P2Y12 inhibitor (P2Y12i) nonadherence in the first year after acute coronary syndrome (ACS) is a major obstacle to optimal clinical outcomes. Differences in definitions and methods for measuring P2Y12i adherence complicate identifying suboptimal therapy after ACS, leading to disparate estimates across studies. We assessed P2Y12i adherence using a range of adherence/persistence definitions and methods in the same population to better understand how such variations affect adherence estimates.
We conducted a population-based retrospective cohort study using administrative claims data. These datasets were linked using unique encoded identifiers and analyzed at ICES. We included patients aged ≥65 years who filled a prescription for P2Y12i (clopidogrel or ticagrelor) ≤7 days after ACS discharge between April 2014 and March 2018. We excluded patients who may have exhibited different adherence patterns (coronary artery bypass grafting during index hospitalization, major bleeding, prescribed oral anticoagulants) and, to avoid misclassifying them as nonadherent, those who died within 1 year after index ACS discharge. We estimated 1-year P2Y12i adherence using 2 proportion of days covered (PDC)-based definitions: mean PDC and proportion with “good” adherence (PDC ≥80%), and persistence with 4 different permissible gaps between prescriptions of 3, 7, 14 (base case), and 30 days. The use of data for this study was authorized under section 45 of Ontario’s Personal Health Information Protection Act, which does not require review by a research ethics board.
The study cohort comprised 21,680 patients (mean age [± SD] 75.64 ± 7.89 years, 60.3% men), of whom 11,917 (55.0%) used clopidogrel and 9,763 (45.0%) used ticagrelor. Most experienced myocardial infarction (85.7%) and underwent percutaneous coronary intervention (68.8%) during the index ACS hospitalization. Common co-morbidities included hypertension (76.9%), diabetes (36.4%), and dyslipidemia (46.4%). Medication use within 90 days before to 7 days after the index episode included statins (93.6%), β blockers (77.9%), and angiotensin-converting enzyme inhibitors/angiotensin II receptor antagonists (84.4%). The mean 1-year PDC was 88.0% (SD: 23.5%), and 82.8% of patients had “good” adherence (≥80% PDC). However, only 60.7% of patients continuously persisted in the 1 year after ACS, using a 14-day allowable gap. Persistence ranged from a high of 67.2% using a 30-day allowable gap to a low of 45.0% using only a 3-day allowable gap ( Figure 1 ).
Our findings illustrate that P2Y12i adherence and persistence estimates in the first year after ACS are greatly sensitive to the definitions used. Mean PDC provided the most optimistic estimates, followed by the proportion with ≥80% PDC (“good” adherence), with persistence estimates being the most conservative. Although the PDC-based methods imply reasonable 1-year P2Y12i adherence, they may give a false sense of security about the true levels of regular medication use. PDC methods do not consider continuous medication use, which is essential for P2Y12i use after ACS to prevent thrombosis and can oversimplify complex adherence behaviors and overlook the timeliness of refills. Although persistence methods evaluate continuous medication use, the permissible gaps must be chosen carefully, with consideration of the medication’s pharmacologic properties because excessively small gaps may be too stringent and not clinically relevant.
Choice of the permissible gap is crucial and should be informed by drug pharmacokinetics and pharmacodynamics, and by study goals. Given P2Y12i’s antiplatelet effect largely dissipates by day 7 without therapy, and continuous protection from thrombotic events in the first year after ACS is crucial, even small medication gaps can impair the clinical efficacy of P2Y12i. Our study revealed that using persistence methods, almost half of patients had a P2Y12i therapy gap >7 days within the year after ACS, a sufficiently long gap to potentially lead to adverse clinical outcomes. It is important to note that typical adherence methods, such as PDC and medication possession ratio, do not detect these important gaps in therapy. Readers of P2Y12i adherence and persistence studies should pay close attention to the methods and definitions used before they presume good adherence on the basis of study results because adherence and persistence estimates can vary widely, and single estimates may not give a full picture of patients’ medication-taking behaviors.
