Pharmacoepidemiology and Drug Safety

BackgroundIn pharmacoepidemiological studies, days of treatment (DoT) duration associated with individual electronic drug utilization records (DUR) are usually missing. Researcher-defined duration (RDD) calculation approaches, as opposed to data-driven approaches, can be used to estimate DoT based on the specific choices and assumptions made by investigators. These are usually underreported or even undocumented. We aimed to develop a framework for the standardization of terminology, formulas, implementation, and reporting of possible RDD approaches. MethodsA systematic classification of RDD calculation approaches was developed via expert consensus. Universal concepts used to operationalise RDDs were identified and described using standard terminologies. An open-source R function, CreateDoT, was created to implement the formulas universal concepts as input parameter. A step-by-step workflow was developed to facilitate implementation and reporting. ResultsRDD approaches were classified in two main classes: I) daily dose (DD)-based calculation approaches (n=3 formulas), and II) fixed-duration approaches (n=2). Seven universal concepts were identified to describe the five corresponding generalized formulas for DoT calculation. Input parameters of the CreateDoT function can be retrieved from source data through its mapping to universal concepts, or inputted by the investigator based on the chosen calculation approach. The input file structure itself represents a standard reporting template for documenting investigators assumptions and methodological choices adopted for DoT calculation. ConclusionsThe CreateDoT framework can facilitate the documentation and reporting of RDD approaches for DoT calculation, increasing transparency and reproducibility of pharmacoepidemiological studies regardless of the data model used, and facilitates sensitivity analyses to evaluate the impact of alternative assumptions in DoT calculation.

BackgroundLong-acting injectable buprenorphine (LAI-BUP) is safe and effective, however is dramatically underutilized in comparison to oral formulations. Little is known regarding how buprenorphine prescribers view LAI-BUP, and which medication attributes they prioritize when selecting treatment for opioid use disorder (OUD). MethodsA secondary analysis of a national, cross-sectional online survey of U.S. physicians who prescribe buprenorphine for OUD was conducted. Respondents reported OUD caseload, LAI-BUP use, and the importance of medication attributes relevant to treatment selection (e.g., efficacy, safety, ease of administration, ease of prescribing, and administrative requirements). Providers were categorized as no LAI-BUP use or, among LAI-BUP prescribers, Low vs High use based on a median split. Group comparisons used chi-square (or Fishers exact) tests for categorical variables and Jonckheere-Terpstra tests for ordinal responses. ResultsAmong 125 respondents, 39 (31.2%) reported no patients receiving LAI-BUP. The remaining 86 (68.8%) were LAI-BUP prescribers, split evenly into Low and High (ns=43; 34.4%) groups using a median cut of 23.2%. LAI-BUP use did not differ meaningfully by specialty, region, or practice setting. Greater LAI-BUP use was reported by providers with larger OUD panels. Ratings of key medication attributes were uniformly high. ConclusionsLAI-BUP remains underused, with uptake highest among clinicians managing larger OUD caseloads. Measured attitudes toward medication attributes did not explain these differences. Future work should assess clinic workflow, staffing, reimbursement, and REMS burden, testing targeted implementation strategies using mixed-methods trials. Identifying what shifts clinicians from no use to low and high use may guide scalable implementation interventions.

BackgroundBiomedical Large Language Models (LLMs) combined with prompt engineering offer domain-specific reasoning, yet their application to individual-level causality assessment remains unexplored. This study evaluated five combinations of biomedical LLMs, prompting strategies, and causality algorithms by comparing their agreement with two human expert evaluators. Research design and methodsA total of 150 Individual Case Safety Reports (ICSRs) were analyzed: 140 reports from Food and Drug Administration Adverse Event Reporting System (FAERS), and 10 myocarditis/pericarditis ICSRs from Vaccine AERS (VAERS). Assessments were conducted using the Naranjo and WHO-UMC algorithms. Biomedical LLMs tested included TinyLlama 1.1B, Medicine LLaMA-3 8B, and MedLLaMA v20, combined with Chain-of-Thought (CoT) or Decomposition prompting. Agreement was measured using Gwets Agreement Coefficient 1 (AC1) and percentage agreement, alongside performance metrics and qualitative error analysis. ResultsThe Medicine LLaMA-3 8B-Naranjo-CoT combination achieved the highest agreement with human assessors for the final classification of causality (64%). Biomedical LLMs demonstrated low inter-rater agreement on critical items of causality assessment such as identification of listed AE, temporal plausibility, alternative causes, and objective evidence of AEs. Frequent model failures included irrelevant responses. ConclusionsBiomedical LLMs showed improved performance over general purpose models previously tested but remain suboptimal for reliable causality assessment of ICSRs.

