Pharmacoepidemiology and Drug Safety最新文献

One of the key challenges in pharmacoepidemiological studies is that of uncontrolled confounding, which occurs when confounders are poorly measured, unmeasured or unknown. Self-controlled designs can help address this issue, as their key comparison is not between people, but periods of time within the same person. This controls for all time-stable confounders (genetics) and in the absence of time-varying confounding negates the need for an external control group. However, these benefits come at the cost of strong assumptions, not all of which are verifiable. This review briefly introduces the reader to different types of self-controlled study designs, their terminology and highlights key publications through an annotated reference list. We include a practical description of how these designs can be implemented and visualised using recent examples, and finish by discussing recent developments. We hope this review will serve as a starting point for researchers looking to apply self-controlled designs in their own work.

Background: Drug-drug interactions (DDIs), highly prevalent amongst the elderly, can lead to avoidable medication-related harm. Cardiovascular and central nervous system (CNS) drugs are commonly implicated. To date, there is no consensus on how to measure DDIs, making comparisons across countries challenging.

Objective: To (i) establish a common data model (CDM) to measure DDI prevalence in the older (aged ≥ 70 years) community-dwelling population of three European countries and (ii) compare and describe cardiovascular and CNS DDI prevalence rates across these countries.

Methods: This cross-country study will apply a harmonised method of DDI identification and analysis using the WHO ATC classification system and national pharmacy claims data from three European countries (Ireland, Italy, Spain). Patients aged ≥ 70 years dispensed ≥ 2 medications during 2016 will be identified from each country's national database. 'Severe' cardiovascular and CNS DDIs (i.e., may result in a life-threatening event/permanent detrimental effect) will be identified using the British National Formulary and Stockley's Drug Interactions. Two separate lists of 'severe' DDIs, per medications reimbursed, will be applied to each database: (i) DDIs relevant to each individual country and (ii) DDIs relevant to all three countries. DDIs will be defined as co-dispensed (same day) and concomitantly (±7 days) dispensed.

Results: Descriptive statistics, including DDI prevalence and 95% confidence intervals, will be reported for each country. Prevalence will be pooled and compared across countries using random effects models and meta-regression, where feasible.

Conclusion: The EuroDDI study will develop a harmonised method to measure and compare DDI prevalence across health-related databases in Europe.

Purpose: This study aimed to obtain a better understanding of the characteristics of the risk management plans (RMP) and the background regulatory policies governing them, in the European Union (EU) and Japan. This was done by descriptively comparing the safety concerns (SCs) listed in the RMP and examining their relationships with product labeling.

Methods: Information regarding SCs was collected from the published RMP of both the EU and Japan for the targeted products-all of which were commonly approved in both regions. The concordance rate of the SCs for each product between the EU- and Japan-RMP was calculated. The warning information for each product was collected from the product labeling, summary of product characteristics for the EU, and package insert for Japan, and compared with the SCs listed in the corresponding RMP.

Results: A total of 259 products that were approved for sale in both the EU and Japan (1998-2023), for which RMP were available in both regions, were analyzed. While 51.0% of the SCs labeled as important identified risks (IIRs) in the EU-RMP were concordant with those in the Japan-RMP, 20.4% of the SCs listed as IIRs in the Japan-RMP were concordant with those in the EU-RMP. The concordance rate between the SCs identified as IIRs and the warning information was 18.6% for the EU-RMP and 88.4% for the Japan-RMP.

Conclusions: The low SC concordance rate between the EU- and Japan-RMP indicates a different approach to selecting RMP SCs by the two regulatory authorities.

Purpose: To assess whether exposure to an extended-release (ER) oxycodone with abuse deterrent properties (ADF) reduced tampering of oxycodone in a real-world, postmarket setting to address the thinking behind Category 4 labeling by the FDA.

Methods: Data from an observational cross-sectional study of the general adult population (2022) was used under a causal framework to estimate the confounding-adjusted odds of tampering oxycodone after exposure to two types of ADF ER oxycodone. The tampering behaviors of those who used only single entity immediate-release (SE-IR) oxycodone was used as a comparison. The tampering outcome was defined as use by snorting, smoking, or injecting any oxycodone (ER or SE-IR). A directed acyclic graph was used to identify covariates. Average treatment effect among the treated was estimated using inverse propensity score weighting combined with survey weights in a regression.

Results: In 2022, 0.14% and 3.0% among the general population reported using the two ER oxycodone groups, while 2.4% used SE-IR oxycodone. Propensity score analyses with both comparators (common support > 98%) balanced demographic, health, and drug use covariates. After adjustment for selection and confounding bias, among those who used ER oxycodone group 1 the odds ratio of tampering with any form of oxycodone was elevated but not statistically significant (2.25; 95% CI: 0.94, 5.39). The odds ratio of tampering by users of ER oxycodone group 2 was significantly elevated (1.90; 95% CI: 1.08, 3.19).

