L'usage secondaire des données médico-administratives afin d’optimiser l’usage des médicaments chez les patients atteints de maladies respiratoires chroniques : adhésion aux médicaments, identification de cas et intensification du traitement

Medication adherence in patients with asthma and chronic obstructive pulmonary disease (COPD) is notoriously low and is associated with suboptimal therapeutic outcomes. To intervene effectively, family physicians need to assess medication adherence efficiently and accurately. Otherwise, failure to detect nonadherence may further reduce patient disease control and result in unnecessary treatment escalation that can increase the risk of adverse events and lead to more complex and costly drug regimens. The overarching goal of this thesis was to investigate how the use of secondary healthcare data can be leveraged to optimize medication adherence in clinical practice. Methodological considerations to facilitate our understanding of treatment escalation in asthma using secondary healthcare data were also examined. In the first part of my doctoral research program, I led a project which aimed at developing e-MEDRESP, a novel web-based tool built from pharmacy claims data that provides to family physicians with objective and easily interpretable information on patient adherence to asthma/COPD medications. This tool was developed in collaboration with family physicians and patients using a framework inspired by user-centered design principles. As part of a feasibility study, e-MEDRESP was subsequently implemented in electronic medical records across several family medicine clinics in Quebec (346 patients, 19 physicians). Findings showed that its integration within physician workflow was feasible. Physicians reported that the tool helped to: 1) better evaluate their patients’ medication adherence; and 2) adjust prescribed therapies, with mean ± sd ratings (5-point Likert scale) of 4.8±0.7 and 4.3±0.9, respectively. A pre-post analysis did not reveal improvement in adherence among patients whose physician consulted e-MEDRESP during a medical visit. However, significant improvements in adherence for inhaled corticosteroids (Proportion of days covered (PDC): 26.4% (95% CI: 14.3-39.3%)) and long-acting muscarinic agents (PDC: 26.4% (95% CI: 12.4-40.2%)) were observed among patients whose adherence level was less than 80% in the 6-month period prior to the medical visit. The second part of this research program consisted of two studies which laid the groundwork to estimate the association between medication adherence and treatment escalation in asthma using Canadian healthcare administrative data, a phenomenon that is currently under-explored in the literature. Prior to embarking in this study, it is important to ensure that healthcare administrative databases can be used to identify asthma patients and treatment escalations in an adequate manner. First, a systematic review was conducted to obtain an overview of the available evidence supporting the validity of algorithms to identify asthma patients in healthcare administrative databases. The algorithm developed by Gershon et al. (Canadian Respiratory Journal, 2009;16(6):183-188) comprising ≥2 ambulatory medical visits or ≥1 hospitalization for asthma over two years had the best trade-off between sensitivity (84 %) and specificity (77%). Second, an operational definition of treatment escalation was developed through a Delphi study that incorporated an expert consensus process. This definition includes 7 steps and was inspired by the 2020 Global for Initiative for Asthma treatment guidelines. I plan to integrate the definitions obtained from these two studies in a future cohort study which aims to examine the association between medication adherence and treatment escalation in asthma. My research provides compelling evidence on the importance of developing and evaluating the feasibility of implementing tools which can aid physicians in assessing medication adherence in clinical practice and extends the literature on treatment escalation in asthma.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.