BackgroundPost-marketing surveillance is essential for complementing the safety profiles of medicinal products, especially for populations generally excluded from clinical trials such as pregnant individuals. However, the absence of a standardised pregnancy indicator in the electronic transmissions of adverse event reports hampers their correct identification in pharmacovigilance databases and complicates the study of safety concerns related to pregnancy exposures. Three recently developed rule-based algorithms with the common aim to systematically retrieve pregnancy-related reports differ in scope and are tailored to different databases (A. FAERS, B. EudraVigilance, C. VigiBase). AimTo compare the design and outputs of the three pregnancy algorithms. MethodsThis study was a collaboration among the authors of the three pregnancy algorithms. We harmonised their rules, implemented them in an R package to enable execution in both VigiBase and FAERS, and analysed key characteristics of reports flagged by each algorithm. ResultsThe pregnancy algorithms A, B, and C flagged 235653, 279515, and 446957 reports respectively in VigiBase, and 265015, 260734, 350479 in FAERS. Reports exclusively retrieved by each algorithm (994, 3248, and 142324 in VigiBase, and 1528, 1100, and 59643 in FAERS) were mostly explained by Algorithm A having no age restriction, Algorithm B excluding normal pregnancy and ineffective contraception, and Algorithm C excluding paternal exposure. ConclusionsDifferences in flagging were largely related to varying scopes. Understanding commonalities and differences is crucial for empowering professionals working with pregnancy-related pharmacovigilance to select and use the most appropriate algorithm for their specific needs. Key pointsO_LIThree independently developed algorithms were designed to retrieve pregnancy-related adverse event reports and support research into pregnancy-specific safety concerns. C_LIO_LIBy applying these algorithms to VigiBase and FAERS, we highlighted overlaps and differences in the reports they flag, reflecting heterogeneous scope and implementation. C_LIO_LIAwareness of these distinctions is essential to select and apply the most suitable algorithm for their specific needs. C_LI

ImportanceGlucocorticoid steroids are increasingly prescribed during the periconceptual period with the hypothesis that they reduce intrauterine inflammation and improve pregnancy rates. There is no robust evidence to support this practice, and the potential harm has not been well characterised. ObjectiveTo examine the risk of adverse perinatal outcomes associated with non-medically indicated glucocorticoid steroid use during the periconceptual period. DesignPopulation-wide linked retrospective cohort study. SettingVictoria, Australia. ParticipantsAfter excluding women with medical indications for steroid use (autoimmune disease, chronic asthma and previous organ transplant), our total cohort included 805,353 births between 2009 and 2021. ExposurePrescriptions of glucocorticoid steroids dispensed during the periconceptual period (12 weeks prior to conception - end of first trimester). Main Outcome(s) and Measure(s)Four primary outcomes were examined - spontaneous preterm birth before 37 completed weeks gestation, small for gestational age (<10th birthweight centile), major congenital abnormality and perinatal mortality. A doubly robust inverse probability weighted regression adjustment model was used to estimate the association between glucocorticoid steroid exposure and outcomes and presented as adjusted relative risks (aRR) with corresponding 95% confidence intervals (95% CI). ResultsThere were 12,301 (1.5%) pregnancies exposed to glucocorticoid steroids during the periconceptual period and 793,052 unexposed. Among the steroid-exposed cohort, major congenital abnormalities occurred in 4.5% of pregnancies, compared with 3.5% among those unexposed to steroids. This resulted in a 23% increased risk of major congenital abnormality (aRR 1.23, 95%CI 1.13-1.34). There were no significant associations between steroid exposure and spontaneous preterm birth (2.4 vs 2.5%; aRR 1.05, 95%CI 0.93-1.19), small for gestational age neonates (9.4 vs 9.3%; aRR 1.02, 95%CI 0.97-1.07) or perinatal mortality (0.5 vs 0.7%; aRR 1.05, 95%CI 0.87-1.26). Conclusions and RelevanceIn our cohort, periconceptual steroid exposure was associated with an increased risk of major congenital abnormality. In the absence of clear clinical indications, avoiding the prescription of periconceptual steroids is critically important. KEY POINTSO_ST_ABSQuestionC_ST_ABSWhat are the risks of adverse perinatal outcomes among patients prescribed glucocorticoid steroids during the periconceptual period? FindingsPericonceptual steroid use is associated with an increased risk of major congenital abnormality. MeaningIn the absence of an indication, periconceptual steroid use does not have clear evidence of benefit, there is now also evidence of significant harm and this practice should not be recommended.

PurposeMedication use during pregnancy is common, but patterns may differ across occupational subgroups, particularly in healthcare and social assistance where working conditions, education, and health literacy vary. We aimed to compare medications dispensed to pregnant women across occupations within this sector. MethodsWe conducted a nationwide population-based study using the French EDP-Sante database, linking occupational and socioeconomic data with healthcare reimbursement records. Pregnancies from 2009-2015 among women aged 15-55 years were identified. Medication dispensing during the year before and during pregnancy was described by occupational groups using the Anatomical Therapeutic Chemical classification. Comparisons were performed within healthcare occupations. ResultsAmong 321,569 pregnancies, 38,860 (12%) involved women in healthcare and social assistance. Before pregnancy, 91-95% received at least one medication (mean: 8-12). Psychotropic medication (including antidepressants and benzodiazepines) dispensing was higher among hospital service workers (18% [95% confidence interval (CI):17-19]), healthcare assistants (16% [95% CI:15-16]), and social workers (14% [95% CI:13-15]) than among physicians (11% [95% CI:10-13]), rehabilitation professionals (10% [95% CI:8-12]), and nurses (10% [95% CI:9-10]). Similar patterns were observed for analgesic and musculoskeletal medications. During pregnancy, differences were less pronounced. Folic acid, recommended before and early in pregnancy, was lower among hospital service workers (30% [95% CI:29-32]) than other healthcare workers (39-53%). ConclusionMedication dispensing is common among pregnant healthcare workers, with marked variation across occupations. These differences may reflect disparities in working conditions or socioeconomic level, underlining the need for tailored maternal health support in specific occupational groups.

