Search

Introduction: Hip fractures among older people are a major public health issue, which can impact quality of life and increase mortality within the year after they occur. A recent observational study found an increased risk of hip fracture in subjects who were new users of tramadol compared with codeine. These drugs have somewhat different indications. Tramadol is indicated for moderate to severe pain and can be used for an extended period; codeine is indicated for mild to moderate pain and cough suppression. Objective: In this observational study, we compared the risk of hip fracture in new users of tramadol or codeine, using multiple databases and analytical methods. Methods: Using data from the Clinical Practice Research Datalink and three US claims databases, we compared the risk of hip fracture after exposure to tramadol or codeine in subjects aged 50–89 years. To ensure comparability, large-scale propensity scores were used to adjust for confounding. Results: We observed a calibrated hazard ratio of 1.10 (95% calibrated confidence interval 0.99–1.21) in the Clinical Practice Research Datalink database, and a pooled estimate across the US databases yielded a calibrated hazard ratio of 1.06 (95% calibrated confidence interval 0.97–1.16). Conclusions: Our results did not demonstrate a statistically significant difference between subjects treated for pain with tramadol compared with codeine for the outcome of hip fracture risk.

Objective To develop a lossless distributed algorithm for generalized linear mixed model (GLMM) with application to privacy-preserving hospital profiling. Materials and Methods The GLMM is often fitted to implement hospital profiling, using clinical or administrative claims data. Due to individual patient data (IPD) privacy regulations and the computational complexity of GLMM, a distributed algorithm for hospital profiling is needed. We develop a novel distributed penalized quasi-likelihood (dPQL) algorithm to fit GLMM when only aggregated data, rather than IPD, can be shared across hospitals. We also show that the standardized mortality rates, which are often reported as the results of hospital profiling, can also be calculated distributively without sharing IPD. We demonstrate the applicability of the proposed dPQL algorithm by ranking 929 hospitals for coronavirus disease 2019 (COVID-19) mortality or referral to hospice that have been previously studied. Results The proposed dPQL algorithm is mathematically proven to be lossless, that is, it obtains identical results as if IPD were pooled from all hospitals. In the example of hospital profiling regarding COVID-19 mortality, the dPQL algorithm reached convergence with only 5 iterations, and the estimation of fixed effects, random effects, and mortality rates were identical to that of the PQL from pooled data. Conclusion The dPQL algorithm is lossless, privacy-preserving and fast-converging for fitting GLMM. It provides an extremely suitable and convenient distributed approach for hospital profiling.

ObjectivesTo assess the uptake of second-line antihyperglycemic agents among patients with type-2 diabetes mellitus (T2DM) receiving metformin.DesignSerial cross-sectional study (2011-2021).SettingTen US and seven non-US electronic health record and administrative claims databases in the Observational Health Data Sciences and Informatics network.Participants4.8 million patients with T2DM receiving metformin.Main Outcomes MeasuresCalendar-year trends in the proportional initiation of second-line antihyperglycemic agents, glucagon-like peptide-1 receptor agonists (GLP-1 RAs), sodium-glucose cotransporter 2 inhibitors (SGLT2is), dipeptidyl peptidase-4 inhibitors, and sulfonylureas, for each database. We also evaluated the relative drug class-level uptake across cardiovascular risk groups.ResultsWe identified 4.6 million patients with T2DM in US databases, 61,382 from Spain, 32,442 from Germany, 25,173 from the UK, 13,270 from France, 5,580 from Scotland, 4,614 from Hong Kong, and 2,322 from Australia. During 2011-2021, the combined proportional initiation of cardioprotective antihyperglycemic agents, GLP-1 RAs and SGLT2is, increased across all data sources, with the combined initiation of these drugs as second-line agents in 2021 ranging from 35.2% to 68.2% in the US databases, 15.4% in France, 34.7% in Spain, 50.1% in Germany, and 54.8% in Scotland. From 2016 to 2021, in some US and non-US databases, uptake of GLP-1 RAs and SGLT2is increased more significantly among populations without cardiovascular disease compared to those with established cardiovascular disease, without any data source providing evidence of a greater increase in their uptake in the populations with cardiovascular disease.ConclusionsDespite the increase in overall uptake of cardioprotective antihyperglycemic agents as second-line treatment for T2DM, their uptake was lower in patients with cardiovascular disease over the last decade. A strategy to ensure medication use concordant with guideline recommendations is essential to improve outcomes of patients with T2DM.

