The Healthcare Integrated Research Database (HIRD) is a real-world data source for health-related research. Data elements of the HIRD, including sourcing, timeliness, and validity, demographic and healthcare-related characteristics of individuals in the HIRD, and aspects of the HIRD relevant to real-world evidence generation, are described. The HIRD includes health insurance claims and other health-related information for individuals enrolled in health insurance plans offered or managed by Elevance Health and has been utilized for research for almost two decades. Individuals in the HIRD reside throughout the United States. Data in the HIRD are available since January 1, 2006, and are updated monthly. As of July 31, 2024, the researchable population of the HIRD included over 91 million individuals with medical benefits, and over 24 million individuals were actively enrolled. The median age of individuals in the HIRD is 36 years (interquartile range, IQR: 22, 54) and 50% of individuals in the HIRD are female. The median duration of continuous enrollment in the HIRD is 2.0 years (IQR: 0.8, 4.7). For those actively enrolled, the median duration of continuous enrollment is 3.8 years (IQR: 1.7, 8.3). Other important characteristics of the HIRD include the ability to trace data back to their source, to support both deterministic and probabilistic linkage with external data sources, and to link family members within health plans. The HIRD has been a trusted resource to generate real-world evidence for a variety of health-related research, including regulatory required safety studies, comparative effectiveness studies, and health outcomes and economics research.

Background Long-acting Gonadotropin-releasing hormone analogs(GnRHa), including leuprolide, goserelin, histrelin, buserelin, triptorelin, have been widely used for a variety of diseases including prostate cancer, breast cancer, endometriosis, uterine leiomyomas, and central precocious puberty. However, their real-world safety profile differences have not been adequately compared. Objective: We aimed to investigate the adverse event (AE) profile differences of long-acting GnRHa reported by the US Food and Drug Administration Adverse Event Reporting System (FAERS). Methods: All indications were searched long-acting GnRHa, as primary suspect drugs, from FAERS data (January 2004 to September 2023). We performed disproportionality analyses by reporting odds ratios (ROR) and conducted univariate and multivariate logistical regression analyses to determine the odds ratio (OR) of serious AEs associated with long-acting GnRHa under different exposure factors. Results: Reproductive system and breast disorders possessed the greatest proportion in AEs of the five long-acting GnRHa. Buserelin and histrelin demonstrated a higher percentage of AEs in the gastrointestinal and psychiatric disorders separately, which also exhibited a higher risk of important medical events in logistic regression analysis and the risk was lower in patients under 18 years old than those over 18 . Conclusions：Significant disparities exist between the adverse event profiles of long-acting GnRHa.The identification of high-risk factors and the enhancement of AEs monitoring are crucial during clinical application.

Masking is a reporting bias where drug safety signals are muffled by elevated reporting of other medications in spontaneous reporting databases. While impacts of masking are often limited, its effect on restricted designs, such as active comparators, can be consequential. We used data from the United States Food and Drugs Administration Adverse Event Reporting System (1999Q3-2013Q3) to study masking in a real-world example. Rosiglitazone, a thiazolidinedione with elevated reporting after safety concerns over cardiovascular risks, was the masking candidate. We hypothesized stimulated reporting masked signals for another thiazolidinedione, pioglitazone. We computed estimates of proportional reporting ratios and information components, using the Bayesian confidence propagation neural network, for pioglitazone-myocardial infarction and pioglitazone-cardiac failure under unrestricted and active comparator designs, both with and without the mask, and before (1999Q3-2007Q1) and after (2007Q1-2013Q3) safety concerns. Relative change-in-estimates were computed to compare results with and without rosiglitazone. From 1999Q3-2007Q1, relative change-in-estimates of proportional reporting ratio for pioglitazone-myocardial infarction was 0.00 in unrestricted design and 0.10 in active comparator; For pioglitazone-cardiac failure, the change was 0.01 and 0.62, respectively. From 2007Q2-2013Q3, relative change in estimate for pioglitazone-myocardial infarction was 0.41 in unrestricted design and 18.00 in active comparator; the change for pioglitazone-cardiac failure was 0.04 and 1.03, respectively. Relative changes in estimates of information component mirrored these trends. In conclusion, masking can influence signal detection in active comparator designs where external events impact reporting rates in reference sets. Evaluating masking in related contexts is essential for drug safety monitoring and resource allocation for follow-up studies.