Comparing adherence results across studies is challenging given disparate methods and populations. Few studies have directly compared the estimates of various adherence and persistence measures in the same population, and none have done so for P2Y12i. One study comparing statin adherence and persistence using multiple definitions also found estimates were greatly variable based on the definition/metric used, with the small-gap persistence method being the most conservative. Although these results may seem intuitive, estimating the magnitude of differences using varying definitions in the same population as we have done provides important insight to clinicians, policymakers, and researchers to better understand the reported P2Y12i adherence rates that have differed in other studies. , , , Various measures for assessing medication adherence have been studied, yet there is no single gold standard. In some instances, examination of continuous use using persistence measures may be most clinically relevant.
Despite using various definitions and measures to accurately assess P2Y12i adherence, we observed suboptimal adherence to P2Y12i across measurement methods. This emphasizes the need to seek solutions to enhance medication adherence. Strategies to improve adherence include conducting adherence checks during clinic visits, using pill boxes, providing personalized reminders, minimizing cost burden, and implementing structured pharmacy care programs. Regular education for both clinicians and patients about the significance of P2Y12i therapy after ACS or percutaneous coronary intervention is also crucial in improving adherence and minimizing adverse outcomes. In addition, prospectively identifying patients at risk of nonadherence can enable targeted interventions for sustained adherence, as investigated in previous work.
Although our study offers valuable insights, it also comes with inherent limitations. Our analysis did not extend to the evaluation of the association of adherence and persistence with clinical outcomes, indicating a direction for future research. Moreover, our retrospective cohort study, relying on the Ontario Drug Benefit Program database, inherently focuses on patients aged ≥65 years, thereby excluding younger patients. Furthermore, our use of prescription claims data might not accurately reflect actual medication consumption, potentially overestimating adherence rates.
In conclusion, using >1 measure of adherence and persistence in the same study is complementary and may allow a better understanding of drug exposure duration and intensity, and an improved ability to detect suboptimal medication use patterns.
CRediT authorship contribution statement
Jungyeon Moon: Writing – review & editing, Writing – original draft, Visualization, Formal analysis. Aya F. Ozaki: Writing – review & editing, Formal analysis. Alice Chong: Writing – review & editing, Methodology, Formal analysis, Data curation. Maneesh Sud: Writing – review & editing, Formal analysis. Jiming Fang: Writing – review & editing, Methodology, Formal analysis, Data curation. Peter C. Austin: Writing – review & editing, Methodology, Formal analysis, Data curation. Dennis T. Ko: Writing – review & editing, Project administration, Funding acquisition, Formal analysis. Cynthia A. Jackevicius: Writing – review & editing, Validation, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization.
Declaration of competing interest
The authors have no competing interests to declare.
Acknowledgment
The opinions, results and conclusions reported in this paper are those of the authors and are independent from the funding sources. The data sources included the Canadian Institute for Health Information (CIHI) Discharge Abstract Database, Ontario Registered Persons Database, Ontario Drug Benefit database and the Statistics Canada database, all at ICES which is an independent, nonprofit research institute whose legal status under Ontario’s health information privacy law allows it to collect and analyze health care and demographic data, without consent, for health system evaluation and improvement. This document used data adapted from the Statistics Canada Postal CodeOM Conversion File, which is based on data licensed from Canada Post Corporation, and/or data adapted from the Ontario Ministry of Health Postal Code Conversion File, which contains data copied under license from ©Canada Post Corporation and Statistics Canada. Parts of this material are based on data and/or information compiled and provided by CIHI and the Ontario MOH. The analyses, conclusions, opinions and statements expressed herein are solely those of the authors and do not reflect those of the funding or data sources; no endorsement is intended or should be inferred. We thank IVQIA Solutions Canada Inc. for use of their Drug Information File.
Author Contributions
Dr. Jackevicius and Dr. Ko were responsible for concept and design, obtaining funding, and administrative, technical, or material support. Dr. Jackevicius supervised the work. Dr. Moon drafted the manuscript. Alice Chong, Dr. Fang, and Dr. Austin were responsible for the statistical analysis and data acquisition. All authors were responsible for the interpretation of data and for critical revision of the manuscript for important intellectual content.
This study was funded by a Foundation grant (FDN-154333) from the Canadian Institutes of Health Research. Dr. Ko has received support from the Jack Tu Research Chair in Cardiovascular Outcomes Research. Dr. Sud is funded by the Eliot Phillipson Clinician-Scientist Program at the University of Toronto and by a Canadian Institute of Health Research Post-Doctoral Fellowship. The authors acknowledge that this study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC).
References
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