BackgroundLong-acting injectable buprenorphine (LAIB) has been positioned as a potentially transformative option for opioid use disorder (OUD), in part because patient experiences reported in qualitative studies emphasize reduced daily burden, increased "freedom," reduced stigma, and fewer pressures related to diversion--while also noting barriers such as insufficient information, early adverse experiences, and concerns about coercion. MethodsWe conducted a cross-sectional online survey of adults recruited from the Behavioral Health Research Panel (BHRP). Eligibility included age [&ge;]18, English literacy, and OUD diagnosis or problematic opioid use within the past 5 years. Survey content assessed buprenorphine experience, knowledge and attitudes toward LAIB, attribute preferences, and open-text feedback. Descriptive statistics were generated; analyses were stratified by buprenorphine experience (experienced vs naive). ResultsAmong 105 participants, 82.9% reported prior buprenorphine use, and 17.1% were buprenorphine-naive. Overall, 53.3% preferred a long-acting injection regimen (weekly/monthly/3-monthly) versus 46.7% preferring a daily oral tablet/film. Convenience and adherence-related themes (e.g., not missing doses, fewer visits) drove LAIB preference, while oral-route preference and concerns about side effects and safety were prominent among those favoring oral formulations. ConclusionsIn this national convenience sample, preferences were nearly evenly split between daily oral and long-acting injectable buprenorphine regimens, with a slight overall preference for LAIB. Findings align with the qualitative literature, emphasizing the practical and psychosocial benefits of LAIB, alongside persistent needs for improved education, shared decision-making, and attention to tolerability, safety perceptions, and cost/coverage barriers.

ImportanceGlucagon-like peptide-1 receptor agonists (GLP-1RAs) are fast-growing treatments for type 2 diabetes, obesity, and sleep apnea and are under investigation as potential treatments for many other conditions. The National Institutes of Healths (NIHs) All of Us Research Program offers a robust observational data source for studying questions related to GLP-1RA use in real-world settings. ObjectiveThis article describes key characteristics of All of Us participants who have been prescribed GLP-1RAs. The goals are to present the utility of the All of Us data and describe the strengths and limitations of using this resource for future research on GLP-1RAs. DesignUsing the All of Us Controlled Tier Curated Data Repository version 8 (CDRv8), we provide a descriptive analysis of the cohort with GLP-1RA records using cross-sectional surveys, longitudinal electronic health record (EHR) data, and longitudinal Fitbit data. SettingThe All of Us Research Program is a large, federally funded, longitudinal cohort study established in 2018 by NIH. Recruitment efforts are nationwide and target a range of populations to advance precision medicine for all. ParticipantsParticipants are U.S. residents, aged 18 or older at the time of study consent, who were enrolled between May 6, 2018, and October 1, 2023. ExposuresThe GLP-1RA cohort included participants with at least two GLP-1RA prescription records on different days at any time point based on their EHRs. Main OutcomesFrequencies and medians for a range of sociodemographic characteristics, health care utilization patterns, comorbid conditions, GLP-1RA prescription trends, laboratory and observation availability, and Fitbit data. ResultsThe All of Us GLP-1RA cohort is large (n=15 477), with high data availability across a range of relevant data types. These participants are older and have more comorbid conditions than the entire CDRv8 population. Prescription trends indicate rapid uptake of GLP-1RA drugs since 2014. Conclusions and RelevanceAll of Us CDRv8 is a valuable resource for research on GLP-1RAs across a large, heterogeneous cohort of participants. The variety and availability of data offer many possibilities for future observational, real-world research to address unanswered questions about GLP-1RA use and replicate recent findings generated from other datasets. Key pointsO_ST_ABSQuestionC_ST_ABSWhat data are available and what are the patterns of GLP-1 receptor agonist (GLP-1RA) prescriptions among participants in the All of Us Research Program? FindingsIn this descriptive cohort study of 633 534 All of Us participants, 15 477 participants had at least two records of GLP-1RA prescriptions. These participants tended to be older, have more comorbid conditions, and have higher health care utilization than the All of Us population as a whole. MeaningThe All of Us Research Program has a robust array of data to support observational studies of people receiving GLP-1RA prescriptions.