Purpose We describe a novel model for collaboration between academia and the pharmaceutical industry, focused on post-marketing pharmacovigilance. Methods Otsuka Pharmaceutical, a global Japan-based pharmaceutical company, and Columbia University, a major university, have established a collaboration dedicated to working together in pharmacoepidemiology and post-marketing pharmacovigilance. An oversight committee, made up of individuals from each institution, meets on a regular basis to set policy and provide oversight. Results The primary aim of this novel academia-industry collaboration is to provide expert research guidance for the industry pharmacovigilance group on questions involving pharmacoepidemiology. University epidemiologists may also be consulted by other divisions of industry, such as the clinical trials group. The first aim of the collaboration is to provide epidemiologic input to industry by determining the incidence, prevalence, and outcomes of diseases; drafting the epidemiology components of risk management plans for drugs; and planning retrospective database analyses. A second major aim is to provide educational services to industry by conducting workshops on basic epidemiology and biostatistics; leading a monthly lecture/journal club series; hosting seminars on medical topics; and providing a writing workshop to assist in preparing abstracts and papers for presentation and publication. University epidemiologists provide oversight/evaluation through quarterly presentations and updates to the industry partner’s external advisory committee as well as to University leadership. Conclusions This unique academia-pharmaceutical industry collaboration enhances understanding of the medical and epidemiologic challenges faced within a pharmacovigilance department of a global pharmaceutical company. We offer it as a model for others performing mandatory regulatory post-marketing pharmacovigilance activities.

Valproate-containing medicines (VPA) are first-line treatment for epilepsy; however, they pose teratogenic risks, restricting their use in women of childbearing age. We aimed to estimate the secular trends in use of VPA and alternative treatments in young women, and to characterise dose/strength, treatment duration and indication in new VPA users We conducted a population-based cohort study using primary care records from the Netherlands, Spain and the UK, primary and outpatient specialist care records from Germany and Belgium (and hospital records from Finland, all mapped to the OMOP Common data model. All women present in the databases aged ≥12 years and ≤55 years on 1st of January of each year in the period 2010-2022 (or latest available), with at least 365 days of prior observation were included. Incidence and prevalence of VPA use in young women decreased between 2010 and 2021, while prevalence of the alternative treatments pregabalin and gabapentin increased, especially in CPRD (it rises from 0.5% to 1.5%). Median age of new VPA users ranged between 40-43 years. Anxiety and depressive disorder were frequent comorbidities, and use of hormonal contraceptives low. Average treatment duration varied substantially across databases . Incidence and prevalence of use of VPA among young women declined since 2015. Conversely, alternative antiepileptics have increased in uptake, particularly gabapentinoids. The use of standardized federated analytics allowed for a rapid assessment of VPA utilisation, supporting the regulatory agencies in their decision making and improving patient safety across Europe.

Background. Drug shortages are an increasing and worldwide problem. Oral antibiotics are one of the most used medicines worldwide and have recently been affected by drug shortages. Despite this, little is known about the impact of antibiotic shortages on clinical and prescribing practices. Aim. To explore the impact of oral antibiotic shortages on national antibiotic utilisation. Methods. A longitudinal study of oral antibiotic shortages and antibiotic utilisation was conducted using Australian reimbursement and regulatory data from January 2022 to December 2023. All nationally reimbursed oral antibiotics were included in the study. The number and duration of reported antibiotic shortages per product were determined for each active ingredient. The clinical impact was assessed using national utilisation in Defined Daily Doses per 100,000 inhabitants. Changes in trends were analysed using Joinpoint regression. Results. Shortages were reported for eighteen of the twenty-one (86%) oral antibiotics reimbursed in Australia. No clear relation between the number and duration of shortages was observed for most antibiotics. Changes in utilisation coinciding with shortages were observed for eight active ingredients. For cefaclor (-20% decrease in utilisation) and roxithromycin (-26% decrease), the impact of shortages is most clearly reflected by decreases in utilisation. For the other six, either minor or mixed changes in utilisation were observed. Conclusions. Antibiotic shortages were common in Australia during 2022 and 2023. The impact of shortages appears complex with various factors most likely influencing impact. Australia’s healthcare professionals seem successful in mitigating the effects of shortages on drug utilisation.