Post-approval safety surveillance of medical products using observational healthcare data can help identify safety issues beyond those found in pre-approval trials. When testing sequentially as data accrue, maximum sequential probability ratio testing (MaxSPRT) is a common approach to maintaining nominal type 1 error. However, the true type 1 error may still deviate from the specified one because of systematic error due to the observational nature of the analysis. This systematic error may persist even after controlling for known confounders. Here we propose to address this issue by combing MaxSPRT with empirical calibration. In empirical calibration, we assume uncertainty about the systematic error in our analysis, the source of uncertainty commonly overlooked in practice. We infer a probability distribution of systematic error by relying on a large set of negative controls: exposure-outcome where no causal effect is believed to exist. Integrating this distribution into our test statistics has previously been shown to restore type 1 error to nominal. Here we show how we can calibrate the critical value central to MaxSPRT. We evaluate this novel approach using simulations and real electronic health records, using H1N1 vaccinations during the 2009-2010 season as an example. Results show that combining empirical calibration with MaxSPRT restores nominal type 1 error. In our real-world example, adjusting for systematic error using empirical calibration has a larger impact than, and hence is just as essential as, adjusting for sequential testing using MaxSPRT. We recommend performing both, using the method described here., Supplemental Materials are available at https://github.com/ohdsi-studies/Eumaeus/tree/main/extras/EmpiricalCalibrationMaxSprtSuppl

Evidence for the effectiveness and safety of the third-generation beta-blockers other than atenolol in hypertension remains scarce. We assessed the effectiveness and safety of beta-blockers as first-line treatment for hypertension using three databases in the United States: two administrative claim databases and one electronic health record-based database from 2001 to 2018. In each database, comparative effectiveness of beta-blockers for the risks of acute myocardial infarction, stroke, and hospitalization for heart failure was assessed, using large-scale propensity adjustment and empirical calibration. Estimates were combined across databases using random-effects meta-analyses. Overall, 118,133 and 267,891 patients initiated third-generation beta-blockers (carvedilol and nebivolol) or atenolol, respectively. The pooled hazard ratios of acute myocardial infarction, stroke, hospitalization for heart failure, and most metabolic complications were not different between the third-generation beta-blockers versus atenolol after propensity score matching and empirical calibration (hazard ratio 1.07, 95% CI 0.74 to 1.55 for acute myocardial infarction; hazard ratio 1.06, 95% CI 0.87 to 1.31 for stroke; hazard ratio 1.46, 95% CI 0.99 to 2.24 for hospitalized heart failure). Third-generation beta-blockers were associated with significantly higher risk of stroke than angiotensin-converting enzyme inhibitors (hazard ratio 1.29, 95% CI 1.03 to 1.72), and thiazide diuretics (hazard ratio 1.56, 95% CI 1.17 to 2.20). In conclusion, this study found many patients with first-line beta-blocker monotherapy for hypertension and no statistically significant differences in the effectiveness and safety comparing atenolol with third-generation beta-blockers. Patients on third-generation beta-blockers had a higher risk of stroke than those on angiotensin-converting enzyme inhibitors and thiazide diuretics.