BackgroundSafety data for most medications in pregnancy remain limited, yet pharmacological treatment is often necessary. Evidence on real-world medication use in pregnancy including over-the-counter products and folic acid is scarce, especially in Belgium. MethodsWe conducted a drug utilisation study using self-reported data from BELpREG, a prospective, web-based pregnancy registry established in November 2022. Pregnant individuals aged [&ge;]18 years receiving healthcare in Belgium can enrol voluntarily at any stage in pregnancy and complete online questionnaires at enrolment and every four weeks until delivery. All participants with follow-up beyond the first trimester were included, and trimester-specific cohorts were constructed based on completion of questionnaires after each trimester. Data were extracted in July 2025. ResultsThis study included 2,096 participants, of whom 1,767 were followed through trimester 2 and 1,136 through trimester 3. Median gestational age at enrolment was 16 weeks. Prevalence estimates of medication use were 80.2% in the six months before conception, 85.8% in trimester 1, 92.0% in trimester 2, and 94.9% in trimester 3. The most common classes were analgesics, vaccines, antihistamines, antianemic preparations, and drugs for acid-related disorders. Paracetamol was most frequently used (35.4% in trimester 1), typically short term (median 3 days), followed by doxylamine-pyridoxine (26.7% in trimester 1). Folic acid supplementation was nearly universal, though only 59.9% met national guideline-concordant criteria. Maternal vaccine uptake was substantial but incomplete, with 67.2% receiving pertussis, 41.5% influenza, and 21.5% COVID-19 vaccination. Exposure to potentially inappropriate or teratogenic medications was rare. ConclusionsMedication use during pregnancy in Belgium was nearly universal, with high use of paracetamol and doxylamine-pyridoxine. Folic acid and vaccine uptake were substantial, but often not guideline-concordant. Key PointsO_LIMedication use during pregnancy in Belgium was nearly universal, with over 85% of participants reporting use in the first trimester and 95% in the third. C_LIO_LIParacetamol (35% in the first trimester) and doxylamine-pyridoxine (27% in the first trimester) were the most frequently used medications. C_LIO_LIFolic acid use was widespread, yet only about 60% of participants followed national timing and duration recommendations. C_LIO_LIMaternal vaccine uptake was substantial, particularly for pertussis (67%), though not universal despite guideline recommendations C_LIO_LIBELpREGs self-reported data capture both prescription and over-the-counter medications, offering a complete picture of real-world use during pregnancy. C_LI Plain Language SummaryThis study looked at how often and when people in Belgium use medications, folic acid supplements, and vaccines during pregnancy. Using data from the BELpREG pregnancy registry, more than 2,000 pregnant participants completed online questionnaires about their health and medication use throughout pregnancy (every four weeks). Almost everyone reported taking at least one medication: 86% during the first trimester and 95% during the third. The most common medicines were paracetamol and doxylamine-pyridoxine. Nearly all participants used folic acid, but only about 60% followed national recommendations for starting timely before pregnancy and continuing through the first trimester. Many received recommended vaccines during pregnancy: about 67% for pertussis, 42% for influenza, and 22% for COVID-19; but uptake was still incomplete. Exposure to potentially inappropriate or teratogenic medications was rare. Because BELpREG collects self-reported data, including both prescribed and over-the-counter products, it provides a comprehensive picture of real-world medication use in pregnant people. Further, these findings help identify gaps between guideline recommendations and actual practice. Social Media QuoteMedication use in pregnancy is nearly universal in Belgium. Paracetamol and doxylamine-pyridoxine top the list. Folic acid and vaccine uptake are high but often not guideline-concordant. BELpREG data reveal unique self-reported real-world patterns. #BELpREG

BackgroundThe self-management of type 2 diabetes (T2D) typically requires enacting various lifestyle changes, which can challenge people living with T2D. Clinical encounters between people with T2D and their clinicians, however, are often focused on metabolic management, leaving less time available for other self-management topics. The QBSAFE cards help patients articulate aspects of their experience with diabetes and prioritize issues for discussion. MethodsThis report details secondary outcomes of a randomized controlled trial; primary outcomes are reported elsewhere. All data was collected at Fair Haven Community Health Care, a federally qualified primary care clinic. 11 clinicians were randomly assigned to provide either usual care or usual care with QBSAFE cards to 155 of their patients with type 2 diabetes and hemoglobin A1c >8%. All patient encounters were video recorded for analysis. Patients and clinicians were not blinded to arm allocation but were kept unaware of the specific aims of the trial. Encounter video reviewers were blinded to arm allocation, but not to specific aims of the trial. The outcomes of interest for this report were the extent to which the QBSAFE cards were used as intended, their effect on the topics of discussion, and whether they enabled clinicians to notice and respond to each patients situation; comparisons between arms were conducted by a linear mixed model with fixed effect of arm and cluster effect of clinician, analyzed in both intent-to-treat and per-protocol populations. Findings12 patients were excluded post-randomization (A1c <8%). Of 143 eligible patients, 137 encounters (65 in the usual care arm, 72 in QBSAFE) yielded evaluable videos. QBSAFE was used as intended in 61 (85%) QBSAFE arm encounters. Conversations about burden of treatment related to non-pharmacological interventions (17 vs 33, p= 0{middle dot}04) and taking medications (11 vs 33, p= 0{middle dot}0008) and about the patients challenging environment (2 vs 10, p= 0{middle dot}04) were more prevalent in the QBSAFE group. There was no difference in the rate of conversations about metabolic management or of new care plans as a result of conversations between groups. InterpretationWhile there was a difference in the types of conversations observed between the two study arms, this difference was small and only apparent in a few domains. Future work could aim to modify the QBSAFE cards to more effectively stimulate patient-centered discussions and to further prepare clinicians to respond to a variety of issues raised during the clinical visit. FundingThis work was supported by funding from the National Institute of Diabetes and Digestive and Kidney Diseases (R01DK129616).