Purpose: Record linkage based on quasi-identifiers remains an important approach as not every data source provides a comprehensive unique identifier. In this study, reasons for the failure of a linkage based on quasi-identifiers were examined. Furthermore, informed algorithms using information on gold-standard links were developed to investigate the potentially achievable linkage quality based on quasi-identifiers. Methods: Linkage algorithms were applied on German claims and cancer registry data using information on gold-standard links. Informed linkage algorithms based on deterministic linkage, logistic regression, random forests, gradient boosting and neural networks were derived and compared. Descriptive analyses were performed to identify reasons for failure of linkage such as discrepancies between data sources. Results: A linkage approach based on gradient boosting performed best and reached a precision of 77%, a recall of 81% and an F*-measure of 64%. Of 641 patients in GePaRD, 8% were not uniquely identifiable using birth year, sex, area of residence, year and quarter of diagnosis, whereas 33% of 42,817 cancer registries patients of Bremen and Lower Saxony were not uniquely identifiable with these quasi-identifiers. Conclusions: Linkage of German claims and cancer registry data based on quasi-identifiers does result in insufficient linkage quality since subjects cannot be uniquely identified. It is advisable to use unique identifiers from a subsample, if available, to derive informed linkage algorithms for the entire sample. In this case the machine learning technique gradient boosting has been found to outperform other methods.

Purpose: This observational study aimed to assess the comparative effectiveness of immunotherapy (IO) in the second-line setting of advanced Non-Small Cell Lung Cancer (aNSCLC). Methods: Population-based data was obtained from the Cancer Registry of Norway (CRN) and linked with the hospital data from the Norwegian Patient Registry (NPR) and treatment data from the Norwegian Prescribed Drug Registry (NorPD). Adult patients with aNSCLC diagnosis in Norway in 2015-2021 were included. The final cohort included patients using IO in the second line setting after first-line treatment with platinum-containing doublet chemotherapy. We performed target trial emulating key eligibility criteria commonly used in aNSCLC randomized clinical trials (RCTs) and constructed three pair-wise 1:1 Propensity score (PS) matched comparison cohorts: atezolizumab versus pembrolizumab, atezolizumab versus nivolumab and nivolumab versus pembrolizumab. Age, sex, morphology, ECOG performance status, PDL1 expression level, peripheral vascular disease, chronic lung disease, diabetes without complications, heart failure, moderate to severe kidney disease, cerebrovascular disease, number of comorbidities per patients and etoposide use were included in the PS. The primary outcome was overall survival (OS). Patients were followed until experiencing death, emigration or end of study. A Cox proportional hazards regression model was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Results: Analyses included a total of 463 patients, 215 users of atezolizumab, 72 nivolumab and 176 pembrolizumab. The median follow-up time was 8.4 (IQR 3.4—20.7) months. We found no significant difference in OS in any of the three pair-wise comparisons; atezolizumab vs. pembrolizumab (HR 0.92; 95% CI: 0.68—1.23), atezolizumab vs. nivolumab (HR 1.14; 95% CI: 0.78—1.67) and nivolumab vs. pembrolizumab (HR 1.11; 95% CI: 0.73—1.71). Conclusion: The study indicates that the three IO drugs atezolizumab, nivolumab and pembrolizumab, when used in second line aNSCLC, demonstrate a similar level of effectiveness. To substantiate these results, further investigation with larger-scale studies is required.

