Some studies show that the diagnosis of Autism Spectrum Disorder could be considered reliable and stable in children aged 18 to 24 months. Nevertheless, the diagnostic stability of early ASD diagnosis has not yet been fully demonstrated. This observational study examines the one-year diagnostic stability of autism spectrum disorder diagnosis in a clinical sample of 147 children diagnosed between 18 and 48 months of age. The ADOS-2 scores were used in order to stratify children in three levels of symptom severity: Autism (AD; comparison score 5–7), Autism Spectrum Disorder (ASD; comparison score 3–4), and Sub-Threshold Symptoms; (STS; comparison score 1–2). Results: Overall, the largest part of children and toddlers diagnosed with autism spectrum disorder between 18 and 48 months continued to show autistic symptoms at one-year follow-up evaluation. Nevertheless, a significant percentage of children with higher ADOS severity scores exhibited a reduction of symptom severity and, therefore, moved towards a milder severity class one year later. Conversely, the number of subjects of the STS group meaningfully increased. Therefore, at one-year follow-up a statistically significant (χ²(2) = 181.46, p < 0.0001) percentage of subjects (25.2% of the total) who had received a categorical diagnosis of Autistic Disorder or Autism Spectrum Disorder in baseline no longer met the criteria for a categorical diagnosis. Furthermore, children who no longer met the criteria for autism spectrum disorder continue to show delays in one or more neurodevelopmental areas, possibly related to the emergence of other neurodevelopmental/neuropsychiatric disorders. Overall, the comprehensive results of the study account for a high sensibility but a moderate stability of ASD early diagnosis.

Background/Objectives: Minimally verbal autistic children face significant communication challenges, often unmet by traditional therapies. Social robots, like NAO, offer predictable, structured interactions that may improve engagement and language skills. This study aimed to evaluate the effectiveness of NAO-assisted therapy in improving communication and social interaction in minimally verbal autistic children compared to standard therapeutic approaches. Methods: In a single-blind, randomized, controlled study, 37 autistic children aged 4–12 years were assigned to either NAO-assisted therapy or standard speech therapy. Participants were assigned to either an NAO-assisted therapy group or a standard speech therapy control group. The intervention included 12 weekly 45 min sessions. Communication outcomes were measured using the Language Development Level Test (TVL) and mand request observations. Results: All 37 participants completed the 12 sessions without adverse events, highlighting the intervention’s feasibility and safety. Children in the NAO-assisted therapy group showed greater improvements in verbal communication (on average, 159 ± 49% more children exhibited improvement across verbal aspects (range: 107–284%; p < 0.001)) particularly in spontaneous communication, compared to the control group. The therapy also increased mand production (from 6.8 ± 4.3 in session 1 to 16.7 ± 7.7 in session 12; p < 0.001; average gain: 0.9 per session), demonstrating steady growth in communicative initiative. These findings underscore the structured and engaging nature of NAO-assisted therapy in supporting consistent progress in communication skills. Conclusions: NAO-assisted therapy is a promising, safe, and effective intervention for enhancing communication in minimally verbal autistic children, offering unique benefits in promoting spontaneous and consistent verbal engagement.

Autism spectrum disorder is a neurodevelopmental disorder with a rising prevalence disorder. This high-cost/high-burden condition needs evidence-based behavioral treatments that are able to reduce the impact of symptoms on children’s functioning. This retrospective chart review study compared the impact of different types of early interventions on toddlers diagnosed with an autism spectrum disorder developmental profile. Analyses were conducted on 90 subjects (mean = 27.76 months, range 18–44 months; M:F = 4.29:1), of which 36 children underwent the usual treatment, 13 children underwent an intervention based on early intensive behavioral intervention (EIBI) and 41 children received the Early Start Denver Model, for one year, with the same weekly frequency of about 6 h a week. A significant decrease in the severity of autism symptoms was observed for all children when looking at the Ados-2 severity score (average difference = 3.05, SD = 0.71, p = < 0.001) and the Ados-2 social subscale (average difference = 2.87, SD = 0.59, p < 0.001). Otherwise, for most of the Griffiths subscales, we found a significant improvement only for those children who underwent the Early Start Denver Model intervention (General Quotient average difference = 14.47, SD = 3.22, corrected p < 0.001). Analyzing the influence of age on the investigated scores, we found a significant association with the Eye–hand Coordination Quotient (p = 0.003), Performance Quotient (p = 0.042) and General Quotient (p = 0.006). In all these domains, a mild negative correlation with age was observed, as measured by the Pearson’s correlation coefficient (r = −0.32, p = 0.002; r = −0.21, p = 0.044; r = −0.25, p = 0.019, respectively), suggesting less severe developmental skills at the start of treatment for older children. Our results are consistent with the literature that underlines the importance of early intervention, since prompt diagnosis can reduce the severity of autism symptoms; nevertheless, in toddlers, our study demonstrated that an intervention model based on naturalistic developmental behavioral principles such as the Early Start Denver Model is more effective on children’s developmental profile. Further studies are required to assess the extent of effectiveness of different early intervention models in community settings.