Background Angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) have been postulated to affect susceptibility to COVID-19. Observational studies so far have lacked rigorous ascertainment adjustment and international generalisability. We aimed to determine whether use of ACEIs or ARBs is associated with an increased susceptibility to COVID-19 in patients with hypertension. Methods In this international, open science, cohort analysis, we used electronic health records from Spain (Information Systems for Research in Primary Care [SIDIAP]) and the USA (Columbia University Irving Medical Center data warehouse [CUIMC] and Department of Veterans Affairs Observational Medical Outcomes Partnership [VA-OMOP]) to identify patients aged 18 years or older with at least one prescription for ACEIs and ARBs (target cohort) or calcium channel blockers (CCBs) and thiazide or thiazide-like diuretics (THZs; comparator cohort) between Nov 1, 2019, and Jan 31, 2020. Users were defined separately as receiving either monotherapy with these four drug classes, or monotherapy or combination therapy (combination use) with other antihypertensive medications. We assessed four outcomes: COVID-19 diagnosis; hospital admission with COVID-19; hospital admission with pneumonia; and hospital admission with pneumonia, acute respiratory distress syndrome, acute kidney injury, or sepsis. We built large-scale propensity score methods derived through a data-driven approach and negative control experiments across ten pairwise comparisons, with results meta-analysed to generate 1280 study effects. For each study effect, we did negative control outcome experiments using a possible 123 controls identified through a data-rich algorithm. This process used a set of predefined baseline patient characteristics to provide the most accurate prediction of treatment and balance among patient cohorts across characteristics. The study is registered with the EU Post-Authorisation Studies register, EUPAS35296. Findings Among 1 355 349 antihypertensive users (363 785 ACEI or ARB monotherapy users, 248 915 CCB or THZ monotherapy users, 711 799 ACEI or ARB combination users, and 473 076 CCB or THZ combination users) included in analyses, no association was observed between COVID-19 diagnosis and exposure to ACEI or ARB monotherapy versus CCB or THZ monotherapy (calibrated hazard ratio [HR] 0·98, 95% CI 0·84–1·14) or combination use exposure (1·01, 0·90–1·15). ACEIs alone similarly showed no relative risk difference when compared with CCB or THZ monotherapy (HR 0·91, 95% CI 0·68–1·21; with heterogeneity of >40%) or combination use (0·95, 0·83–1·07). Directly comparing ACEIs with ARBs demonstrated a moderately lower risk with ACEIs, which was significant with combination use (HR 0·88, 95% CI 0·79–0·99) and non-significant for monotherapy (0·85, 0·69–1·05). We observed no significant difference between drug classes for risk of hospital admission with COVID-19, hospital admission with pneumonia, or hospital admission with pneumonia, acute respiratory distress syndrome, acute kidney injury, or sepsis across all comparisons. Interpretation No clinically significant increased risk of COVID-19 diagnosis or hospital admission-related outcomes associated with ACEI or ARB use was observed, suggesting users should not discontinue or change their treatment to decrease their risk of COVID-19.

The use of antipsychotic medications (APMs) could be different among countries due to availability, approved indications, characteristics and clinical practice. However, there is limited literature providing comparisons of APMs use among countries. To examine trends in antipsychotic prescribing in Taiwan, Hong Kong, Japan, and the United States, we conducted a cross-national study from 2002 to 2014 b y using the distributed network approach with common data model. We included all patients who had at least a record of antipsychotic prescription in this study, and defined patients without previous exposure of antipsychotics for 6 months before the index date as new users for incidence estimation. We calculated the incidence, prevalence, and prescription rate of each medication by calendar year. Among older patients, sulpiride was the most incident [incidence rate (IR) 11.0–23.3) and prevalent [prevalence rate (PR) 11.9–14.3) APM in Taiwan, and most prevalent (PR 2.5–3.9) in Japan. Quetiapine and haloperidol were most common in the United States (IR 8.1–9.5; PR 18.0–18.4) and Hong Kong (PR 8.8–13.7; PR 10.6–12.7), respectively. The trend of quetiapine use was increasing in Taiwan, Hong Kong and the United States. As compared to older patients, the younger patients had more propensity to be prescribed second-generation APM for treatment in four countries. Trends in antipsychotic prescribing varied among countries. Quetiapine use was most prevalent in the United States and increasing in Taiwan and Hong Kong. The increasing use of quetiapine in the elderly patients might be due to its safety profile compared to other APMs.

