BackgroundIntegrative care practices (ICPs) in the management of autism spectrum disorders (ASDs) do not currently receive sufficient rigorous evaluation. Moreover, they lack formalisation despite their important place in clinical practice, particularly in France, where they are implemented in day care hospital structures. We present here the methodology of the EPIGRAM study financed by the French General Directorate of Healthcare Organisations (PREPS 2013), the profile of the inclusive population (M0), and the experience of families after 12 months of care. MethodIn order to homogenise practices, a manual has been formalised following the methodology of the clinical pathway and the recommendations of the High Authority of health. We then conducted an open, multi-centre, observational study in a natural environment in routine care evaluating ICPs in ASDs for children aged 3 to 6 years. The main inclusion criteria were the diagnosis (typical or atypical autism, according to the ICD10) and the duration of care (2 to 4 half-days per week) during the first year of care. The children were clinically assessed using several instruments including the PEP-3 (Psychoeducational Profile-3) and the CARS (Children Autism Rating Scale). In order to maintain a blind assessment for the main variable, randomly assigned videos of the PEP-3 were rated by two independent psychologists. A questionnaire, completed separately by the families and the professionals at inclusion and at one year, complemented the study. ResultsWe included 89 children demonstrated to have severe autism through the CARS (mean=44; SD=6.51) and moderate autism through he PEP-3 maladaptive behaviour category (CAT MB) (mean=30; SD=29.89). The children were all diagnosed before the age of 3 years 6 months. Most of the children had severe language impairments. With PEP-3, the impairment is more pronounced in gross motor skills and characteristic verbal behaviours. Thirty percent of the families had a low socioeconomic status and 56 % were first generation immigrants. Seventy-seven percent of the children were enrolled in kindergarten with school life support. Children of immigrant parents have more severe autism (CARS : mean=45.8 (±6.51) vs. 41.2 (±5.58), t-test, P=0.001) in addition to a more marked cognitive deficit (CVP : median=24.5 [min–max: 9.75–45.5] vs. 52 [min–max: 26–76], Wilcoxon, P=0.001). At entry, the collection of the signs at the structured clinical observation met the signs noted by the families as being the object of their concern. With standardised tools, these observations constituted an efficient functional and multidimensional analysis for the integrative care project. At 12 months, the families’ level of satisfaction was high and consistent with that of the practitioners. ConclusionOur results show that the day units take care of a population of children with predominantly severe and complex ASDs, for the most part, from families with low socioeconomic status and first-generation immigration. The relevance of the multidimensional observation and the care project of the ISPs were confirmed by the families at entry and after 12 months of care.