Key pointsO_LIWe report findings from a phase 2 study of MY008211A among Chinese men and women aged [&ge;]18 years with paroxysmal nocturnal hemoglobinuria C_LIO_LIIncreases in hemoglobin of [&ge;]20 g/L were maintained for up to 44 weeks of treatment with MY008211A in all 34 patientsiv C_LI Explanation of noveltyParoxysmal nocturnal hemoglobinuria is characterized by red blood cell (RBC) destruction and a prothrombotic state.v Treatments exist such as complement 5 inhibitors but these carry the risk for iatrogenic extravascular hemolysis and anemia.vi As reported here, the novel, oral complement factor B inhibitor MY008211A yielded increases in hemoglobin and RBC levels, while adverse events over 44 weeks were largely mild to moderate in severity, and infections generally consisted of respiratory infections.vii Paroxysmal nocturnal hemoglobinuria (PNH) is a life-threatening disease characterized by red blood cell (RBC) destruction, blood clots, and impaired bone marrow function.viii We evaluated the efficacy and safety of 3 dosages of MY008211A, a novel complement factor B inhibitor,ix for treating PNH.x This was a multicenter, open-label, phase 2, dose-finding study of MY008211A among Chinese men and women with complement inhibitor-naive PNH and signs of active hemolysis.xi Patients with hemoglobin <100 g/L were assigned to oral MY008211A 400 mg twice daily (BID), 600 mg BID, or 800 mg once daily (QD) for 12 weeks and could then continue treatment with 400 mg BID during a 32-week extension.xii The primary endpoint was the proportion of patients achieving an increase in hemoglobin concentration of [&ge;]20 g/L vs baseline on day (D)84, without RBC transfusions after 4 weeks of dosing.xiii Safety assessments included adverse events (AEs).xiv Fifteen, 9, and 10 patients were assigned to MY008211A 400 mg BID, 600 mg BID, and 800 mg QD, respectively.xv All patients completed the study and its 32-week extension.xvi On D84, all 34 patients achieved increases in hemoglobin concentration of [&ge;]20 g/L from baseline;xvii all patients maintained this increase at D308.xviii Through D308, grade [&ge;]3 AEs occurred in 5 (33%), 5 (56%), and 4 (40%) patients in the 400-, 600-, and 800-mg groups, respectively.xix There were no deaths.xx In this multicenter, open-label study of 3 dosages of MY008211A for PNH, all patients achieved and maintained increases in hemoglobin of [&ge;]20 g/L from baseline without RBC transfusions.

ObjectivesTo estimate potential launch prices of generic semaglutide following patent expiry from 2026 and to quantify the global obesity and type 2 diabetes (T2DM) burden in countries where generic access may become possible. MethodsWe used World Bank population data and World Obesity and Diabetes Atlas prevalence estimates to calculate obesity and T2DM burden in countries where semaglutide patents expire in 2026 or were not filed. Patent status was identified using MedsPaL and cross-checked with regional databases. We updated established cost-plus pricing methodologies using 2024-2025 Indian API shipment data to estimate production costs for oral and injectable semaglutide, incorporating formulation, packaging, taxation, and profit assumptions. ResultsTen countries with 2026 patent expiry represent 44% of the global population and 48% of the global obesity burden. No patent filings were identified in 150 additional countries. By the end of 2026, generic injectable semaglutide could be distributed in 160 countries where 69% of global T2DM and 84% of clinical obesity occurs. Estimated generic injectable costs ranged from $28-140 per person-year, while oral formulations ranged from $186-380 per person-year. Injection devices contributed disproportionately to total cost. ConclusionPatent expiry could substantially expand access to semaglutide at dramatically lower prices, particularly in high-burden settings. However, device costs, secondary patents, and health system constraints may limit equitable uptake without coordinated policy action. Study ImportanceO_ST_ABSWhat is already known about this subject?C_ST_ABSO_LISemaglutide is highly effective for obesity and cardiometabolic disease but remains unaffordable in many low- and middle-income countries due to high branded prices and patent protections. C_LIO_LIPrevious cost-plus analyses show that generic competition can substantially reduce prices of essential medicines after patent expiry. C_LI What are the new findings in your manuscript?O_LIUsing 2024-2025 API shipment data, we estimate generic injectable semaglutide could be produced for $28-140 per person-year following 2026 patent expiry. C_LIO_LIBy 2026, generic semaglutide could be available in 160 countries comprising 69% of global T2DM and 84% of clinical obesity burden. C_LI How might your results change the direction of research or the focus of clinical practice?O_LIProvides an evidence base for procurement planning and price negotiations ahead of patent expiry. C_LIO_LIHighlights the importance of addressing device costs and secondary patents to ensure equitable global access. C_LI