Evidence derived from existing health-care data, such as administrative claims and electronic health records, can fill evidence gaps in medicine. However, many claim such data cannot be used to estimate causal treatment effects because of the potential for observational study bias; for example, due to residual confounding. Other concerns include P hacking and publication bias. In response, the Observational Health Data Sciences and Informatics international collaborative launched the Large-scale Evidence Generation and Evaluation across a Network of Databases (LEGEND) research initiative. Its mission is to generate evidence on the effects of medical interventions using observational health-care databases while addressing the aforementioned concerns by following a recently proposed paradigm. We define 10 principles of LEGEND that enshrine this new paradigm, prescribing the generation and dissemination of evidence on many research questions at once; for example, comparing all treatments for a disease for many outcomes, thus preventing publication bias. These questions are answered using a prespecified and systematic approach, avoiding P hacking. Best-practice statistical methods address measured confounding, and control questions (research questions where the answer is known) quantify potential residual bias. Finally, the evidence is generated in a network of databases to assess consistency by sharing open-source analytics code to enhance transparency and reproducibility, but without sharing patient-level information. Here we detail the LEGEND principles and provide a generic overview of a LEGEND study. Our companion paper highlights an example study on the effects of hypertension treatments, and evaluates the internal and external validity of the evidence we generate.

Introduction Observational studies estimating severe outcomes for paracetamol versus ibuprofen use have acknowledged the specific challenge of channeling bias. A previous study relying on negative controls suggested that using large-scale propensity score (LSPS) matching may mitigate bias better than models using limited lists of covariates. Objective The aim was to assess whether using LSPS matching would enable the evaluation of paracetamol, compared to ibuprofen, and increased risk of myocardial infarction, stroke, gastrointestinal (GI) bleeding, or acute renal failure. Study design and setting In a new-user cohort study, we used two propensity score model strategies for confounder controls. One replicated the approach of controlling for a hand-picked list. The second used LSPSs based on all available covariates for matching. Positive and negative controls assessed residual confounding and calibrated confidence intervals. The data source was the Clinical Practices Research Datalink (CPRD). Results A substantial proportion of negative controls were statistically significant after propensity score matching on the publication covariates, indicating considerable systematic error. LSPS adjustment was less biased, but residual error remained. The calibrated estimates resulted in very wide confidence intervals, indicating large uncertainty in effect estimates once residual error was incorporated. Conclusions For paracetamol versus ibuprofen, when using LSPS methods in the CPRD, it is only possible to distinguish true effects if those effects are large (hazard ratio > 2). Due to their smaller hazard ratios, the outcomes under study cannot be differentiated from null effects (represented by negative controls) even if there were a true effect. Based on these data, we conclude that we are unable to determine whether paracetamol is associated with an increased risk of myocardial infarction, stroke, GI bleeding, and acute renal failure compared to ibuprofen, due to residual confounding. Electronic supplementary material The online version of this article (10.1007/s40264-020-00950-3) contains supplementary material, which is available to authorized users.

