Doctor: Alexandre MALMARTEL
Title: Personalizing non-pharmacological interventions in randomized controlled trials
Supervisors: Philippe RAVAUD and Viet-Thi TRAN
Doctoral school: Ecole doctorale Pierre Louis de Santé Publique : Epidémiologie et Sciences de l’Information Biomédicale (ED 393)
Date of thesis defense: Novembre 17th, 2023
Jury:
Christelle NGUYEN, PU-PH, Université Paris Cité,
Clarisse DIBAO-DINA, PU-PH, Université de Tours,
Alexandra ROUQUETTE, PU-PH, Université Paris Saclay,
Viet-Thi TRAN, MCU-PH, Université Paris Cité,
Thesis topic:
Personalized medicine is a medical model in which decisions, practices and interventions are tailored to each patient’s individual characteristics in order to optimize the treatment (timing, dose, nature…) and thus improve outcomes. Non-pharmacological interventions (e.g. physiotherapy, psychotherapy, behavioral interventions…) need to be personalized to take account of this individual variability. However, the development and the implementation of personalization in non-pharmacological interventions are poorly described or standardized.
As a first step, we conducted a systematic review of personalized non-pharmacological intervention protocols. This confirmed that the description of personalization was insufficient for the interventions to be reproducible and enabled us to develop a classification of the different personalization methods used in clinical trials.
Secondly, we focused on the development of personalized non-pharmacological interventions and on the identification of the variables on which personalization is based. Using the example of smoking cessation, we developed a 3-step method for identifying the personomic markers (e.g. psychosocial situation, personal preferences, beliefs, etc.) to be used to personalize a non-pharmacological intervention: 1/ identification of variables in the literature, 2/ prioritization of variables by doctors, then 3/ by patients.
Our research will enable to standardize the selection of the tailoring variables using a valid and reproductible method, and to describe personalization transparently and accurately in clinical trials, thereby improving their replicability.
