"It's a problem, because I think that early intervention services can really make a difference in kids who are at risk for developmental delay," said Dr. Joanne Cox, a pediatrics professor at Children's Hospital Boston, who was not involved in the research.

Early intervention services try to get kids who lag behind in physical or mental skills to catch up with their peers.

Doctors, other health providers and parents can refer a child for early intervention, and states set the standards for which children are eligible for government-coordinated services.

Dr. Brian Tang, lead author of the study and a clinical instructor at Stanford University School of Medicine, said there's been a big push in the public health and medical communities to screen for developmental delays and get kids into early intervention programs.

Among children who spent time as a newborn in the neonatal intensive care unit (NICU) of a hospital, the risk of having delays is higher, and "I thought it was important to see how good of a job we're doing in terms of getting (them) referrals," Tang said.

He and his colleagues collected data from a quality improvement program in California that has tracked the medical care and health outcomes of children who have been hospitalized in a NICU.

All of the more than 5,000 children in the study were eligible for three follow-up visits with a specialist to check up on their developmental progress after they left the NICU. The visits occurred before the child turned three.

The researchers measured how often kids at high risk of delays got a referral for early intervention during one of these follow-up visits.

They considered the children high risk if they failed a screen for developmental progress or scored near the bottom of the pack on a standardized test.

Of the 185 children who failed the screening test, some had already been in early intervention and most of those who weren't received a referral for it during their first follow-up visit.

But 42 of the 118 who had no early intervention before failing the screen -- 36 percent -- did not get a referral to early intervention or other services at the first follow-up visit when they were four to eight months old.

Similarly, 136 of the 401 kids (34 percent) who scored low on the standard test and had no previous early intervention did not get any referrals.

Tang said he suspected that doctors might be taking a wait-and-see approach before referring the children to early intervention.

But at the second follow-up visit, when the kids were 12 to 16 months old, the numbers were about the same -- 34 to 37 percent of the children without previous early intervention did not get a referral for it at that point either.

"I think it was really shocking to see that... a significant proportion of them weren't getting referrals," Tang told Reuters Health.

Based on their test scores, all of them would have been good candidates for early intervention, he said.

It's not clear why these children were not recommended for early intervention.

"I would question whether there are problems with access, like are there not enough (early intervention providers) in the area. Is that a reason why they didn't refer them?" Cox told Reuters Health.

Tang said it's also possible that some parents might not want their child referred.

The researchers write in their report in the journal Pediatrics that major budget cuts to early intervention services in California, and more strict criteria limiting which kids are eligible, might have influenced how readily doctors referred the children.

"I think the evidence is pretty strong that early intervention does help children who are delayed, but I understand there is conflict with limited resources," Tang said.

He said he is now exploring what barriers are stopping children at high risk for developmental delays from getting referred.

SOURCE: bit.ly/Kbuh6J Pediatrics, online May 21, 2012.

• A Permanent Economic Emergency