[Behavioral assessments in infancy/toddlerhood and autism spectrum disorder: Effectiveness of a semi-structured behavioral observation incorporated into a legal health examination].

Objectives　We investigated the relationship between the diagnosis of autism spectrum disorder (ASD) based on legal health examinations and semi-structured behavioral observations conducted by the Social Attention and Communication Surveillance-Japan (SACS-J), wherein public health nurses evaluated sociality, verbal communication development, and tool employment in infants and toddlers through interactive play.Methods　Public health nurses evaluated the behavioral features of 372 babies. These babies were born in 2011 or 2012 in a town, received a routine of legal health examinations at one and a half year (20 months) and at three years (38 months), and remained traceable until December 2016. The nurses used SACS-J items and the babies underwent routines of the examinations at 15, 20, 27, and 38 months. We statistically compared the relationships of the behavioral features at each of the above-mentioned months between two groups: Children with ASD and children with typical development.Results　Eight children were medically diagnosed with ASD, while five children received a diagnosis other than ASD. We compared gender, conditions at birth, physical development, and SACS-J behavioral observation items between the ASD group and the typical-development group. The ratio of the mean of the results from the ASD group was high (P<0.05), which indicates that the ASD group was significantly slower than the typical-development group in acquisition timing of "the sitting" and "the walk" (P<0.05). Significant differences in SACS-J items were "eye contact" at 15 months (P<0.05), at 20, 27, and 38 months (P<0.001) and "joint attention" at 15 months (P<.001). Further significant differences were "joint attention - adults do" at 20 months (P<0.05), "joint attention - children do" at 20 months (P<0.01), "pointing" at 20 months (P<0.05), "showing" at 27 months (P<0.001), "verbal development" at 15 months (P<0.01), "Use of language" at 20 months (P<0.01), "2-word sentence" at 27 months, and "3-word sentence" at 38 months (P<0.001). The ASD group scored significantly higher than the typical-development group in "fine motor skill" at 15 months (P<0.001) and at 27 months (P<0.01).Conclusion　Introducing an evaluation of standardized behavioral observations by public nurses in the early stages of development, prior to the legal health examination of babies at one and a half year, revealed the possibility of the early identification of children suspected of ASD at the public health activity level by a local government. Related health guidance and upbringing-and-development support are necessary in the community.