Explore the vast range of titles available.

ObjectiveTo examine COVID-19 vaccine effectiveness over six 7-day intervals after the first dose and assess underlying bias in observational data.Design and settingRetrospective cohort study using Columbia University Irving Medical Center data linked to state and city immunisation registries.Outcomes and measuresWe used large-scale propensity score matching with up to 54 987 covariates, fitted Cox proportional hazards models and constructed Kaplan-Meier plots for two main outcomes (COVID-19 infection and COVID-19-associated hospitalisation). We conducted manual chart review of cases in week 1 in both groups along with a set of secondary analyses for other index date, outcome and population choices.ResultsThe study included 179 666 patients. We observed increasing effectiveness after the first dose of mRNA vaccines with week 6 effectiveness approximating 84% (95% CI 72% to 91%) for COVID-19 infection and 86% (95% CI 69% to 95%) for COVID-19-associated hospitalisation. When analysing unexpectedly high effectiveness in week 1, chart review revealed that vaccinated patients are less likely to seek care after vaccination and are more likely to be diagnosed with COVID-19 during the encounters for other conditions. Secondary analyses highlighted potential outcome misclassification for International Classification of Diseases, Tenth Revision, Clinical Modification diagnosis, the influence of excluding patients with prior COVID-19 infection and anchoring in the unexposed group. Long-term vaccine effectiveness in fully vaccinated patients matched the results of the randomised trials.ConclusionsFor vaccine effectiveness studies, observational data need to be scrutinised to ensure compared groups exhibit similar health-seeking behaviour and are equally likely to be captured in the data. While we found that studies may be capable of accurately estimating long-term effectiveness despite bias in early weeks, the early week results should be reported in every study so that we may gain a better understanding of the biases. Given the difference in temporal trends of vaccine exposure and patients’ baseline characteristics, indirect comparison of vaccines may produce biased results.

The electrocardiogram (ECG) is the most frequently performed cardiovascular diagnostic test, but it is unclear how much information resting ECGs contain about long term cardiovascular risk. Here we report that a deep convolutional neural network can accurately predict the long-term risk of cardiovascular mortality and disease based on a resting ECG alone. Using a large dataset of resting 12-lead ECGs collected at Stanford University Medical Center, we developed SEER, the Stanford Estimator of Electrocardiogram Risk. SEER predicts 5-year cardiovascular mortality with an area under the receiver operator characteristic curve (AUC) of 0.83 in a held-out test set at Stanford, and with AUCs of 0.78 and 0.83 respectively when independently evaluated at Cedars-Sinai Medical Center and Columbia University Irving Medical Center. SEER predicts 5-year atherosclerotic disease (ASCVD) with an AUC of 0.67, similar to the Pooled Cohort Equations for ASCVD Risk, while being only modestly correlated. When used in conjunction with the Pooled Cohort Equations, SEER accurately reclassified 16% of patients from low to moderate risk, uncovering a group with an actual average 9.9% 10-year ASCVD risk who would not have otherwise been indicated for statin therapy. SEER can also predict several other cardiovascular conditions such as heart failure and atrial fibrillation. Using only lead I of the ECG it predicts 5-year cardiovascular mortality with an AUC of 0.80. SEER, used alongside the Pooled Cohort Equations and other risk tools, can substantially improve cardiovascular risk stratification and aid in medical decision making.

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.

Objective: Background incidence rates are routinely used in safety studies to evaluate an association of an exposure and outcome. Systematic research on sensitivity of rates to the choice of the study parameters is lacking. Materials and Methods: We used 12 data sources to systematically examine the influence of age, race, sex, database, time-at-risk, season and year, prior observation and clean window on incidence rates using 15 adverse events of special interest for COVID-19 vaccines as an example. For binary comparisons we calculated incidence rate ratios and performed random-effect meta-analysis. Results: We observed a wide variation of background rates that goes well beyond age and database effects previously observed. While rates vary up to a factor of 1,000 across age groups, even after adjusting for age and sex, the study showed residual bias due to the other parameters. Rates were highly influenced by the choice of anchoring (e.g., health visit, vaccination, or arbitrary date) for the time-at-risk start . Anchoring on a healthcare encounter yielded higher incidence comparing to a random date, especially for short time-at-risk. Incidence rates were highly influenced by the choice of the database (varying by up to a factor of 100), clean window choice and time-at-risk duration, and less so by secular or seasonal trends. Conclusion: Comparing background to observed rates requires appropriate adjustment and careful time-at-risk start and duration choice. Results should be interpreted in the context of study parameter choices.