Purpose: This scoping review contributes to understanding the utilization of end-of-life anticancer drugs in gastrointestinal cancer patients, a topic posing significant public health challenges due to its prevalence, impact on morbidity and mortality, and associated healthcare costs. Insight into unique treatment patterns at the end of life is vital for enhancing the appropriateness of cancer care for these patients. Methods: Extensive searches were conducted in Medline and Embase to locate articles on the use of anticancer drugs in the end-of-life phase of gastrointestinal cancer patients. Results: We identified 30 publications and described treatment frequencies alongside methodological characteristics. On average, 7.0% and 13.2% of patients received anticancer drugs in the last two and four weeks of life, respectively, with 3.3% commencing new anticancer regimens in the final four weeks. Conclusion: This review offers a comprehensive overview of primary studies reporting end-of-life treatments in gastrointestinal cancer patients. However, methodological inconsistencies present significant challenges, highlighting a notable proportion of patients potentially experiencing overtreatment. Therefore, there is a pressing need for more standardized research methods to ensure robust evaluations and enhance the quality of care for these patients.

Background: Pregnant individuals are at high risk for developing severe illness related to COVID-19. We adapted the “COVID-19 infectiOn aNd medicineS In pregnancy” (CONSIGN) CONSIGN study protocol as part of an international collaboration to understand the natural history of COVID-19 disease among pregnant individuals in the U.S. Methods: We identified individuals aged 12 to 55 years with documented live-birth deliveries in the Sentinel Distributed Database who had at least one qualifying diagnosis for COVID-19 or positive-confirmed nucleic acid amplification test for SARS-CoV-2. We conducted separate 1:1 propensity score matched analyses comparing pregnant individuals with COVID-19 to: 1) pregnant individuals without COVID-19 during six months prior to or during pregnancy; and 2) non-pregnant individuals with COVID-19. Results: From January 2020 to December 2022 , we identified 52,355 pregnant individuals with COVID-19 matched to 52,355 pregnant individuals without COVID-19. Outpatient medication use in the 30 days surrounding COVID-19 date (or matched date) was generally low but similar between pregnant individuals with and without COVID-19. We identified 40,518 pregnant individuals with COVID-19 matched to 40,518 non-pregnant individuals with COVID-19. Medication use in 30 days prior to COVID-19 was higher among non-pregnant than pregnant individuals with COVID-19. More pregnant individuals than matched non-pregnant individuals were classified as non-severe (87.2% vs. 79.9%); however, more non-pregnant individuals could not have their COVID-19 severity determined (19.0% vs. 10.0%). Conclusions: Medication use is generally low in pregnancies with COVID-19, compared to pregnancies without COVID-19 and non-pregnant individuals with COVID-19. Interpretation of medication utilization patterns is challenging due to changing treatment and prevention recommendations during the pandemic.

A document by Carla Rodriguez-Watson. Click on the document to view its contents.

Introduction: Use observational methods to evaluate reliability of evidence generated by a study of the effect of glucagon-like peptide 1 receptor agonists (GLP-1RA) on chronic lower respiratory disease (CLRD) outcomes among type-2 diabetes mellites (T2DM) patients. Research Design and Methods: We independently reproduced a study comparing effects of GLP-1RA versus dipeptidyl peptidase-4 inhibitors (DPP4-i) on CLRD outcome among patients with T2DM and prior CLRD. We reproduced inputs and outputs using the original study data (national administrative claims) and evaluated robustness to alternate design/analysis decisions. To evaluate generalizability, we applied the protocol and meta-analyzed across a research network including diverse array of populations and data sources. We also produced additional analyses evaluating individual drugs within the GLP-1RA class. Results: We confirmed alignment of study inputs and outputs and closely reproduced effect estimates and sensitivity analyses. Adjusted effect estimates were robust to empirical calibration. Network meta-analysis confirmed original findings, but indicated weaker effects than originally published. Meta-analyzing drugs within the GLP-1RA class against DPP4-I provided some evidence that effects vary within the GLP-1RA class, indicating stronger effects for exenatide and weaker effects of dulaglutide. Conclusions: This study supports the reliability of the original study by 1) confirming the findings in a range of alternate databases and populations 2) demonstrating effects for multiple drugs within the GLP-1RA class, and 3) independently confirming the reproducibility original study and its findings. We propose that clinicians treating patients with T2DM and a history of CLRD consider GLP-1RA in absence of strong motivating reasons to select another therapy.