Objective: Observational evidence suggests that patients with type 2 diabetes mellitus (T2DM) are at increased risk for acute pancreatitis (AP) versus those without T2DM. A small number of AP events were reported in clinical trials of the sodium glucose co-transporter 2 inhibitor canagliflozin, though no imbalances were observed between treatment groups. This observational study evaluated risk of AP among new users of canagliflozin compared with new users of six classes of other antihyperglycemic agents (AHAs). Methods: Three US claims databases were analyzed based on a prespecified protocol approved by the European Medicines Agency. Propensity score adjustment controlled for imbalances in baseline covariates. Cox regression models estimated the hazard ratio of AP with canagliflozin compared with other AHAs using on-treatment (primary) and intent-to-treat approaches. Sensitivity analyses assessed robustness of findings. Results: Across the three databases, there were between 12,023–80,986 new users of canagliflozin; the unadjusted incidence rates of AP (per 1000 person-years) were between 1.5–2.2 for canagliflozin and 1.1–6.6 for other AHAs. The risk of AP was generally similar for new users of canagliflozin compared with new users of glucagon-like peptide-1 receptor agonists, dipeptidyl peptidase-4 inhibitors, sulfonylureas, thiazolidinediones, insulin, and other AHAs, with no consistent between-treatment differences observed across databases. Intent-to-treat and sensitivity analysis findings were qualitatively consistent with on-treatment findings. Conclusions: In this large observational study, incidence rates of AP in patients with T2DM treated with canagliflozin or other AHAs were generally similar, with no evidence suggesting that canagliflozin is associated with increased risk of AP compared with other AHAs.

BackgroundVaccine safety surveillance commonly includes a serial testing approach with a sensitive method for “signal generation” and specific method for “signal validation.” Whether serially combining epidemiological designs improves both sensitivity and specificity is unknown.MethodsWe assessed the overall performance of serial testing using three administrative claims and one electronic health record database. We compared Type I and II errors before and after empirical calibration for historical comparator, SCCS, and the serial combination of those designs against six vaccine exposure groups with 93 negative control and 279 imputed positive control outcomes.ResultsHistorical comparator mostly had lower Type II error than SCCS. SCCS had lower Type I error than the historical comparator. Before empirical calibration, serial combination increased specificity and decreased sensitivity. Type II errors mostly exceeded 50%. After empirical calibration, Type I errors returned to nominal; sensitivity was lowest when the methods were combined.ConclusionWe recommend against the serial approach in vaccine safety surveillance. While serial combination produced fewer false positive signals compared to the most specific method, it generated more false negative signals compared to the most sensitive method. Using the noisy historical comparator in front of SCCS deteriorated overall performance in evaluating safety signals.Key MessagesUsing the serial approach in vaccine safety surveillance did not improve overall performance: specificity increased but sensitivity decreased.Without empirical calibration, Type II errors exceeded 50%; after empirical calibration, Type I error rates returned to nominal with negligible change to Type II error rates.While prior research has suggested high sensitivity of the historical comparator method in distinguishing true safety signals, there were cases when self-controlled case series was more sensitive.Vaccine safety surveillance is becoming increasingly important, so monitoring systems should closely consider the utility and sequence of epidemiological designs.

Objectives We propose a one-shot, privacy-preserving distributed algorithm to perform logistic regression (ODAL) across multiple clinical sites. Materials and Methods ODAL effectively utilizes the information from the local site (where the patient-level data are accessible) and incorporates the first-order (ODAL1) and second-order (ODAL2) gradients of the likelihood function from other sites to construct an estimator without requiring iterative communication across sites or transferring patient-level data. We evaluated ODAL via extensive simulation studies and an application to a dataset from the University of Pennsylvania Health System. The estimation accuracy was evaluated by comparing it with the estimator based on the combined individual participant data or pooled data (ie, gold standard). Results Our simulation studies revealed that the relative estimation bias of ODAL1 compared with the pooled estimates was Conclusions This study demonstrates that ODAL is privacy-preserving and communication-efficient with small bias and high statistical efficiency.

Background: Observational health data has the potential to be a rich resource to inform clinical practice and regulatory decision making. However, the lack of standard data quality processes makes it difficult to know if these data are research ready. The EHDEN COVID-19 Rapid Col-laboration Call presented the opportunity to assess how the newly developed open-source tool Data Quality Dashboard (DQD) informs the quality of data in a federated network. Methods: 15 Data Partners (DPs) from 10 different countries worked with the EHDEN taskforce to map their data to the OMOP CDM. Throughout the process at least two DQD results were collected and compared for each DP. Results: All DPs showed an improvement in their data quality between the first and last run of the DQD. The DQD excelled at helping DPs identify and fix conformance is-sues but showed less of an impact on completeness and plausibility checks. Conclusions: This is the first study to apply the DQD on multiple, disparate databases across a network. While study-specific checks should still be run, we recommend that all data holders converting their data to the OMOP CDM use the DQD as it ensures conformance to the model specifications and that a database meets a baseline level of completeness and plausibility for use in research.

