February 1, 2010

Kids with suspected development problems may not get referrals

Many pediatricians score high on screening their patients for developmental delays but barely make a passing grade in referring children with suspected delays for further testing or treatment, according to a study from Johns Hopkins Children’s Center and other institutions to appear in the February issue of Pediatrics.

Because screening is effective only if followed by referral and treatment, pediatricians need two separate formalized systems in their practices—one for screening and one for referral—the investigators write in their report, which was available online Jan. 25.

The report also notes that pediatricians, and their patients, would fare better if the doctor’s office placed the referral on a patient’s behalf instead of handing the family a phone number to do so on its own.

In 2006, the American Academy of Pediatrics issued guidelines that call for informal clinical evaluation of development of all children younger than 3 years during all well-child visits, and the use of a standard developmental screening test in all children at ages 9 months, 18 months and 24 (or 30) months. The AAP recommends that children who fail a screening test be referred to a specialist for more testing and to an early intervention program.

The new study, launched as an AAP project to improve screening and referral, analyzed the procedures over nine months among 17 pediatric practices in 15 states, whose physicians and staff were trained on the new AAP guidelines.

Although screening rates improved after the training, from 68 percent to 85 percent of patients, referrals remained low, with only 61 percent of those with suspected delays sent for further testing and therapy.

Among pediatricians who referred patients, many failed to follow up with the family to check whether they acted on the referral. The investigators said that many families don’t follow up, and that some don’t understand the reason for the referral in the first place.

In addition, the investigators said, even when pediatricians did refer, their patterns were all over the map, with most failing to refer children to both a specialist and an early intervention program; rather, they chose one or the other.

“The ultimate goal of screening is to improve outcomes for children with developmental delays, but in our study we found that many pediatricians did not act properly even when serious red flags were present,” said lead researcher Tracy King, a pediatrician at the Johns Hopkins Children’s Center. “This is where we should focus our efforts: making sure that more children with suspected delays get referred for testing and therapy.”

King and colleagues said that referrals work best if they are placed by the pediatrician’s office on the patient’s behalf and tracked much like any other medical referral. Historically, referrals for developmental delays have not been treated as medical referrals but rather informally, by handing the parents a phone number and telling them to follow up with an early intervention program or a therapy provider.

Follow-up is critical because children with developmental delays who are treated promptly fare better than those who get delayed treatment or no treatment at all, King said. For example, a child with a speech problem who receives early speech-language therapy will improve faster and do better in the long run than a child who is either not identified until later or whose referrals are not acted upon.

To increase referral and follow-up, each practice should establish a separate referral log and tracking system that prompts staff, nurses and doctors not only to refer a patient but to call and check with the family periodically to ensure action.

Other findings from the study include:

While nine of the 17 practices reported conducting developmental screening prior to the training, none of the 17 were following AAP screening guidelines.

After the training, all practices successfully incorporated AAP-recommended screening into their clinic routines.

Fourteen percent of all screened children during the nine-month study failed a developmental screening, a red flag signaling possible developmental delays.

Referrals ranged from 27 percent to 100 percent among the 17 practices, with an average of 61 percent.

Unlike screening rates, referral rates did not increase over time. They dropped down even further during the second half of the nine-month study, with just over half the children who failed a screening test getting a referral.

The most common deterrent to universal use of standard screening tools was fear that the time it takes might slow down patient flow.

Co-investigators are S. Darius Tandon, of Johns Hopkins; Michelle Macias, of Medical University of South Carolina; Jill Healy and Paula Duncan, both of University of Vermont; Nancy Swigonski, Indiana University School of Medicine; and Stephanie Skipper and Paul Lipkin, both of Kennedy Krieger Institute.

The research was funded in part by the AAP, the Centers for Disease Control and Prevention and the Commonwealth Fund, among others.