BackgroundUsing the FDA Adverse Event Reporting System (FAERS) database, this pharmacovigilance investigation systematically assessed the adverse events (AEs) associated with Fedratinib use in real-world clinical practice. MethodsUsing the FAERS database, we performed a disproportionality analysis incorporating four distinct signal detection methodologies: ROR, PRR, BCPNN, and MGPS. Subgroup analyses were conducted to evaluate the effects of age and gender on Fedratinib-associated AEs. Furthermore, a time-to-onset analysis was performed to characterize the temporal patterns of AEs. ResultsThe FAERS database comprised 10,011,422 individual case safety reports, of which 1,284 were classified as Fedratinib-related AEs, encompassing 38 significant preferred terms (PTs). The most frequently reported adverse reactions included diarrhoea, nausea, vomiting, constipation, abdominal discomfort, fatigue, anaemia, platelet count decreased, and Wernickes encephalopathy. New AEs emerged from the study, such as blood potassium increased, hyperkalaemia, gout, hypocalcaemia, renal failure, renal disorder, and gallbladder disorder. Higher rates were observed in males over 65 years of age. Most cases occurred within the first month of Fedratinib treatment, with the incidence of related AEs decreasing over time. ConclusionThis current study marks the debut investigation regarding Fedratinibs safety in actual clinical practice, which providing substantive evidence to inform future pharmacovigilance investigations of this drug.

Artificial intelligence (AI), particularly large language models (LLMs), is increasingly explored in healthcare, yet its real-world usability and safety in high-risk clinical pharmacy tasks remain uncertain. Vancomycin therapeutic drug monitoring (TDM), which requires precise pharmacokinetic calculations and context-sensitive interpretation within a narrow therapeutic window, provides a stringent test case for AI-assisted decision support. This proof-of-concept study developed and evaluated a hybrid clinical decision support system (TDM-AID) integrating a validated deterministic pharmacokinetic calculation engine, GPT-4o-based structured clinical interpretation, and retrieval-augmented guideline support. Thirty retrospective adult vancomycin TDM cases were assessed using a weighted six-domain rubric covering pharmacokinetic accuracy, AUC estimation, prospective prediction, timing recommendations, clinical judgment, and documentation quality. Two independent expert pharmacists evaluated system outputs against benchmark consultations. The overall median performance was 78% (IQR 12%), classified as Acceptable, and 73% (IQR 14%) when deterministic calculations were excluded. Foundational pharmacokinetic calculations achieved 100% accuracy. Clinical judgment demonstrated Good performance (83%), whereas prospective prediction was limited (58%), and timing recommendations were absent in all cases. Safety violations occurred in 17% of cases, including dose recommendations exceeding 4 g/day. Inter-rater reliability was good (ICC 0.87). These findings suggest that hybrid AI-driven decision support is technically feasible and usable as a pharmacist-augmenting draft generator; however, limitations in predictive reasoning, timing logistics, and safety enforcement necessitate deterministic safeguards and mandatory expert oversight before clinical implementation.

BackgroundGLP-1 receptor agonists (GLP-1RAs) and SGLT2 inhibitors (SGLT2Is) have established cardiovascular benefits for patients with type 2 diabetes mellitus (T2DM), with similar class-level effectiveness found in previous studies. However, real-world comparative effectiveness assessments of individual agents remain limited. ObjectivesTo compare the cardiovascular effectiveness of individual GLP-1RAs and SGLT2Is. MethodsWe conducted a multi-national, retrospective, new-user active-comparator cohort study using 10 US and non-US administrative claims and electronic health record databases. The study included 1,245,211 adults with T2DM receiving metformin who initiated second-line therapy with one of six GLP-1RAs (albiglutide, dulaglutide, exenatide, liraglutide, lixisenatide, semaglutide) or one of four SGLT2Is (canagliflozin, dapagliflozin, empagliflozin, ertugliflozin). Empagliflozin (393,499; 31.6%), semaglutide (235,585; 18.9%), dapagliflozin (208,666; 16.8%), and dulaglutide (207,348; 16.8%) were most commonly used. A secondary subgroup analysis included 316,242 patients with established cardiovascular diseases (CVD). Primary outcomes were 3-point major adverse cardiovascular events (MACE: acute myocardial infarction, stroke, sudden cardiac death) and 4-point MACE (adding hospitalization/ER visit with heart failure). Secondary outcomes included the individual components. Hazard ratios (HRs) were estimated for pairwise agent comparisons while on-treatment (per-protocol) and over total follow-up using Cox proportional hazards models, with propensity score adjustments, negative control calibration, and pre-specified study diagnostics to guard against potential confounding. Random-effects meta-analysis produced summary HR estimates across data sources that passed diagnostics. ResultsAcross the study cohort, individual GLP-1RAs and SGLT2Is demonstrated broadly similar cardiovascular effectiveness, both within and across drug classes. For example, semaglutide and empagliflozin showed comparable risks for 3-point MACE (meta-analytic HR 1.05; 95% CI 0.79-1.39) and 4-point MACE (meta-analytic HR 0.95; 95% CI 0.81-1.12), with consistent findings in the CVD subgroup. Study diagnostics confirmed adequate equipoise, covariate balance and statistical power to detect similarity in HRs between 0.8 and 1.2 for commonly used agents. ConclusionsIn this large-scale real-world study, individual GLP-1RAs and SGLT2Is exhibited largely comparable cardiovascular benefits, including in patients with established CVD. These findings align with network meta-analytic estimates from major cardiovascular outcome trials and broadly support current treatment guidelines. Clinical choices should be guided by relevant factors such as safety, adherence, tolerability, cost, and patient preference, where further work is needed.