Studies of the effects of medical interventions increasingly take place in distributed research settings using data from multiple clinical data sources including electronic health records and administrative claims. In such settings, privacy concerns typically prohibit sharing of individual patient data, and instead, cross-network analyses can only utilize summary statistics from the individual databases such as hazard ratios and standard errors. In the specific but very common context of the Cox proportional hazards model, we show that combining such per site summary statistics into a single network-wide estimate using standard meta-analysis methods leads to substantial bias when outcome counts are small. This bias derives primarily from the normal approximations of the per site likelihood that the methods utilized. Here we propose and evaluate methods that eschew normal approximations in favor of three more flexible approximations: a skew-normal, a one-dimensional grid, and a custom parametric function that mimics the behavior of the Cox likelihood function. In extensive simulation studies, we demonstrate how these approximations impact bias in the context of both fixed-effects and (Bayesian) random-effects models. We then apply these approaches to three real-world studies of the comparative safety of antidepressants, each using data from four observational health care databases.

Author(s): Khalid, Sara; Yang, Cynthia; Blacketer, Clair; Duarte-Salles, Talita; Fernandez-Bertolin, Sergio; Kim, Chungsoo; Park, Rae Woong; Park, Jimyung; Schuemie, Martijn J; Sena, Anthony G; Suchard, Marc A; You, Seng Chan; Rijnbeek, Peter R; Reps, Jenna M | Abstract: Background and objectiveAs a response to the ongoing COVID-19 pandemic, several prediction models in the existing literature were rapidly developed, with the aim of providing evidence-based guidance. However, none of these COVID-19 prediction models have been found to be reliable. Models are commonly assessed to have a risk of bias, often due to insufficient reporting, use of non-representative data, and lack of large-scale external validation. In this paper, we present the Observational Health Data Sciences and Informatics (OHDSI) analytics pipeline for patient-level prediction modeling as a standardized approach for rapid yet reliable development and validation of prediction models. We demonstrate how our analytics pipeline and open-source software tools can be used to answer important prediction questions while limiting potential causes of bias (e.g., by validating phenotypes, specifying the target population, performing large-scale external validation, and publicly providing all analytical source code).MethodsWe show step-by-step how to implement the analytics pipeline for the question: 'In patients hospitalized with COVID-19, what is the risk of death 0 to 30 days after hospitalization?'. We develop models using six different machine learning methods in a USA claims database containing over 20,000 COVID-19 hospitalizations and externally validate the models using data containing over 45,000 COVID-19 hospitalizations from South Korea, Spain, and the USA.ResultsOur open-source software tools enabled us to efficiently go end-to-end from problem design to reliable Model Development and evaluation. When predicting death in patients hospitalized with COVID-19, AdaBoost, random forest, gradient boosting machine, and decision tree yielded similar or lower internal and external validation discrimination performance compared to L1-regularized logistic regression, whereas the MLP neural network consistently resulted in lower discrimination. L1-regularized logistic regression models were well calibrated.ConclusionOur results show that following the OHDSI analytics pipeline for patient-level prediction modelling can enable the rapid development towards reliable prediction models. The OHDSI software tools and pipeline are open source and available to researchers from all around the world.

