Primary care physicians report a lack of self-perceived competency, a desire for education, and a need for improvement in primary care for children with autism. Physician education is needed to improve primary care for children with autism. Practice parameters and models of care should address physician-reported barriers to care.

Previous studies suggest over half of children with autism are using complementary alternative medicine (CAM). In this study, physicians responded (n = 539, 19% response rate) to a survey regarding CAM use in children with autism. Physicians encouraged multi-vitamins (49%), essential fatty acids (25%), melatonin (25%) and probiotics (19%) and discouraged withholding immunizations (76%), chelation (61%), anti-infectives (57%), delaying immunizations (55%) and secretin (43%). Physicians encouraging CAM were more likely to desire CAM training, inquire about CAM use, be female, be younger, and report greater autism visits, autism education and CAM knowledge. Physicians were more likely to desire CAM training, inquire about CAM and view CAM as a challenge for children with autism compared to children with other neurodevelopmental and chronic/complex conditions.

We propose that, in addition to a primary care medical home, CSHCN-HI benefit from a unique medical home center that can provide sufficient resources and expertise to organize their complex care coordination needs. Medical home centers, designed specifically to manage the care of children with complex high intensity medical and care coordination needs, have the potential to reduce excess health care utilization and improve patient outcomes by providing this group of children with customized, accessible and integrated services.

Forty-six subjects received primary medical care within an autism-specific medical home intervention ( www.autismmedicalhome.com ) and 157 controls received standard primary medical care. Subjects and controls had autism spectrum disorder diagnoses. Thirty-four subjects (74%) and 62 controls (40%) completed pre and post surveys. Controlling for pre-survey medical home status, subjects had 250% greater odds of receipt of a medical home at the study end compared to controls (p = 0.021). Compared to controls, subjects receiving the intervention reported significantly more satisfaction (p = 0.0004), greater shared decision making (p = 0.0005) and fewer unmet needs (p = 0.067). However, subjects reported no change in family stress (p = 0.204).