Screening for Autism in Iranian Preschoolers: Contrasting M-CHAT and a Scale Developed in Iran.

Samadi et al. (2015) compared two autism screeners for Iranian preschoolers. One was the well-known M-CHAT. The other was the Hiva, a new scale built in Iran.

They ran both tools on the same group of children and counted how many autism cases each one caught.

The Hiva scale flagged more children with autism than the M-CHAT. It gave a higher autism rate in the sample.

In short, the home-grown tool was better at finding cases in Iranian kids.

Tsai et al. (2019) in Taiwan also worked with an M-CHAT version. They reported strong accuracy: 86 % sensitivity and 96 % specificity. Their result looks opposite to Ali’s, but the kids were younger (16–30 months) and the tool was fine-tuned for Taiwan. Age and wording matter.

Pandey et al. (2008) showed the M-CHAT misses more toddlers at 18 months than at 24 months. Ali’s preschool sample is older still, so the M-CHAT may lose power again—supporting the idea that culture and age both shape performance.

Bao et al. (2017) reviewed 18 screeners used in low- and middle-income countries. They found wide swings in accuracy and called for local norms. Ali’s study answers that call by building and testing the Hiva scale.

Tsai et al. (2012) built their own 15-item Taiwanese screener and got near-perfect accuracy. Like Ali, they proved a local tool can beat an imported one when language and culture differ.

If you screen young children from immigrant or non-Western families, the standard M-CHAT may under-count. Ask parents if the items make sense in their culture. When in doubt, pair the M-CHAT with a locally validated tool or plan a second-stage assessment. Ali’s work gives you evidence to justify extra steps and reduce missed cases.