databases yields insights into sensitivity and specificity trade-offs when definitions are altered. Outcomes should be thoroughly evaluated prior to use for background rates for their plausibility for use across a global network.

Parasitic infections related to anti-type 2 immunity monoclonal antibodies: a disproportionality analysis in the food and drug administration's adverse event reporting system (FAERS).

Introduction: Monoclonal antibodies (mAbs) targeting immunoglobulin E (IgE) [omalizumab], type 2 (T2) cytokine interleukin (IL) 5 [mepolizumab, reslizumab], IL-4 Receptor (R) α [dupilumab], and IL-5R [benralizumab]), improve quality of life in patients with T2-driven inflammatory diseases. However, there is a concern for an increased risk of helminth infections. The aim was to explore safety signals of parasitic infections for omalizumab, mepolizumab, reslizumab, dupilumab, and benralizumab. Methods: Spontaneous reports were used from the Food and Drug Administration's Adverse Event Reporting System (FAERS) database from 2004 to 2021. Parasitic infections were defined as any type of parasitic infection term obtained from the Standardised Medical Dictionary for Regulatory Activities® (MedDRA®). Safety signal strength was assessed by the Reporting Odds Ratio (ROR). Results: 15,502,908 reports were eligible for analysis. Amongst 175,888 reports for omalizumab, mepolizumab, reslizumab, dupilumab, and benralizumab, there were 79 reports on parasitic infections. Median age was 55 years (interquartile range 24-63 years) and 59.5% were female. Indications were known in 26 (32.9%) reports; 14 (53.8%) biologicals were reportedly prescribed for asthma, 8 (30.7%) for various types of dermatitis, and 2 (7.6%) for urticaria. A safety signal was observed for each biological, except for reslizumab (due to lack of power), with the strongest signal attributed to benralizumab (ROR = 15.7, 95% Confidence Interval: 8.4-29.3). Conclusion: Parasitic infections were disproportionately reported for mAbs targeting IgE, T2 cytokines, or T2 cytokine receptors. While the number of adverse event reports on parasitic infections in the database was relatively low, resulting safety signals were disproportionate and warrant further investigation.

European Health Data & Evidence Network-learnings from building out a standardized international health data network.

OBJECTIVE: Health data standardized to a common data model (CDM) simplifies and facilitates research. This study examines the factors that make standardizing observational health data to the Observational Medical Outcomes Partnership (OMOP) CDM successful. MATERIALS AND METHODS: Twenty-five data partners (DPs) from 11 countries received funding from the European Health Data Evidence Network (EHDEN) to standardize their data. Three surveys, DataQualityDashboard results, and statistics from the conversion process were analyzed qualitatively and quantitatively. Our measures of success were the total number of days to transform source data into the OMOP CDM and participation in network research. RESULTS: The health data converted to CDM represented more than 133 million patients. 100%, 88%, and 84% of DPs took Surveys 1, 2, and 3. The median duration of the 6 key extract, transform, and load (ETL) processes ranged from 4 to 115 days. Of the 25 DPs, 21 DPs were considered applicable for analysis of which 52% standardized their data on time, and 48% participated in an international collaborative study. DISCUSSION: This study shows that the consistent workflow used by EHDEN proves appropriate to support the successful standardization of observational data across Europe. Over the 25 successful transformations, we confirmed that getting the right people for the ETL is critical and vocabulary mapping requires specific expertise and support of tools. Additionally, we learned that teams that proactively prepared for data governance issues were able to avoid considerable delays improving their ability to finish on time. CONCLUSION: This study provides guidance for future DPs to standardize to the OMOP CDM and participate in distributed networks. We demonstrate that the Observational Health Data Sciences and Informatics community must continue to evaluate and provide guidance and support for what ultimately develops the backbone of how community members generate evidence.

Drug Utilisation Patterns of Alternatives to Ranitidine-Containing Medicines in Patients Treated with Ranitidine: A Network Analysis of Data from Six European National Databases.

INTRODUCTION: Ranitidine, a histamine H2-receptor antagonist (H2RA), is indicated in the management of gastric acid-related disorders. In 2020, the European Medicines Agency (EMA) recommended suspension of all ranitidine-containing medicines in the European Union (EU) due to the presence of N-nitrosodimethylamine (NDMA) impurities, which were considered to be carcinogenic. The aim of this study was to investigate the impact of regulatory intervention on use patterns of ranitidine-containing medicines and their therapeutic alternatives. OBJECTIVES: The aim was to study drug utilisation patterns of ranitidine and report discernible trends in treatment discontinuation and switching to alternative medications. METHODS: This retrospective, population-based cohort study was conducted using primary care records from six European countries between 2017 and 2023. To explore drug utilisation patterns, we calculated (1) incident use of ranitidine, other H2RAs, and other alternative drugs for the treatment of gastric ulcer and/or gastric bleeding; (2) ranitidine discontinuation; and (3) switching from ranitidine to alternative drugs (H2RAs, proton-pump inhibitors [PPIs], and other medicinal products for acid-related disorders). RESULTS: During the study period, 385,273 new ranitidine users were observed, with most users being female and aged 18-74 years. Ranitidine was the most commonly prescribed H2RA in the pre-referral period (September 2017-August 2019), with incidence rates between 0.8 and 9.0/1000 person years (PY). A steep decline to 0.3-3.8/1000 PY was observed in the referral period (September 2019-March 2020), eventually dropping to 0.0-0.4/1000 PY in the post-referral period (April 2020-March 2022). Switching from ranitidine to alternative drugs increased in the post-referral period, with the majority of patients switching to PPIs. Discontinuation of ranitidine use ranged from 270 to 380/1000 users in 2017 and decreased over time. CONCLUSIONS: Ranitidine was commonly used prior to referral, but it was subsequently discontinued and replaced primarily with PPIs.

