Methodological Choices for the Clinical Development of Medical Devices

This document focuses on aspects of the clinical assessment of efficacy for a new medical device or a new health technology that is made after the pre-clinical phase and feasibility studies.

Its objectives are:

• to provide an up-to-date overview of methods that can be used to assess the clinical benefit of a new medical device or health technology;
• to describe possible study designs for carrying out a high-quality clinical assessment.

Since a clinical assessment has the aim of comparing the efficacy of two treatments, a well‑conducted double-blind randomised controlled trial is the type of study that offers the highest level of evidence, allowing the efficacy of these treatments to be assessed and medical decisions to be supported. Randomisation, blinding and intention-to-treat analysis (for studies aiming to demonstrate the superiority of a treatment) are the main criteria with regard to methodological quality. In most cases, a randomised controlled trial can be performed.

However, some situations are incompatible with conventional randomised controlled trials, particularly when randomisation and/or blinding are impossible.

The National Committee for the Evaluation of Medical Devices and Health Technologies (CNEDiMTS) wished to ascertain the situations in which conventional randomised controlled trials are not possible and to identify the methods and conditions that allow a high-quality comparative clinical assessment to be performed. This document focuses on other types of randomised trials and observational studies. All these study designs have their own limitations and should be reserved for clearly defined situations. There must be a scientific justification for choosing them, since randomised controlled trials remain the standard for any clinical study that compares the efficacy of multiple treatments or treatment strategies.