It is unknown how insurance status affects elements of evaluation at developmental behavioral (DB) pediatric sites.
To compare DB referrals, evaluation, and treatment for children with Medicaid and private insurance.
Fifty-six developmental behavioral pediatricians at 12 sites recorded anonymous data on structured forms for ≤15 consecutive referrals. Children with Medicaid (n = 309) and private insurance (n = 393) were compared on sociodemographic factors, referral concerns, evaluation elements, and resulting diagnoses. All significant bivariate findings were verified in multivariable models controlling for site and sociodemographic characteristics.
Those with Medicaid were significantly less likely to be white (30% vs 63%) and to have parents who went beyond high school (50% vs 92%) and who spoke English (89% vs 97%) (all p < 0.001). Referral sources were similar, except that fewer children with Medicaid were self-referred (12% vs 22%; p < 0.01). Both groups presented with multiple concerns, ∼3/child, especially speech and language delays, autism spectrum disorder, and attention-deficit hyperactivity disorder. Children with Medicaid tended to present more often with concerns about other behavior problems (14% vs 7%; p = 0.05). Wait times to appointments were similar (∼20 weeks), and visits were over 2 hours in length. Only 1 of 92 services and recommendations differed significantly.
There were few differences in care, but most DB patients presented with multiple concerns and had complex evaluations. Regardless of the insurance type, they experienced long wait times that may be detrimental to therapeutic outcomes.
*Pediatrics, Children's Hospital at Montefiore/Albert Einstein College of Medicine Bronx, NY;
†Pediatrics, Boston University, Boston, MA;
‡Pediatrics, Children's Hospital of Philadelphia, Philadelphia, PA;
§Pediatrics, Brown University, Providence, RI;
‖Pediatrics, Case Western Reserve University, Cleveland, OH.
Address for reprints: Ruth E. K. Stein, MD; e-mail: firstname.lastname@example.org
Disclosure: The authors declare no conflict of interest.
Received July , 2017
Accepted December , 2017