Background: Despite cardiovascular benefits of sodium-glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP1RA) in patients with type 2 diabetes mellitus (T2DM) and atherosclerotic cardiovascular disease (ASCVD), their utilization remains low globally. This study aimed to evaluate the utilization of SGLT2i and GLP1RA in patients with T2DM and ASCVD, as well as the factors associated with medications in South Korea. Methods: This retrospective study was conducted from 2015 to 2020, using National Patient Sample claims data. The study population included adult patients with confirmed T2DM and ASCVD diagnosed between March 1 and October 31 each year. Demographic and clinical characteristics, and influencing factors were investigated. Results: Among 57,576 study participants, SGLT2i use increased from 1.2 % to 10.51 % during the study period, whereas GLP1RA use increased slightly from 0 % to 1.17 %. Older age, comorbid chronic kidney disease, concurrent use of dipeptidyl peptidase 4 inhibitors (DPP4i), and prescriptions from specific physician specialties negatively influenced SGLT2i use. Conversely, comorbid dyslipidemia, heart failure, concurrent use of sulfonylurea (SU), and prescriptions from cardiologists positively influenced SGLT2i use. For GLP1RA, older age, concurrent DPP4i use, and specific physician specialty were negative factors, whereas female sex, dyslipidemia, insulin, and SU use were positive factors. Conclusions: Despite increasing utilization, 88.35 % of eligible patients remained untreated with SGLT2i and GLP1RA as of 2020. This study highlights the disparities in utilization based on patient characteristics and physician specialties, emphasizing the need to remove barriers and enhance clinical benefits for high-risk patients.

Purpose: Hypertension (HT), dyslipidemia (DL), and diabetes mellitus (DM) are major risk factors for cardiovascular diseases. Despite the wide availability of medications to reduce this risk, poor adherence to medications remains an issue. The aim of this study is to evaluate medication adherence of prevalent users in these disease medications (HT, DL, DM) using claims data. Factors associated with non-adherence were also examined. Methods: Of 7,538 participants of the Tsuruoka Metabolomics Cohort Study, 3,737（HT: 2,724, DL: 2,137, DM: 665) were identified as prevalent users of these disease medications. Information on lifestyle was collected through a questionnaire. Adherence was assessed by a proportion of days covered (PDC) and participants with PDC ≥ 0.8 were defined as adherent. Predictors of non-adherence were determined by performing multivariable logistic regression. Results: Medication adherence differed by medication groups. Those with DL-only showed the lowest adherence (79.6%), followed by those with DM-only (81.7%) and those with HT-only (89.8%). The associations were observed especially among those with no comorbidities. Determinants associated with non-adherence in each medication group were skipping breakfast and poor understanding of medications among those with HT, females, having comorbidities, having the medical history of stroke, and drinking habit among those with DL, and skipping breakfast among those with DM. Conclusion: While participants showed high medication adherence, differences were observed across medication groups. The identified predictors of non-adherence could help target those in need of adherence support.