IntroductionObservational data enables large-scale vaccine safety surveillance but requires careful evaluation of potential sources of bias. One potential source of bias is an index date selection procedure for the unvaccinated cohort or unvaccinated comparison time. Here, we evaluate different index date selection procedures for two vaccines: COVID-19 and influenza.MethodsFor each vaccine, we extracted patient baseline characteristics on the index date and up to 450 days prior and then compared them to the characteristics of the unvaccinated patients indexed on an arbitrary date or indexed on a date of a visit. Additionally, we compared vaccinated patients indexed on the date of vaccination and the same patients indexed on a prior date or visit.ResultsCOVID-19 vaccination and influenza vaccination differ drastically from each other in terms of populations vaccinated and their status on the day of vaccination. When compared to indexing on a visit in unvaccinated population, influenza vaccination had markedly higher covariate proportions and COVID-19 vaccination had lower proportions of most covariates on the index date. In contrast, COVID-19 vaccination had similar covariate proportions when compared to an arbitrary date. These effects attenuated but were still present with a longer lookback period. The effect of day 0 was present even when patients served as their own controls.ConclusionPatient baseline characteristics are sensitive to the choice of the index date. In vaccine safety studies, unexposed index event should represent vaccination settings. Study designs previously used to assess influenza vaccination must be reassessed for COVID-19 to account for a potentially healthier population and lack of medical activity on the day of vaccination.

Background Observational data enables large-scale vaccine safety surveillance but requires careful evaluation of the potential sources of bias. One potential source of bias is the index date selection procedure for the unvaccinated cohort or unvaccinated comparison time (“anchoring”). Objective Here, we evaluated the different index date selection procedures for 2 vaccinations: COVID-19 and influenza. Methods For each vaccine, we extracted patient baseline characteristics on the index date and up to 450 days prior and then compared them to the characteristics of the unvaccinated patients indexed on (1) an arbitrary date or (2) a date of a visit. Additionally, we compared vaccinated patients indexed on the date of vaccination and the same patients indexed on a prior date or visit. Results COVID-19 vaccination and influenza vaccination differ drastically from each other in terms of the populations vaccinated and their status on the day of vaccination. When compared to indexing on a visit in the unvaccinated population, influenza vaccination had markedly higher covariate proportions, and COVID-19 vaccination had lower proportions of most covariates on the index date. In contrast, COVID-19 vaccination had similar covariate proportions when compared to an arbitrary date. These effects attenuated, but were still present, with a longer lookback period. The effect of day 0 was present even when the patients served as their own controls. Conclusions Patient baseline characteristics are sensitive to the choice of the index date. In vaccine safety studies, unexposed index event should represent vaccination settings. Study designs previously used to assess influenza vaccination must be reassessed for COVID-19 to account for a potentially healthier population and lack of medical activity on the day of vaccination.

BACKGROUND Observational data enables large-scale vaccine safety surveillance but requires careful evaluation of the potential sources of bias. One potential source of bias is the index date selection procedure for the unvaccinated cohort or unvaccinated comparison time (“anchoring”). OBJECTIVE Here, we evaluated the different index date selection procedures for 2 vaccinations: COVID-19 and influenza. METHODS For each vaccine, we extracted patient baseline characteristics on the index date and up to 450 days prior and then compared them to the characteristics of the unvaccinated patients indexed on (1) an arbitrary date or (2) a date of a visit. Additionally, we compared vaccinated patients indexed on the date of vaccination and the same patients indexed on a prior date or visit. RESULTS COVID-19 vaccination and influenza vaccination differ drastically from each other in terms of the populations vaccinated and their status on the day of vaccination. When compared to indexing on a visit in the unvaccinated population, influenza vaccination had markedly higher covariate proportions, and COVID-19 vaccination had lower proportions of most covariates on the index date. In contrast, COVID-19 vaccination had similar covariate proportions when compared to an arbitrary date. These effects attenuated, but were still present, with a longer lookback period. The effect of day 0 was present even when the patients served as their own controls. CONCLUSIONS Patient baseline characteristics are sensitive to the choice of the index date. In vaccine safety studies, unexposed index event should represent vaccination settings. Study designs previously used to assess influenza vaccination must be reassessed for COVID-19 to account for a potentially healthier population and lack of medical activity on the day of vaccination.