Background and AimsThe glucagon-like peptide-1 receptor agonist (GLP-1 RA) semaglutide has demonstrated efficacy for the secondary prevention of cardiovascular disease among patients with overweight/obesity without diabetes mellitus. However, the comparative effectiveness of GLP-1 RA versus other antiobesity medications (e.g. phentermine-topiramate) not been evaluated. MethodsThis was a retrospective, observational, cohort study using target trial emulation methodology using the Truveta electronic health record database of more than 120 million patients. Adult patients with a body mass index (BMI) >=27 kg/m2, a history of cardiovascular disease (prior ischemic stroke, transient ischemic attack, or myocardial infarction, or known coronary artery disease, heart failure, or peripheral artery disease) without diabetes mellitus were included in the study. The primary endpoint was time to first major adverse cardiovascular or cerebrovascular event (MACCE, defined as stroke or myocardial infarction). ResultsIn total, 35,240 were included in the bupropion-naltrexone versus GLP-1 RA comparison, and 27,051 were included in the phentermine-topiramate versus GLP-1 RA comparison. In the pre-weighting cohort, GLP-1 RA use was associated with decreased hazard of MACCE compared to bupropion-naltrexone (HR 0.50 [95% confidence interval (CI) 0.36-0.69]) and phentermine-topiramate (HR 0.43 [95% CI 0.30-0.60]). In the propensity score-overlap weighted cohort, GLP-1 RA prescription was not associated with a lower hazard of MACCE than bupropion-naltrexone (aHR 0.69 [95% CI 0.47-1.00]) but was associated with a lower hazard compared to phentermine-topiramate (aHR 0.61 [95% CI 0.41-0.91]; adjusted absolute rate difference 0.98 per 1000 person-years). ConclusionsPrescription of a GLP-1 RA was associated with a lower risk of subsequent MACCE than phentermine-topiramate.

BackgroundMedication shortages are a considerable and ongoing issue in healthcare, disrupting consumer access. Since 2021, Australias national medicines regulator has issued Serious Scarcity Substitution Instruments (SSSIs), allowing pharmacists to substitute a specific therapeutically equivalent strength and/or formulation of a medicine without prior approval from a prescriber. The impact of SSSIs on utilisation of medicines has not been investigated. ObjectiveDetermine whether SSSIs are effective in addressing medicine shortages and meeting patient need. MethodsThis retrospective cohort study used aggregated pharmacy claims to examine the utilisation of 12 medicines which had an SSSI. We calculated the percentage change in defined daily doses dispensed per 1000 population per day in the 11 months after SSSI implementation, compared with the previous two years. A percentage change of less than 20% was used to indicate success. ResultsFollowing product shortages, utilisation fell for 10 of the 12 medicines examined. For eight of these medicines (amoxicillin, cefalexin, estradiol, fluoxetine, insulin degludec with insulin aspart, isosorbide mononitrate, vigabatrin, and warfarin) decreases in utilisation were minimised to <20%. On average, SSSIs where all permitted substitute products were scarce (e.g. abatacept) were associated with larger decreases in use (between -22% and -68%) than those for which none or only some of the substitutes were in shortage (between -45% and +7%, respectively). ConclusionsWhile product shortages led to decreases in medicines consumption, SSSIs appeared to be successful in limiting decreases. However, SSSIs were less likely to be successful when many of the permitted substitute products were also scarce. Key pointsO_LIThis study is the first to evaluate the effectiveness of Australias Serious Scarcity Substitution Instruments (SSSIs) in mitigating medicine shortages using national dispensing data and interrupted time series analysis. C_LIO_LITwo-thirds of SSSIs successfully limited utilisation declines to less than 20%, with effectiveness strongly linked to the availability of substitute products. C_LIO_LIBy demonstrating variable utilisation outcomes across medicines, this study adds empirical evidence to international debates on substitution policies, suggesting that nationally standardised frameworks like Australias SSSIs may function best when supported by robust supply intelligence. C_LIO_LISSSIs are a valuable policy tool for maintaining continuity of care during shortages, but timely implementation and ensuring substitute supply are critical for optimal impact. C_LI

