The primary health care provider plays a critical though often underappreciated role in the identification and ongoing care coordination of children and youth with autism spectrum disorders (ASDs). Consistent and accessible care is important for initial identification of children at risk for ASD, but also for ongoing management and identification of the medical and nonmedical needs of the child and family. Integration of care across medical, educational and social service systems is challenging to both families and health care providers due to the time required, cost of staff time and services, knowledge, training, availability of services, and lack of support for the collaborative effort itself (Swiezy et al. 2008). Effective communication and collaboration between families and health care providers helps to facilitate the implementation of interventions. The Maternal Child Health Bureau and the American Academy of Pediatrics promote the Medical Home model of primary care practice. The Medical Home provides care that is accessible, continuous, comprehensive, family centered, compassionate, culturally effective, and coordinated with special services for children with special health care needs (CSHCN) (Medical Home Initiative 2002). This review addresses the enhanced responsibilities that children and youth with ASD present to the primary health care provider and the real barriers that need to be addressed by research and public policy to effect the recommendations for care that both parents and providers see as important for implementation of a true Medical Home.

Care of Children with Autism in the Medical Home

A familiar and consistent source of health care is important for all children, though is especially important for CSHCN. Research to date indicates that CSHCN benefit from components of Medical Home care such as family involvement and timeliness of care with improved health status in terms of fewer missed school days and fewer unmet needs with provision of care coordination (Homer et al. 2008). Whether care through a Medical Home leads to better overall health outcome is not yet known. The National Survey of Children’s Health (NHIS) documented that children with ASD have greater health care utilization than other children. This data set includes 483 children with ASD whose service utilization was compared to almost 85,000 children with typical development. The survey documented increased number of acute and chronic health care visits, use of PT/OT/speech therapies, mental health counseling, and chronic use of prescription medications for children with ASD compared to children without special health care needs (p < .001) (Gurney et al. 2006). Of the children with ASD in this sample, three quarters received therapy for an emotional, behavioral, or developmental problem. Families of children with ASD reported significantly more respiratory, food, and skin allergies, with food allergies accounting for the relative difference. This survey was based upon parent report of both diagnosis and service use and thus could be affected by parent recall, as well as inexact reporting of autism or other comorbid diagnoses. It did, however, identify differences in health care utilization that require further analysis. While much of the information analyzed from national data sets refer to the health care provider as being the pediatrician, in practice, many children receive primary care from family physicians, nurse practitioners, and physician’s assistants. Whether professionals with varied training backgrounds effect the Medical Home differently is not known at this time.

Parent Satisfaction and the Medical Home

Families of children with ASD often express dissatisfaction with their pediatric care provider related to components of the Medical Home, including family focused care, advocacy for special services, care coordination, and availability (Carbone et al. 2009; Brachlow et al. 2007; Liptak et al. 2006). In another analysis of the National Survey for Children’s Health, the caregivers’ perception were compared amongst the health care delivered to 495 children with ASD, 6,716 children with Asthma, and 11,403 other CSHCN. The caregivers reported on attributes of the Medical Home including preventive care, family focus, compassion, cultural appropriateness, accessibility, continuity, and coordination of care. Children with ASD were half as likely as other CSHCN to have family centered, comprehensive, and coordinated care (Brachlow et al. 2007). The families reported time with the provider as the major barrier to provision of the Medical Home. While they had equal access to primary care, families of children with ASD reported more difficulty accessing subspecialty care than children with asthma, but comparable difficulty to other CSHCN. The children with ASD had similar needs for additional services such as physical therapy, medical equipment, special education, and counseling as broad categories, to other CSHCN. However, families of children with ASD were less likely to report that their care providers discussed or coordinated the care received through these providers. Minority racial and ethnic status further depressed access to specialty care for children with ASD although children with ASD who had medical assistance or low cost insurance assistance, such as SCHIP, had equal access to general health care (Liptak et al. 2008). Parents of children with autism surveyed regarding their experiences with primary care providers expressed more dissatisfaction if there was a delay in initial diagnosis and if they utilized a greater number of therapies (Harrington et al. 2006). This may in part reflect the observations of Carbone et al. (2009) who noted that families whose physicians did not listen to their early concerns regarding delays had lower expectations for their participation in care related to ASD.

Few families in the focus group convened by Carbone et al. (2009) seemed to be aware of the concept of a Medical Home. They did not think of their pediatricians as providing care coordination. Care coordination is a significant need: over half of the families of children with ASD in the National Survey of Children with Special Health Care Needs reported that a caregiver had to reduce or stop work because of their child’s needs and over 25% of families reported that they must spend 10 or more hours per week coordinating their child’s care (Kogan et al. 2008). In this sample, 26.5% of children with ASD have a medical home compared to 48% of other children with SCHN. Having a Medical Home was associated with the report of decreased financial impact of obtaining medical care. Parents of children with ASD responding to a survey regarding their primary care providers felt their pediatricians did not answer questions about their child’s disability and did not have adequate knowledge to counsel them about complementary therapies compared to parents of children with motor or intellectual disabilities (p < .04). Overall they thought their pediatricians were sensitive and kept up with medical practice. They would have liked their pediatrician to connect them with other parents, have more family input to the care plan, and have more information on the child’s condition and preventive care (Liptak et al. 2006).