Objective To identify primary care recorded factors associated with the short-term human papillomavirus (HPV) vaccination among girls and female adolescents. Methods We used the Spanish Pharmacoepidemiological Research Database for Public Health System to identify girls aged 9-18 years in 2007-2016 with ≥1 year of primary care data (start date) and collect baseline characteristics and consultations. The link between those factors and the 1-year HPV vaccination was assessed through multivariate adjusted Odds Ratios (OR; 95% CI). Results Among 388,669 girls identified, 14,999 were vaccinated against HPV during the first year. The vaccination decreased among girls consulting for social (0.62; 0.55-0.70) or adaptation (0.60; 0.38-0.93) problems, prescriptions of benzodiazepines (0.58; 0.35-0.94) or drugs for gastrointestinal symptoms (0.58; 0.38-0.89), with recorded tuberculosis test (0.68; 0.52-0.87), B and C meningitis (0.46; 0.43-0.50) vaccination, consulting the doctor in the last 3 months (0.92; 0.87-0.98) and with severe thinness (0.68; 0.59-0.78) or obesity (0.88; 0.78-0.99). The odds increased among girls with abdominal pain (1.10; 1.04-1.16), prescriptions in the last 3 months (1.17; 1.10-1.25), and vaccinations against chickenpox (1.16; 1.07-1.27), influenza (1.23; 1.08-1.40), hepatitis (1.46; 1.35-1.58) and diphtheria-tetanus (1.91; 1.82-2.01). Conclusions The correlation between the HPV and other vaccinations suggested medical (contra)indications or willingness and knowledge/beliefs that might affect different vaccinations similarly. The higher frequency of the HPV vaccination among females requiring specific treatments or with abdominal pain suggested vulnerability to HPV complications. While a decreased HPV vaccination linked to certain social and personal situations (and benzodiazepines treatment) requires further research.

Letter to the Editor Pharmacoepidemiology and Drug SafetyAuthors: Christopher Lee1, Silvia Rizzi 1, Ana Luiza Bierrenbach1, Luis Antunes1, Niina-Maria Nissinen1, Olga Rocha1, Jennifer Campbell2, Caroline Minassian3, Susan Hodgson2, Anne Broe1

Background: Valsartan is commonly used for cardiac conditions. In 2018, the Food and Drug Administration recalled generic valsartan due to the detection of impurities. Our objective was to determine if heart failure patients receiving valsartan at the recall date had a greater likelihood of unfavorable outcomes than patients using comparable antihypertensives. Methods: We conducted a cohort study of Optum’s de-identified Clinformatics® Datamart (July 2017-January 2019). Heart failure patients with commercial or Medicare Advantage insurance who received valsartan were compared to persons who received non-recalled angiotensin receptor blockers (ARBs) and angiotensin converting enzyme-inhibitors (ACE-Is) for 1 year prior and including the recall date. Outcomes included a composite for all-cause hospitalization, emergency department (ED), and urgent care (UC) use and a measure of cardiac events which included hospitalizations for acute myocardial infarction and hospitalizations/ED/UC visits for stroke/transient ischemic attack, heart failure or hypertension at 6-months post-recall. Cox proportional hazard models with propensity score weighting compared the risk of outcomes between groups. Results: Of the 87,130 patients, 15% were valsartan users and 85% were users of non-recalled ARBs/ACE-Is. Valsartan use was not associated with an increased risk of all-cause hospitalization/ED/UC use six-months post-recall (HR 1.00; 95%CI 0.96–1.03), compared with individuals taking non-recalled ARBs/ACE-Is. Similarly, cardiac events 6-months post-recall did not differ between individuals on valsartan and non-recalled ARBs/ACE-Is (HR 1.04; 95%CI 0.97–1.12). Conclusions: The valsartan recall did not affect short-term outcomes of heart failure patients. However, the recall potentially disrupted the medication regimens of patients, possibly straining the healthcare system.

Transparency and reproducibility are major prerequisites for conducting meaningful real-world evidence (RWE) studies that are fit for decision-making. Many advances have been made in the documentation and reporting of study protocols and results, but the principles for version control and sharing of analytic code in RWE are not yet as established as in other quantitative disciplines like computational biology and health informatics. In this practical tutorial, we aim to give an introduction to distributed version control systems (VCS) tailored towards the FAIR ( Findable, Accessible, Interoperable and Reproducible) implementation of RWE studies. To ease adoption, we provide detailed step-by-step instructions with practical examples on how the Git VCS and R programming language can be implemented into RWE study workflows to facilitate reproducible analyses. We further discuss and showcase how these tools can be used to track changes, collaborate, disseminate and archive RWE studies through dedicated project repositories that maintain a complete audit trail of all relevant study documents.