Using administrative claims and electronic health records for observational studies is common but challenging due to data limitations. Researchers rely on phenotype algorithms, requiring labor-intensive chart reviews for validation. This study investigates whether case adjudication using the previously introduced Knowledge-Enhanced Electronic Profile Review (KEEPER) system with large language models (LLMs) is feasible and could serve as a viable alternative to manual chart review. The task involves adjudicating cases identified by a phenotype algorithm, with KEEPER extracting predefined findings such as symptoms, comorbidities, and treatments from structured data. LLMs then evaluate KEEPER outputs to determine whether a patient truly qualifies as a case. We tested four LLMs including GPT-4, hosted locally to ensure privacy. Using zero-shot prompting and iterative prompt optimization, we found LLM performance, across ten diseases, varied by prompt and model, with sensitivities from 78 to 98% and specificities from 48 to 98%, indicating promise for automating phenotype evaluation.

Using real-world data and past vaccination data, we conducted a large-scale experiment to quantify bias, precision and timeliness of different study designs to estimate historical background (expected) compared to post-vaccination (observed) rates of safety events for several vaccines. We used negative (not causally related) and positive control outcomes. The latter were synthetically generated true safety signals with incident rate ratios ranging from 1.5 to 4. Observed vs. expected analysis using within-database historical background rates is a sensitive but unspecific method for the identification of potential vaccine safety signals. Despite good discrimination, most analyses showed a tendency to overestimate risks, with 20%-100% type 1 error, but low (0% to 20%) type 2 error in the large databases included in our study. Efforts to improve the comparability of background and post-vaccine rates, including age-sex adjustment and anchoring background rates around a visit, reduced type 1 error and improved precision but residual systematic error persisted. Additionally, empirical calibration dramatically reduced type 1 to nominal but came at the cost of increasing type 2 error.

Abstract Objective To quantify the background incidence rates of 15 prespecified adverse events of special interest (AESIs) associated with covid-19 vaccines. Design Multinational network cohort study. Setting Electronic health records and health claims data from eight countries: Australia, France, Germany, Japan, the Netherlands, Spain, the United Kingdom, and the United States, mapped to a common data model. Participants 126 661 070 people observed for at least 365 days before 1 January 2017, 2018, or 2019 from 13 databases. Main outcome measures Events of interests were 15 prespecified AESIs (non-haemorrhagic and haemorrhagic stroke, acute myocardial infarction, deep vein thrombosis, pulmonary embolism, anaphylaxis, Bell’s palsy, myocarditis or pericarditis, narcolepsy, appendicitis, immune thrombocytopenia, disseminated intravascular coagulation, encephalomyelitis (including acute disseminated encephalomyelitis), Guillain-Barré syndrome, and transverse myelitis). Incidence rates of AESIs were stratified by age, sex, and database. Rates were pooled across databases using random effects meta-analyses and classified according to the frequency categories of the Council for International Organizations of Medical Sciences. Results Background rates varied greatly between databases. Deep vein thrombosis ranged from 387 (95% confidence interval 370 to 404) per 100 000 person years in UK CPRD GOLD data to 1443 (1416 to 1470) per 100 000 person years in US IBM MarketScan Multi-State Medicaid data among women aged 65 to 74 years. Some AESIs increased with age. For example, myocardial infarction rates in men increased from 28 (27 to 29) per 100 000 person years among those aged 18-34 years to 1400 (1374 to 1427) per 100 000 person years in those older than 85 years in US Optum electronic health record data. Other AESIs were more common in young people. For example, rates of anaphylaxis among boys and men were 78 (75 to 80) per 100 000 person years in those aged 6-17 years and 8 (6 to 10) per 100 000 person years in those older than 85 years in Optum electronic health record data. Meta-analytic estimates of AESI rates were classified according to age and sex. Conclusion This study found large variations in the observed rates of AESIs by age group and sex, showing the need for stratification or standardisation before using background rates for safety surveillance. Considerable population level heterogeneity in AESI rates was found between databases.

