Skip to Main content Skip to Navigation

Mesure de l’efficacité clinique d’une intervention précoce dans une population d’enfants avec un Trouble du Spectre de l’Autisme : validité interne et externe du résultat

Abstract : Early intervention in order to decrease the severity of Autism Spectrum Disorder has come to be self-evident . However, there is no consensus on the type of early intervention, as we have previously shown in a literature review . Some interventions have shown weak to moderate effects in randomized controlled trials. But many methodological biases call into question the validity of those results. We have chosen to implement and evaluate the Early Start Denver Model (ESDM) at our intervention center. Our hypothesis was that ESDM intervention will diminish the severity of ASD symptoms in children under 3 years old. In our pilot study, we evaluated the effect of a 12 hours per week ESDM intervention, during one year, in a population of 19 children with ASD. The primary measure was the global development, measured by the Mullen Scales of Early Learning. Our results show a significant improvement in global development pre/post intervention. Effect size in receptive language was the most important with a Cohen’s d of 0.72. These results are comparable to those of previous published studies. This trial enabled also to data collection on the feasibility of implementing this intervention in our context. Most parents and professionals were satisfied of the intervention. Treatment adherence was of 8.3±1.2 hours per week instead of 12 hours per week during the pilot study. This reduction of intervention dose was largely due to absences (holidays, training) of the professionals delivering the intervention. We discuss here the choice of criteria for the measure of effectiveness of the intervention. This choice was difficult due to the lack of tools measuring precisely the change in ASD symptomatology. To test our hypothesis, a two-parallel, multi-center randomized controlled trial was set up. The experimental group received ESDM intervention, delivered by a multidisciplinary team, 12 hours per week, during the 2 years. The control group received treatment as usual. The protocol has been published. The trial was design to have high power and high-level evidences as defined by the Cochrane recommendations. In order to achieve this, we estimated the simple size for our population from a 15 points difference between the groups on the Mullen scale, an alpha risk of 0.5 and a beta risk at 90%. As we were limited by the number of availabilities for ESDM intervention, we used 1:2 randomization. The inclusion phase ended in March 2019. 61 children were included to the ESDM group and 119 children to the control group. We discuss the internal and external validity, as well as the clinical relevance according to the possible outcomes. If outcomes show effectiveness of the intervention, a cost-consequence study will allow us to measure the efficiency of this treatment. A follow-up study has also started and will measure the effectiveness of ESDM intervention on ASD symptomatology and academic achievements five year after initial inclusion. It is evidently essential to measure the long-term effectiveness and efficiency of early interventions; however, intervention research must also collect other measures and observations. Thus, determine mediators and moderators of the intervention will make it possible to move towards personalized interventions, aiming specifically the physiological processes involved in ASD. Better understanding of these factors, and of all potential barriers or facilitators to the transfer in other contexts, will allow us to develop transferable and sustainable interventions. We will conclude on the importance of pilot studies to estimate the different factors involved. This may allow for modification of the initial intervention if necessary, prior to conducting costly effectiveness or efficiency trial. We also stress the importance of involving clinicians in research. Their practical experience of evaluation and intervention allow the formulation of hypotheses and bring complementary data to that of researchers
Document type :
Complete list of metadatas

Cited literature [254 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Friday, February 7, 2020 - 2:34:10 PM
Last modification on : Saturday, February 8, 2020 - 1:37:00 AM


Version validated by the jury (STAR)


  • HAL Id : tel-02470695, version 1



Marie-Maude Geoffray. Mesure de l’efficacité clinique d’une intervention précoce dans une population d’enfants avec un Trouble du Spectre de l’Autisme : validité interne et externe du résultat. Médecine humaine et pathologie. Université de Lyon, 2019. Français. ⟨NNT : 2019LYSE1134⟩. ⟨tel-02470695⟩



Record views


Files downloads