| Objectives:To evaluate the diagnostic accuracy of the Chinese version of the Autism Mental Status Exam(AMSE)for diagnosing autism spectrum disorder(ASD)in child and adolescent Chinese sample by receiver operating characteristic curve(ROC curve)and determinate the best cutoff score of ASE,as a reference of early identification of ASD.Methods:The participants were recruited from the Child and Adolescent Mental Health Outpatient Clinic,Department of Psychiatry,the 1st Affiliated Hospital of Kunming Medical University from July 2016 to January 2017.Subjects who aged from 2 to 18 years old exhibit:ing social interaction and/or communication problems with a suspected ASD diagnosis were enrolled to the present study.Firstly,a board certified Child and Adolescent Psychiatrist with extensive experience(a best estimate clinical diagnosis clinician,BECD clinician)recruited participants according to the inclusion criteria and recommended them to participate in the study.Secondly,psychiatrist A reviewed the Childhood Autism Rating Scale(CARS),the Clancy Autism Behavior Scale(CABS)and the developmental history,which was documented using a standardized intake form listing chief complaint,history of present illness,past medical history,and developmental milestones.Thirdly,another psychiatrist(Psychiatrist B)measured in AMSE score for each subject independently.At last,according to DSM-5 criteria,the BECD clinician who was blind to the AMSE scores reviewed all clinical evaluation administered by psychiatrist A and established the diagnosis as ASD or non-ASD,which was viewed as "gold standard" in the present study.ROC curve of AMSE were used to search for the most effective cutoff score.Sensitivity,specificity,misdiagnosis rate,missed diagnosis rate,Youden index,positive predictive value,negative predictive value,positive likelihood ratio and negative likelihood ratio of the Chinese version of AMSE were also calculated at the most effective cutoff score.The nonparametric measure of area under the ROC curves(AUC)was applied to assess the diagnostic accuracy of AMSE.Results:1.A total of 69 participants was included in this study with age ranged from 2 to 18 years(Mean age:5.32),comprising 56(81.2%)males and 13(18.8%)females.The study sample included 54 children with ASD(46 males and 8 females)and 15 with non-ASD(10 males and 5 females).The sample of non-ASD is comprised of mental retardation(MR),language development disorders and unspecified schizophrenia spectrum disorder.2.The best cutoff score of the Chinese version of AMSE was estimated as 6.The cutoff score of 6 produced a sensitivity of 98.1%and a specificity of 86.7%.The misdiagnosis rate was 13.3%and the missed diagnosis rate was 1.9%.The Youden index was 0.848.The positive predictive value and negative predictive value were 96.4%and 92.9%,and the positive likelihood ratio and negative likelihood ratio were 7.361 and 0.021.The value of AUC was 0.979[95%confidence interval(CI)0.911～0.999]and represents very good diagnostic accuracy.3.The Kappa value was 0.869(P<0.05),showing that the strength of diagnostic agreement by "gold standard" and the Chinese version of AMSE was great.4.The inter-rater reliability of the Chinese version of AMSE was assessed with 13 participants.Two clinicians established a 80.6%inter-rater reliability(P<0.05).Conclusions:Our data suggest that the diagnostic accuracy of the Chinese version of AMSE is satisfied in a sample of Chinese children and adolescents.As a brief and free structured tool,the Chinese version of AMSE successfully differentiates ASD patients from non-ASD patients.Therefore,Chinese version of AMSE can support diagnostic decision making and in turn improves the early diagnostic and therapeutic procedures in clinical practice. |
PDF Full Text Request