Supporting Pharmacovigilance Signal Validation and Prioritization with Analyses of Routinely Collected Health Data: Lessons Learned from an EHDEN Network Study.

INTRODUCTION: Individual case reports are the main asset in pharmacovigilance signal management. Signal validation is the first stage after signal detection and aims to determine if there is sufficient evidence to justify further assessment. Throughout signal management, a prioritization of signals is continually made. Routinely collected health data can provide relevant contextual information but are primarily used at a later stage in pharmacoepidemiological studies to assess communicated signals. OBJECTIVE: The aim of this study was to examine the feasibility and utility of analysing routine health data from a multinational distributed network to support signal validation and prioritization and to reflect on key user requirements for these analyses to become an integral part of this process. METHODS: Statistical signal detection was performed in VigiBase, the WHO global database of individual case safety reports, targeting generic manufacturer drugs and 16 prespecified adverse events. During a 5-day study-a-thon, signal validation and prioritization were performed using information from VigiBase, regulatory documents and the scientific literature alongside descriptive analyses of routine health data from 10 partners of the European Health Data and Evidence Network (EHDEN). Databases included in the study were from the UK, Spain, Norway, the Netherlands and Serbia, capturing records from primary care and/or hospitals. RESULTS: Ninety-five statistical signals were subjected to signal validation, of which eight were considered for descriptive analyses in the routine health data. Design, execution and interpretation of results from these analyses took up to a few hours for each signal (of which 15-60 minutes were for execution) and informed decisions for five out of eight signals. The impact of insights from the routine health data varied and included possible alternative explanations, potential public health and clinical impact and feasibility of follow-up pharmacoepidemiological studies. Three signals were selected for signal assessment, two of these decisions were supported by insights from the routine health data. Standardization of analytical code, availability of adverse event phenotypes including bridges between different source vocabularies, and governance around the access and use of routine health data were identified as important aspects for future development. CONCLUSIONS: Analyses of routine health data from a distributed network to support signal validation and prioritization are feasible in the given time limits and can inform decision making. The cost-benefit of integrating these analyses at this stage of signal management requires further research.

Transforming the Information System for Research in Primary Care (SIDIAP) in Catalonia to the OMOP Common Data Model and Its Use for COVID-19 Research.

Purpose: The primary aim of this work was to convert the Information System for Research in Primary Care (SIDIAP) from Catalonia, Spain, to the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM). Our second aim was to provide a descriptive analysis of COVID-19-related outcomes among the general population. Patients and Methods: We mapped patient-level data from SIDIAP to the OMOP CDM and we performed more than 3,400 data quality checks to assess its readiness for research. We established a general population cohort as of the 1st March 2020 and identified outpatient COVID-19 diagnoses or tested positive for, hospitalised with, admitted to intensive care units (ICU) with, died with, or vaccinated against COVID-19 up to 30th June 2022. Results: After verifying the high quality of the transformed dataset, we included 5,870,274 individuals in the general population cohort. Of those, 604,472 had either an outpatient COVID-19 diagnosis or positive test result, 58,991 had a hospitalisation, 5,642 had an ICU admission, and 11,233 died with COVID-19. A total of 4,584,515 received a COVID-19 vaccine. People who were hospitalised or died were more commonly older, male, and with more comorbidities. Those admitted to ICU with COVID-19 were generally younger and more often male than those hospitalised and those who died. Conclusion: We successfully transformed SIDIAP to the OMOP CDM. From this dataset, a general population cohort of 5.9 million individuals was identified and their COVID-19-related outcomes over time were described. The transformed SIDIAP database is a valuable resource that can enable distributed network research in COVID-19 and beyond.

The added value of text from Dutch general practitioner notes in predictive modeling.

OBJECTIVE: This work aims to explore the value of Dutch unstructured data, in combination with structured data, for the development of prognostic prediction models in a general practitioner (GP) setting. MATERIALS AND METHODS: We trained and validated prediction models for 4 common clinical prediction problems using various sparse text representations, common prediction algorithms, and observational GP electronic health record (EHR) data. We trained and validated 84 models internally and externally on data from different EHR systems. RESULTS: On average, over all the different text representations and prediction algorithms, models only using text data performed better or similar to models using structured data alone in 2 prediction tasks. Additionally, in these 2 tasks, the combination of structured and text data outperformed models using structured or text data alone. No large performance differences were found between the different text representations and prediction algorithms. DISCUSSION: Our findings indicate that the use of unstructured data alone can result in well-performing prediction models for some clinical prediction problems. Furthermore, the performance improvement achieved by combining structured and text data highlights the added value. Additionally, we demonstrate the significance of clinical natural language processing research in languages other than English and the possibility of validating text-based prediction models across various EHR systems. CONCLUSION: Our study highlights the potential benefits of incorporating unstructured data in clinical prediction models in a GP setting. Although the added value of unstructured data may vary depending on the specific prediction task, our findings suggest that it has the potential to enhance patient care.

Challenges of Estimating Global Feature Importance in Real-World Health Care Data.

Feature importance is often used to explain clinical prediction models. In this work, we examine three challenges using experiments with electronic health record data: computational feasibility, choosing between methods, and interpretation of the resulting explanation. This work aims to create awareness of the disagreement between feature importance methods and underscores the need for guidance to practitioners how to deal with these discrepancies.

Does Using a Stacking Ensemble Method to Combine Multiple Base Learners Within a Database Improve Model Transportability?

We investigated a stacking ensemble method that combines multiple base learners within a database. The results on external validation across four large databases suggest a stacking ensemble could improve model transportability.