Conclusions: Tampering of ER oxycodone products by individuals was rare. We found evidence suggestive of elevated odds of tampering behaviors with use of an ADF ER oxycodone.

Purpose: To comply with the Health Insurance Portability and Accountability Act of 1996 (HIPAA) Privacy Rule, many real-world data providers mask a patient's date of birth by supplying only year of birth to data users. The lack of granularity around patient age is a challenge when using RWD, especially for pediatric research studies. In this study, a proxy for patient date of birth is evaluated using electronic health record (EHR) data.

Methods: This validation study leverages a retrospective cohort of EHR data from Mass General Brigham (MGB) patients born between January 1, 2018, and December 31, 2022, to assess the use of the date of a patient's first observed International Classification of Diseases 10th Revision Clinical Modification (ICD-10-CM) day-of birth code (Z37* or Z38*) as a proxy for date of birth. Alternative proxy measures such as date of first other infancy-related ICD-10-CM code and date of first clinical activity were also assessed.

Results: Of 82 398 patients born during the five-year study period, 58 047 (70.4%) had an ICD-10-CM birth code and were included in the primary analysis. The mean difference between true date of birth and first observed birth code was 0.3 days with a standard deviation of 15.0 days. The first observed birth code occurred within 30 days of the true date of birth in 99.9% of cases.

Conclusion: Results from this study suggest that the date of the first day-of ICD-10-CM birth code can be used as a proxy for true patient date of birth in pediatric RWD studies.

Purpose: To characterize select laboratory tests ordered versus reported for patients diagnosed with COVID-19 in administrative healthcare and commercial laboratory data.

Methods: Among patients with an outpatient COVID-19 diagnosis claim in HealthVerity data (01/01/2021-12/31/2022), this study described baseline characteristics and descriptively compared SARS-CoV-2 diagnostic tests and liver function tests from administrative healthcare (insurance claims and hospital billing data) and commercial laboratories, overall and by code type (e.g., CPT, LOINC). Select liver function tests were also described by method-specific and methodless LOINC.

Results: Among 214 998 patients with COVID-19, 46.1% had a SARS-CoV-2 molecular diagnostic test recorded within 7 days of diagnosis (in either administrative or laboratory data); 44.5% had a corresponding CPT in medical claims, while only 10.0% had a corresponding LOINC in laboratory data. In contrast, the six most common liver function tests (albumin, aspartate aminotransferase, total protein, alkaline phosphatase, alanine aminotransferase, and total bilirubin) were identified in 55.7%-56.6% of patients via LOINC, but only in 3.2%-4.2% via CPT claims. Of the total count of select liver function tests performed in the laboratory data, 99.7% of aspartate aminotransferase, 96.1% of direct bilirubin, and 82.9% of lactate dehydrogenase were reported by methodless LOINC rather than method-specific LOINC.

Conclusions: Important differences were identified between orders for SARS-CoV-2 diagnostic tests and liver function tests, as well as missingness of LOINC method, highlighting challenges related to completeness of laboratory data in real-world data sources. These challenges underscore a need to improve data quality when considering the utility of laboratory data for research.

Purpose: Psoriasis (PsO), a chronic inflammatory skin disorder affecting a substantial proportion of populations globally, often necessitates systemic treatment including biologics. This 14-year cohort study, based on Danish national register data, aimed to investigate the enduring safety profile of ustekinumab compared to other systemic psoriasis treatments.

Methods: Using comprehensive Danish national register data, this study scrutinized patients diagnosed with psoriasis or psoriatic arthritis (PsA) who received ustekinumab. The treatment group comparators were non-biological systemic treatment (non-biologic), tumor necrosis factor α inhibitor medicine groups (TNF-α), interleukin (IL)-17 inhibitors (IL-17), and IL-23 inhibitors (IL-23). The study periods for comparisons were 2009-2022 for non-biologic and TNF-α, 2015-2022 for IL-17, and 2018-2022 for IL-23. Outcomes were malignancies, cardiovascular events, serious infections, and serious hypersensitivity reactions. Cox proportional hazards regression models were employed to analyze two estimands: a standard intention-to-treat (ITT) estimand and a continuous-index-treatment (CIT) estimand, which considered switch and re-initiation of treatments within individuals.

Results: Users of ustekinumab were found to be younger on average, with an average age of 45.1 years compared to 51.6, 47.2, 49.0, and 48.4 years in the non-biologic, TNF-α, IL-17, and IL-23 groups, respectively. Also, 57.3% of the ustekinumab users were male, compared to 46.7%, 48.9%, 50.9%, and 58.3% for the non-biologic, TNF-α, IL-17, and IL-23 groups, respectively. Although the hazard ratio estimates varied across comparators, ustekinumab was found to be safe: regardless of PsA status, no discernible safety signals in terms of malignancy, MACE, severe infections, or severe hypersensitivity reactions were observed for ustekinumab when compared to the treatment comparators.

