This review contributes to the growing body of global autism spectrum disorder literature by examining the use of screening instruments in low- and middle-income countries with respect to study design and methodology, instrument adaptation and performance, and collaboration with community stakeholders in research. A systematic review was conducted to understand the use of autism spectrum disorder screening instruments in low- and middle-income countries from studies published between 1992 and 2015. This review found that 18 different autism spectrum disorder screeners have been used in low- and middle-income settings with wide ranges of sensitivities and specificities. The significant variation in study design, screening methodology, and population characteristics limits the ability of this review to make robust recommendations about optimal screening tool selection. Clinical-based screening for autism spectrum disorder was the most widely reported method. However, community-based screening was shown to be an effective method for identifying autism spectrum disorder in communities with limited clinical resources. Only a few studies included in this review reported cultural adaptation of screening tools and collaboration with local stakeholders. Establishing guidelines for the reporting of cultural adaptation and community collaboration procedures as well as screening instrument psychometrics and screening methodology will enable the field to develop best practices for autism spectrum disorder screening in low-resource settings.

This study examined the utility of the Chinese version Modified Checklist for Autism in Toddlers (C-M-CHAT) in a high-risk sample. This study recruited 236 children aged from 18 to 47 months old in rural area of Southern Taiwan, including 113 children with Autism Spectrum Disorder (ASD) and 123 children with Developmental Delay (DD). Using failing any four of the 23 C-M-CHAT items as cutoff, it showed a sensitivity of 77.0% and a specificity of 72.4%. The positive predictive value (PPV) was 71.9% while the negative predictive value (NPV) was 77.4%. Using failing any three of the 14 new critical items--referred to as the ''Brief 14''--as cutoff, it yielded a sensitivity of 70.8% and a specificity of 82.1%. The PPV of the ''Brief 14'' was 78.4% while the NPV was 75.4%. The preliminary results of the C-M-CHAT and the ''Brief 14'' performance demonstrated an acceptable predictive validity and promising utility for use in high-risk, rural populations in Taiwan.

The findings indicated that one of the critical goals in early intervention for young children with ASD and their families is to reduce children's behavioural problems.

No abstract

This study examined the effect of age at completion of an autism screening test on item failure rates contrasting older (>20 months) with younger (<20 months) toddlers in a community primary care sample of 73,564 children. Items related to social development were categorized into one of three age sets per criteria from Inada, Kamio & Koyama (2010). Younger toddlers produced higher rates of item failure than older toddlers and items in both of the later acquired item sets had higher probability rates for failure than the earliest acquired item set (prior to 8 months). Use of the same items and the same scoring throughout the target age range for autism screening may not be the best strategy for identifying the youngest toddlers at risk for autism.