Objective: The current observational research literature shows extensive publication bias and contradiction. The Observational Health Data Sciences and Informatics (OHDSI) initiative seeks to improve research reproducibility through open science. Methods: OHDSI has created an international federated data source of electronic health records and administrative claims that covers nearly 10% of the world’s population. Using a common data model with a practical schema and extensive vocabulary mappings, data from around the world follow the identical format. OHDSI’s research methods emphasize reproducibility, with a large-scale approach to addressing confounding using propensity score adjustment with extensive diagnostics; negative and positive control hypotheses to test for residual systematic error; a variety of data sources to assess consistency and generalizability; a completely open approach including protocol, software, models, parameters, and raw results so that studies can be externally verified; and the study of many hypotheses in parallel so that the operating characteristics of the methods can be assessed. Results: OHDSI has already produced findings in areas like hypertension treatment that are being incorporated into practice, and it has produced rigorous studies of COVID-19 that have aided government agencies in their treatment decisions, that have characterized the disease extensively, that have estimated the comparative effects of treatments, and that the predict likelihood of advancing to serious complications. Conclusions: OHDSI practices open science and incorporates a series of methods to address reproducibility. It has produced important results in several areas, including hypertension therapy and COVID-19 research.

Purpose To compare the incidence of diabetic ketoacidosis (DKA) among patients with type 2 diabetes mellitus (T2DM) who were new users of sodium glucose co‐transporter 2 inhibitors (SGLT2i) versus other classes of antihyperglycemic agents (AHAs). Methods Patients were identified from four large US claims databases using broad (all T2DM patients) and narrow (intended to exclude patients with type 1 diabetes or secondary diabetes misclassified as T2DM) definitions of T2DM. New users of SGLT2i and seven groups of comparator AHAs were matched (1:1) on exposure propensity scores to adjust for imbalances in baseline covariates. Cox proportional hazards regression models, conditioned on propensity score‐matched pairs, were used to estimate hazard ratios (HRs) of DKA for new users of SGLT2i versus other AHAs. When I2

IntroductionTherapeutic options for type 2 diabetes mellitus (T2DM) have expanded over the last decade with the emergence of cardioprotective novel agents, but without such data for older drugs, leaving a critical gap in our understanding of the relative effects of T2DM agents on cardiovascular risk.Methods and analysisThe large-scale evidence generations across a network of databases for T2DM (LEGEND-T2DM) initiative is a series of systematic, large-scale, multinational, real-world comparative cardiovascular effectiveness and safety studies of all four major second-line anti-hyperglycaemic agents, including sodium–glucose co-transporter-2 inhibitor, glucagon-like peptide-1 receptor agonist, dipeptidyl peptidase-4 inhibitor and sulfonylureas. LEGEND-T2DM will leverage the Observational Health Data Sciences and Informatics (OHDSI) community that provides access to a global network of administrative claims and electronic health record data sources, representing 190 million patients in the USA and about 50 million internationally. LEGEND-T2DM will identify all adult, patients with T2DM who newly initiate a traditionally second-line T2DM agent. Using an active comparator, new-user cohort design, LEGEND-T2DM will execute all pairwise class-versus-class and drug-versus-drug comparisons in each data source, producing extensive study diagnostics that assess reliability and generalisability through cohort balance and equipoise to examine the relative risk of cardiovascular and safety outcomes. The primary cardiovascular outcomes include a composite of major adverse cardiovascular events and a series of safety outcomes. The study will pursue data-driven, large-scale propensity adjustment for measured confounding, a large set of negative control outcome experiments to address unmeasured and systematic bias.Ethics and disseminationThe study ensures data safety through a federated analytic approach and follows research best practices, including prespecification and full disclosure of results. LEGEND-T2DM is dedicated to open science and transparency and will publicly share all analytic code from reproducible cohort definitions through turn-key software, enabling other research groups to leverage our methods, data and results to verify and extend our findings.