SPHERE Project




The methodS in Patient-centered outcomes & HEalth ResEarch (SPHERE – UMR 1246) research unit is a two-site inter-regional research unit jointly accredited by the University of Nantes, the University of Tours and the INSERM.


We aim at developing, validating, and applying methods in a pluridisciplinary perspective to address a variety of challenges encountered in health research and decision making, taking into account both individuals’ environments and perceptions.

The SPHERE research program is currently developed in the framework of the “personome” concept introduced by Ziegelstein in 2015 (Ziegelstein RC. Personomics. JAMA Intern Med. 2015 Jun;175(6):888-9) which advocates that individuals should be considered both in terms of their social and family environment and their own personality, health beliefs, and experience of illness.

Our scientific objectives

Our scientific objectives aim to strengthen the development of methods for patient-centered health research in a multidisciplinary perspective involving biostatistics, public health, clinical disciplines (addictology, dermatology, general medicine, nephrology), pharmacology and human and social science disciplines (health psychology, health economics and sociology).

Illustration of the project

Our research

Our project is based on 4 axes which are part of a methodological continuum to address
several challenges related to the development of methods for :

Icone de l'axes

Axis 1

Cluster Randomized Trials and Complex Interventions

The design, conduct and analysis of pragmatic cluster randomized trials aimed at assessing
complex interventions.

Icone de l'axes

Axis 2

Definition, selection, validation and assessment of outcomes

The choice and relevance of outcomes and the case for patient and public involvement.

Icone de l'axes

Axis 3

Methods for the measurement and interpretation of Self-Reported Outcomes

The measurement and interpretation of self-reported outcomes evaluating subjective concepts.

Icone de l'axes

Axis 4

Prediction and Causality

The link between prediction and causality in a non-randomized context, notably using machine learning techniques and network psychometrics.