The Centers for Disease Control and Prevention’s Advisory Committee on Immunization Practices recommends pneumococcal vaccination for adults with chronic or immunocompromising conditions to prevent pneumococcal disease, yet vaccination rates are low and have limited information on regional variation. This study examines factors associated with pneumococcal vaccination in adults with underlying conditions and describes regional variation in vaccination across the U.S. Using IBM MarketScan Commercial Database and Medicare Supplemental Database, this retrospective cohort study included adults ages 19–64 newly diagnosed with chronic (i.e. diabetes, chronic heart, lung, or liver disease, alcohol or tobacco dependence) or immunocompromising (i.e. cancer, chronic renal disease, organ transplant, HIV/AIDS, and asplenia) conditions in 2013. Adults were followed up until the time of pneumococcal vaccination, death, or December 31, 2019, whichever came first. Cox proportional hazards model was used to examine factors associated with vaccination. Vaccination rate was calculated by metropolitan statistical area (MSA) and visually represented on a U.S. map. 255,330 adults were included. Vaccination rate increased from 6.0% to 21.1% among adults with one year and five years of follow-up, respectively. It took 2.4 years on average for adults to receive vaccination after initial diagnosis. Adults ages 50–64, 35–49 (relative to 19–34) or receiving influenza vaccination were more likely to receive pneumococcal vaccinations. Adults with HIV/AIDS were more likely and those with other conditions were less likely to be vaccinated than those with diabetes. Adults being diagnosed by other providers were less likely to be vaccinated than those diagnosed by primary care providers. Vaccination rate varied largely across MSAs, ranging from 0.0% (Ames, IA; Cheyenne, WY) to 34.0% (Ann Arbor, MI). Pneumococcal vaccination remains low and most adults with underlying conditions are unvaccinated. Insights into factors associated with vaccination, including regional variability, can help to increase pneumococcal vaccination.

A growing body of evidence generated from observational data has demonstrated the potential to influence decision-making and improve patient outcomes. For observational evidence to be actionable, however, it must be generated reliably and in a timely manner. Large distributed observational data networks enable research on diverse patient populations at scale and develop new sound methods to improve reproducibility and robustness of real-world evidence. Nevertheless, the problems of generalizability, portability and scalability persist and compound. As analytical methods only partially address bias, reliable observational research (especially in networks) must address the bias at the design stage (i.e., pre-analysis bias) including the strategies for identifying patients of interest and defining comparators. This thesis synthesizes and enumerates a set of challenges to addressing pre-analysis bias in observational studies and presents mixed-methods approaches and informatics solutions for overcoming a number of those obstacles. We develop frameworks, methods and tools for scalable and reliable phenotyping including data source granularity estimation, comprehensive concept set selection, index date specification, and structured data-based patient review for phenotype evaluation. We cover the research on potential bias in the unexposed comparator definition including systematic background rates estimation and interpretation, and definition and evaluation of the unexposed comparator. We propose that the use of standardized approaches and methods as described in this thesis not only improves reliability but also increases responsiveness of observational evidence. To test this hypothesis, we designed and piloted a Data Consult Service - a service that generates new on-demand evidence at the bedside. We demonstrate that it is feasible to generate reliable evidence to address clinicians’ information needs in a robust and timely fashion and provide our analysis of the current limitations and future steps needed to scale such a service.

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.