Developmental screening and early intervention in a childcare setting for young children at risk for autism and other developmental delays: A feasibility trial.

Randomization was by childcare site, not by child, and one site supplied most participants, leaving 16 children in treatment and 4 in control. The analysis did not test whether the treatment group changed more than the control group, so the child outcome changes are exploratory. Only 20 of 42 eligible children enrolled, and the study did not learn why many families did not consent to intervention. Childcare providers received only 2 hours of training and achieved fidelity above 70% for environment and routine but not language. Because children also continued regular childcare, the study cannot separate JASPER from ordinary classroom experience. With such small groups, the estimates depend on model assumptions and cannot describe treatment magnitude well.

The authors reported no conflict of interest.

Do not read the increases in joint engagement, play, and language as proof that JASPER works. The study had 16 treated and 4 control children, randomized by site, and did not test treatment against control, so the changes cannot be separated from ordinary childcare or chance.