Validity aspects of the 23-item Modified Checklist for Autism in Toddlers (M-CHAT) in a clinical population of Latvian children.

Objectives. The aim of the study was to investigate the validity aspects of the M-CHAT Latvian version. M-CHAT is a 23-item screening tool for evaluating risk for ASD.
Materials and Methods. A retrospective study was conducted. Patients who visited the outpatient clinic for Pediatric Psychiatry in Children’s clinical university hospital between October 2019 and January 2021 with filled out M-CHAT scale were included. Information about age, sex, M-CHAT, ADOS results were collected.. ADOS was considered as reference. Data were analysed with IBM SPSS Statistics.
Results. 120 patients were included. 67.5% (N = 81) were boys, 31.5% (N = 39) – girls. Mean age was 34.4 ± 7.8 months. The youngest patient was 14 months old, oldest – 47 months old. Only 30.8% (N = 37) of the toddlers were in the recommended age range for M-CHAT (under 30 months). The mean M-CHAT score was 7.7 (SD = 2.8) in boys and 8.1 (SD = 2.5) in girls, difference was not statistically significant(p = 0.64). Only in 1 of the patients the M-CHAT score fell in the low-risk range, 35%(N = 42) scored in the moderate risk range and 64.2% (N = 77) in the high-risk range. Confirmatory ADOS results were available for 31.7% of patients (N = 38). ADOS was positive in 80% of patients in the high-risk M-CHAT group (20 of the 25 patients). The sensitivity of M-CHAT was 74%, and specificity 55%, but the results lacked statistical significance (p = 0.135), possibly due to the low number of patients analyzed. The internal consistency of M-CHAT (as measured by the Cronbach’s Alpha) was moderate 0.628.
Conclusions. The level of ASD risk as measured by the M-CHAT is very high in clinical help -seeking population. M-CHAT demonstrated acceptable psychometric properties, but more research is needed in larger patient group to establish the clinical utility of the screening instrument in highly psychopathologically saturated clinical populations.