A growing number of autism studies are looking for early risk factors by studying baby sibs, infants who have a sibling with autism and hence a 20-fold increased risk of developing the disorder compared with the general public.

Scientists are also studying baby sibs to assess the potential benefits of early intervention, before the infants even have a diagnosis.

In a pilot project called iBASIS, part of the British Autism Study of Infant Siblings (BASIS), scientists are analyzing whether training parents to respond more attentively and directly to their children helps infants at high risk for autism minimize their symptoms.

With seven baby sibs, the study is too small to determine whether the intervention lowers autism risk, but the researchers found that parent-child interactions improved over the course of the five-month program, according to results published 27 March in the Journal of Autism and Developmental Disorders.

Therapists spent 12 sessions training mothers to better read and respond to their preverbal 8- to 10-month-old children, making audio and visual recordings during each session. They taught the mothers how to respond to infant emotions, facial expressions and gestures, and how to promote joint attention.

The therapists reviewed the recording with the mothers at the following session, and taught them how to improve their interactions. They encouraged the mothers to practice their new skills for 30 minutes every day.

The researchers compared this group with 37 other baby sibs from BASIS who did not receive the therapy, and 33 normal-risk infants.

All seven mothers of high-risk baby sibs who received the training said the program helped increase their awareness of their baby’s emotions.

One mother realized that her baby’s grunting conveyed emotion, a skill she didn’t know he had. “It makes you look really closely at what your interaction with your child is, and there is much more in it than you thought,” said another mother.

The new study follows previous research from the same group showing that 1-year-old baby sibs whose parents foster social engagement during interventions show fewer autism diagnoses at age 3 than baby sibs who receive less interactive parenting. This suggests that teaching caregivers how to recognize and encourage social behaviors may help at-risk children develop more typically, the researchers say.

The researchers are testing the intervention’s effects on a larger group of 54 baby sibs, also part of BASIS.

For most children with autism, interventions don’t begin until at least 3 years of age, when they can be reliably diagnosed. If successful, iBASIS should provide impetus for delivering therapy much earlier.