IntroductionRecreational and medical cannabis use (CU) information is often available within the electronic health record (EHR) in a format that is impractical for health care provider use. Transformation of free-text EHR documentation in notes to discrete elements is possible using natural language processing (NLP) and has the potential to characterize CU efficiently. The objective of this study was to develop an NLP algorithm to identify documentation of CU within EHR unstructured clinical notes. MethodsWe identified EHR notes with cannabis-related terminologies through a keyword search among all Geisinger patients with at least one encounter between 1/1/2013 and 6/30/2022. We trained four NLP models to classify notes into six categories based on time, context, and reliability of CU documentation identified through manual annotation. We compared the demographic characteristics of patients with positive classification for CU using the best-performing model to those of the overall population. ResultsOf the over 1.7 million eligible patients, 150,726 (8.6%) were flagged as cannabis users. The Bio-ClinicalBERT, a transformer-based NLP model, achieved close to human performance in classifying CU (weighted Precision=91.4, Recall=93.3, F-score=92.4). Cannabis users had higher BMI and were at least nine-fold more likely to use tobacco, alcohol, and illicit substances. ConclusionOur study evaluated the prevalence of CU documentation across the entire corpus of EHR notes data without population segmentation. The NLP methodologies used achieved performance close to that of human annotation and laid the foundation for identifying and classifying CU within unstructured data sources, with future applications in research and patient care. Plain Language SummaryMarijuana, also known as cannabis, may impact the health of patients, yet it is not routinely captured in medical records, and when documented, it is often found in unstructured formats (e.g., progress notes) rather than in discrete fields. Incomplete and unstructured capture limits many functional capabilities within the EHR that enhance patient care (e.g., drug interactions, notifications) and limit researchers from identifying patients routinely exposed to marijuana use. The transformation of free-text documentation of cannabis use (CU) into discrete elements can be performed using natural language processing (NLP). The objective of this study was to develop an NLP model to identify CU in unstructured clinical notes in the EHR. We examined the EHRs of Geisinger patients in Pennsylvania over a 10-year period. Among 1.7 million patients, 9% were identified as CU. One of the NLP models tested, Bio-ClinicalBERT, achieved the highest performance. Cannabis users had a higher BMI and were ten-fold more likely to be tobacco users, ten-fold more likely to use alcohol, and nine-fold more likely to use illicit substances. NLP can be used to better understand the risks and benefits of CU at a population level and may improve patient identification to assist clinical decision-making. Future CU epidemiological research should continue to explore other avenues to automate and improve CU documentation by leveraging rapidly evolving technologies, such as artificial intelligence-driven tools.

BACKGROUNDThis study evaluates the reach, scalability, and implementation of a large-scale, multi-site tele-buprenorphine program designed to treat opioid use disorder (OUD) within rural carceral settings. Given that individuals transition frequently between jails and the community, these facilities represent a critical window for OUD intervention, yet they often face significant provider shortages and logistical barriers. We conducted a retrospective chart review of 842 unique patients (1,321 treatment episodes) enrolled in the University of Marylands tele-buprenorphine program across six rural county jails between June 2020 and May 2025. Data extracted from jail records and electronic health records were used to analyze patient demographics, prescribing patterns, and program retention. RESULTSThe patient population was primarily male (71.1%) and White (75.7%), with a mean age of 35.4 years. Participants reported high-severity OUD, with an average of 12.6 years of opioid use. Reflecting broad admission criteria, 55.2% of participants were new treatment initiates not receiving MOUD prior to booking. Patients spent a mean of 35.6 days incarcerated before initiation and were retained in the program for an average of 66 days. Buprenorphine doses were titrated from a mean initiation dose of 8.8 mg to 16.2 mg at discharge. The program demonstrated a 99.5% adherence rate among retained patients. Only 3% of the total sample were discharged for medication diversion or hoarding. CONCLUSIONSTelemedicine is a highly feasible and scalable model for delivering evidence-based MOUD in rural jails. By utilizing a "liberal admission policy" that prioritizes both treatment initiation and maintenance, programs can successfully reach high-risk individuals who lack access to community-based care. These findings suggest that tele-buprenorphine can effectively bridge the treatment gap in underserved jurisdictions, potentially reducing the risk of overdose during the high-risk post-release period.

ObjectivesThe 2025 Global Consensus recommends continuing biologics throughout pregnancy in women with inflammatory bowel disease (IBD). Real-world evidence on biologic treatment patterns and outcomes remains limited. This study compared maternal and neonatal outcomes across different biologic use trajectories during pregnancy. MethodsA retrospective study was performed in pregnant women with IBD, treated and/or delivering at the University Hospitals Leuven, Belgium, between 2017 and 2025. Patients were categorized as continuers, discontinuers, non-users or initiators of biologics during pregnancy ResultsAmong 255 pregnancies, 103 (40.4%) were continuers, 68 (26.7%) discontinuers, 77 (30.2%) non-users, and 7 (2.7%) initiators. Before conception, 67.1% used biologics. Third-trimester disease activity was most frequent in initiators (42.9%, 3/7) and discontinuers (19.1%, 13/68), followed by non-users (14.3%, 11/77) and continuers (13.6%, 14/103). C-sections occurred more often in non-users (41.3%, 26/63) and discontinuers (39.4%, 26/66) than continuers (31.1%, 23/74). Preterm birth was more common among initiators (14.3%, 1/7) and discontinuers (12.1%, 8/66) than continuers (8.0%, 6/75) and non-users (3.2%, 2/62). Low birthweight occurred most in initiators (14.3%, 1/7), continuers (8.1%, 6/74) and discontinuers (6.1%, 4/66). Small-for-gestational-age infants were most frequent among continuers (14.9%, 11/74) and initiators (14.3%, 1/7) than discontinuers (7.6%, 5/66). ConclusionsWomen who discontinued biologics during pregnancy had higher rates of C-sections, preterm birth, and third-trimester disease activity than continuers, supporting continuation of biologics in pregnancy. The higher SGA rates among continuers, however, require further investigation. Initiators showed the poorest outcomes, highlighting the need for adequate disease control before and during pregnancy.