Medication Review Support – ABiMed

This tool will be based on the following methods: (a) an increased collaboration between the pharmacist and the physician during the review, including the transfer of patient data available on the physician's computer, and the proposal of a phone meeting after the review, (b) a rule-based system implementing the existing recommendations for detecting the prescriptions to revise, (c) a visual documentary tool for synthetizing the properties of the 5-20 drugs taken by the patient in a single screen, and (d) a form structuring the patient interview and the expression of his preferences.

The project will be based on recent advances in computer science, in particular (a) automatic language processing to code the patient data sent by the doctor and avoid tedious data entry, (b) visual analytics to give an overview of the properties of the drugs and (c) semantic web, with an ontology of the medication review to structure all of the knowledge involved in the project.

The proposed tool will be subject to two evaluations. An in vitro study on clinical cases, carried out online by pharmacists, will compare the quality of the reviews carried out with or without the tool. An in vivo study, carried out in clinical conditions by pharmacists and doctors on volunteer patients, will measure the satisfaction of the various parties involved, the quality of the pharmacists' reviews and the effect of the polypharmacy prevention rules on doctors.

To come

This project aims to have a major impact on several levels: (a) social: the project can have a significant impact on the health of patients, especially the elderly, (b) economic: the project can reduce health costs, by deprescribing unnecessary drugs and the reduction in hospital costs related to adverse effects, (c) professional: the project can simplify the work of clinicians and strengthen the medical value of the pharmacist, (d) scientific: the project is very innovative in the field of clinical decision support, because it combines semantic web and visualization, (e) ecological: the consumption of drugs is constantly increasing; reducing it would make it more sustainable.

For the moment, 1 journal article and 4 congress papers.

Mouazer A, Tsopra R, Sedki K, Letord C, Lamy JB. Decision-support systems for managing polypharmacy in the elderly : A scoping review. Journal of Biomedical Informatics 2022 ;130 :104074 www.sciencedirect.com/science/article/pii/S1532046422000909

Polypharmacy is a major public health problem. It causes adverse events and drug-drug interactions, but it also has an important cost for health insurances. A solution is the medication review: a pharmacist interviews the patient, reviews his medications and simplify the treatment, by removing non-mandatory drugs or those responsible for adverse events. Medication review is reimbursed by health insurance in France since 2018. However, the review is difficult because the patient data are owned by the physician, and it requires complex knowledge. Few medication reviews are made in practice.
The tools for supporting medication reviews are limited. They are based on the implementation of “deprescribing” rules, on the presentation of the properties of the drugs taken by the patient in plain tables, or on questionnaires for structuring the patient's interview. However, these approaches have never been combined together, and they have not incorporated recent advances in computer science.
In this project, we propose to design and evaluate a tool to support medication reviews for pharmacists, and prevention of polypharmacy for prescribing doctors. This tool will be based on the following methods: (a) an increased collaboration between the pharmacist and the physician during the review, including the transfer of patient data available on the physician's computer, and the proposal of a phone meeting after the review, (b) a rule-based system implementing the existing recommendations for detecting the prescriptions to revise, (c) a visual documentary tool for synthetizing the properties of the 5-20 drugs taken by the patient in a single screen, and (d) a form structuring the patient interview and the expression of his preferences.
The project will be based on recent advances in computer science, in particular (a) automatic language processing to code the patient data sent by the doctor and avoid tedious data entry, (b) visual analytics to give an overview of the properties of the drugs and (c) semantic web, with an ontology of the medication review to structure all of the knowledge involved in the project.
The proposed tool will be subject to two evaluations. An in vitro study on clinical cases, carried out online by pharmacists, will compare the quality of the reviews carried out with or without the tool. An in vivo study, carried out in clinical conditions by pharmacists and doctors on volunteer patients, will measure the satisfaction of the various parties involved, the quality of the pharmacists' reviews and the effect of the polypharmacy prevention rules on doctors.
This project aims to have a major impact on several levels: (a) social: the project can have a significant impact on the health of patients, especially the elderly, (b) economic: the project can reduce health costs, by deprescribing unnecessary drugs and the reduction in hospital costs related to adverse effects, (c) professional: the project can simplify the work of clinicians and strengthen the medical value of the pharmacist, (d) scientific: the project is very innovative in the field of clinical decision support, because it combines semantic web and visualization, (e) ecological: the consumption of drugs is constantly increasing; reducing it would make it more sustainable.
The project is supported by a consortium of four complementary partners: an academic laboratory, a learned society, an association of clinicians and an enterprise that publishes medical software. The results of the project will be published in high-level journals and congresses in medical informatics, informatics, clinical pharmacy and primary care. At the end of the project, the software developed will be marketed by the partner enterprise and / or other enterprises in the sector.