Conclusions: The present study corroborated the enduring safety of ustekinumab in the context of PsO treatment.

Objective: To enhance documentation on programming decisions in Real World Evidence (RWE) studies.

Materials and methods: We analyzed several statistical analysis plans (SAP) within the Vaccine Monitoring Collaboration for Europe (VAC4EU) to identify study design sections and specifications for programming RWE studies. We designed a machine-readable metadata schema containing study sections, codelists, and time anchoring definitions specified in the SAPs with adaptability and user-friendliness.

Results: We developed the RWE-BRIDGE, a metadata schema in form of relational database divided into four study design sections with 12 tables: Study Variable Definition (two tables), Cohort Definition (two tables), Post-Exposure Outcome Analysis (one table), and Data Retrieval (seven tables). We provide a guide to populate this metadata schema and a Shiny app that checks the tables. RWE-BRIDGE is available on GitHub (github.com/UMC-Utrecht-RWE/RWE-BRIDGE).

Discussion: The RWE-BRIDGE has been designed to support the translation of study design sections from statistical analysis plans into analytical pipelines and to adhere to the FAIR principles, facilitating collaboration and transparency between researcher and programmers. This metadata schema strategy is flexible as it can support different common data models and programming languages, and it is adaptable to the specific needs of each SAP by adding further tables or fields, if necessary. Modified versions of the RWE-BRIGE have been applied in several RWE studies within VAC4EU.

Conclusion: RWE-BRIDGE offers a systematic approach to detailing variables, time anchoring, and algorithms for RWE studies. This metadata schema facilitates communication between researcher and programmers.

Purpose: The Excellence Network in RheumatoloGY (ENRGY) was founded in 2021 and encompasses data from more than 700 private practice rheumatology providers throughout the United States, forming a practice-based research network (PBRN).

Methods: Electronic health record (EHR) data from participating practices are aggregated, including structured data (e.g., clinical assessments) and unstructured data from two different EHR platforms. Targeted data quality efforts ensure capture of high-quality clinical data and reduce missingness. ENRGY network membership also provides participating sites access to technology services that enhance patient care. Centralized ethics approval and pre-existing legal agreements along with electronic tools to curate data and contact eligible study participants improves efficiency of multi-center prospective studies.

Results: The ENRGY data warehouse includes linked administrative claims data for commercial and government-provided insurance (31% of patients linked to commercial claims) and patient-generated health data from both in-office and out-of-office settings via a smartphone app as well as biosensor data. ENRGY data and infrastructure can be employed to identify the highest yield sites for prospective studies; identify patients meeting study eligibility criteria; pre-screen individual patient's willingness to participate in specific studies; centralize study data monitoring; and assist in the conduct of prospective studies.

Conclusion: ENRGY data have been used to generate a growing number of scientific publications and may serve as a model for PBRNs in other specialties seeking to harness the potential of data linkages, patient-generated data capture, and centralized study infrastructure for observational and interventional research.

Background: Studies evaluating the hepatic safety of medications have been limited by the inability to control for confounding from receipt of other hepatotoxic drugs.

Objective: The objective of this study was to develop an index (Hepatotoxicity Score) to adjust for concomitant hepatotoxic medication exposure within pharmacoepidemiology studies.

Methods: We identified 193 medications with ≥ 4 reports of hepatotoxicity and created cohorts of outpatient initiators in the Veterans Health Administration (2000-2021). Exposure occurred from initiation through 30 days after discontinuation or up to 1 year. We measured age-/sex-adjusted rates of hospitalization for severe acute liver injury (ALI) by chronic liver disease (CLD), identified drugs with high rates, and used these rates as weights in the score. To demonstrate real-world use, we calculated the score for proton pump inhibitor (PPI) initiators. We summed the weights of the drugs dispensed within 90 days prior to PPI initiation. Hazard ratios (HRs) of severe ALI (95% confidence intervals) were measured with and without adjustment for Hepatotoxicity Score.

Results: Among 89 512 PPI initiators with CLD, HRs of severe ALI were higher for lansoprazole (HR = 2.17 [95% CI, 1.24-3.82]), but not pantoprazole (HR = 0.83 [95% CI, 0.61-1.13]), versus omeprazole. Adjustment for Hepatotoxicity Score attenuated HRs of lansoprazole (HR = 1.99 [95% CI, 1.13-3.50]). Among 2 462 414 PPI initiators without CLD, HRs were not significantly higher for lansoprazole (HR = 1.66 [95% CI, 0.99-2.77]) but were significantly lower for pantoprazole (HR = 0.59 [95% CI, 0.37-0.95]), versus omeprazole. Adjustment for Hepatotoxicity Score attenuated HRs of lansoprazole (HR = 1.52 [95% CI, 0.91-2.54]).

Conclusions: The Hepatotoxicity Score provides a tool to adjust for confounding due to concomitant hepatotoxic drug exposure within hepatic safety studies.