Approach
Citizens are heterogeneous with regard to age, biological sex, gender, genetic constitution, and current or chronic disease status. A “one-size-fits-all” approach to drug therapy is therefore often inadequate. Even more so when treatments involve multiple medications for one or more diseases (polymedication). Polymedication and the lack of personalised therapies raise the incidents and severity of adverse drug reactions (ADRs). In Europe alone, approximately 197,000 annual deaths can be attributed to ADRs.
However, it is extremely challenging to take individual characteristics into account and to assign personalised drug treatment based on currently available risk assessment and dosing tools. While individual risk factors of drug therapy associated with ADRs and poor treatment outcomes are known, a systematic and holistic evaluation considering polypharmacy, multimorbidity, genetic variations as well as demographic and disease-related factors is still missing. Changing the state of the art and moving toward therapies that consider all of these factors would be highly beneficial for citizens as well as health care systems.
This is where SafePolyMed comes in: In order to assist physicians in accurately assessing individuals’ risk of experiencing ADRs, SafePolyMed seeks to develop a novel, evidence-based approach.
Through real-world data analysis of over 1 million individuals from national and international health record databases, SafePolyMed aims to build a well-defined risk score system. Genome-wide genetic information, demographic data, current and chronic health conditions as well as polypharmacy of individual patients will be taken into account to explore underlying causes of poor treatment outcomes and to define more personalised treatment plans.
Moreover, SafePolyMed will develop citizen-centred tools using machine learning (ML) and artificial intelligence (AI) techniques. By educating citizens and empowering them to participate in self-documenting their therapy, the project seeks to improve the communication between patients and physicians as well as equal access to innovative, sustainable, high-quality health-care across Europe.
Real-World Data
SafePolyMed will gather, combine and analyse the electronic health records and genotype data of over 1 million individuals. Genome-wide genotype data, demographic data, underlying health conditions, and polypharmacy will all be combined in regression analyses.
Risk Prediction
The development of an evidence-based risk scoring system using machine learning (ML) on large real-world datasets enables physicians to identify patients at risk, taking into consideration polypharmacy, multimorbidity, genetic variations as well as demographic and disease-related factors. More personalised and effective decisions will be thereby made possible.
Patient Reported Outcome Measures
Since most adverse drug effects occur during the time a patient is not within the structures of the health care system, a citizen-science approach to empower patients to better understand, recognise and report individual safety outcomes is needed. SafePolyMed aims to develop and validate patient-reported outcome measures (PROMs) in collaboration with patient organisations and other stakeholders using the Delphi consensus process. The developed PROMs will be implemented in the medication management center (MMC) and will be further validated in clinical case studies.
Medication Management Center (MMC)
To ensure interoperability within Europe’s national health systems, SafePolyMed will integrate several advanced modules on patient safety and empowerment in a sustainable, comprehensive knowledge hub - a medication management center (MMC). In doing so, SafePolyMed will provide standardised interfaces across regions and countries in Europe.
Model-based Precision Dosing
Safety and efficacy of drug therapy can be affected by multiple factors, and even the administration of standard drugs at therapeutic doses can lead to dose-related adverse drugs reactions (ADRs). SafePolyMed will apply rigorously validated mathematical models to real-world data in order to make individualized treatment based on patient-related factors possible.
Clinical Case Study
Real-world usage of the medication management tool developed in the previous phase will be examined in a clinical case study, involving four sites in four European countries and serving as a “proof-of-principle”. The primary objective is to assess the feasibility of using the developed tool to empower EU citizens to improve drug safety. Secondly, the study aims to verify whether the tool is able to accurately identify patients at risk of experiencing adverse drugs reactions.