The dissatisfaction families report with care provision is not limited to conventional health services, but also extends to community and school related services, including occupational, physical and speech therapies, in addition to behavioral services. Families participating in the National Survey of Children with Special Health Care Needs whose children had ASD reported 3.39 times more difficulty getting therapeutic services than other CSHCN and 2.65 more dissatisfaction with those services (Montes et al. 2009). The major barriers described by families related to information transfer and communication between providers. Future research will need to evaluate the impact of the Medical Home on the health of children and families affected by ASD and evaluate the implementation of family–provider collaboration in care across systems.

Clinician Satisfaction and the Medical Home for ASD

The concern expressed by parents regarding the capacity of the primary care provider to meet the care needs of children with ASD is mirrored by the concerns of the providers regarding how to meet the needs of their patients. Challenges to the provision of a Medical Home for patients with autism include lack of time in a standard office visit for assessment and counseling, lack of capacity for care coordination due to the need for additional time, insufficient expertise in general and in specific with behavioral management, inadequate staffing for screening and care coordination, lack of reimbursement, lack of practice guidelines, and skepticism by families of traditional medicine or vaccines, in addition to encountering frequent use of complementary and alternative medicine by families (Golnik et al. 2009; Carbone et al. 2009, 2010). Pediatricians see themselves as advocates for their patients with SHCN, coordinating care with consultants and report they initiate discussions related to transition to adult services. Family focus groups discussing their care needs, however, did not see their pediatricians providing the care coordination, advocacy efforts, or discussion of transition that was reported by the providers. Many families contact their pediatricians with behavioral concerns about their children with ASD. Pediatricians report that they would like access to additional resources to help these families. The pediatricians interviewed by Carbone et al. (2009) noted that interaction with schools is usually facilitated by families. They recognize that families typically coordinate the nonmedical care of their children. Clearly, strategies to effect family collaboration in planning care, defining expectations, and enhancing communication between families, clinicians, other providers involved in the care of children with ASD, and the educational system would benefit from implementation of the Medical Home and promotion of family focused care.

Pediatricians in primary care expect to provide health supervision and anticipatory guidance, diagnose and treat acute and chronic disorders, and serve as the first contact of care for children with chronic illnesses. (Committee on Pediatric Workforce, AAP 2011). Families report dissatisfaction in limited access to subspecialty care. Families of children with ASD who are concerned that their primary care provider may have insufficient knowledge about comorbid conditions may bypass the primary care provider by seeking subspecialty consultation. This has been documented in referrals for neurology evaluation for children with ASD (Ming and Hashim 2011). In many insurance systems, referral for subspecialty evaluation must be recommended by the primary care provider. Children with ASD are reported to have increased rates of medical conditions such as constipation (Ibrahim et al. 2009), seizures (Bolton et al. 2011), and sleep disorders (Reynolds and Malow 2011). While subspecialty evaluation and care may be indicated, the pediatric provider has training and expertise in the initial evaluation and medical management of these disorders. Symptoms may not be unique to children with autism. The physician may not refer a child to a subspecialist if the clinical assessment does not indicate the referral is necessary. The system of medical referral for subspecialty care, varies across regions and providers. Issues related to under-referral of patients who would benefit from earlier subspecialty care and over-referral increasing the costs of medical care could potentially be improved with structured referral information provided to secondary care providers and secondary care acknowledgement of appropriate use of guidelines in primary care (Akbari et al. 2011). The perception of coordinated care with subspecialists is improved when the primary care provider initiates and communicates with the subspecialty referral source (OMalley and Cunningham 2009). Of the almost 3,500 adults surveyed in the 2007 Health Tracking Household Survey, half found their primary care providers informed about their subspecialty care when the referral came from their physician while only 35% found that to be the case with self referral. It is unknown whether the report of decreased access to subspecialty care for children with autism is increased by the real deficit in general mental health services for children, absence of funding for behavioral health services, desire for complementary therapies that are not covered by insurance, or true underreferral. Only prospective studies designed to examine evidence based health care in children with autism will be able to determine the true care needs of this population.

Toward a Medical Home for Children with ASD

The Medical Home can serve to support the family in the coordination of the multiple service systems required by children with ASD (Committee on Children with Disabilities 2001; Murphy et al. 2011).

Education

The knowledge base of child health care practitioners related to ASD has been a major focus of education and training efforts. Educational efforts regarding the medical and behavioral comorbidities of ASD have been directed at the primary care provider through the pediatric literature (e.g. Buie et al. 2010) and continuing medical education offerings on the local and national level. Education and support for the primary care provider caring for children and youth with ASD has been a targeted area of interest for the American Academy of Pediatrics with publication of the Autism Toolkit in 2007 and subsequent revisions planned for release in 2012 (AAP Toolkit). This product is designed to provide information to the primary care provider on screening for ASD and management of the common comorbid conditions. The Autism Treatment Network (www.atn.org) is examining both the prevalence of medical comorbidities of children enrolled in this national registry and the mechanisms that might enhance the quality of their care in primary and tertiary care systems (Coury 2010; Coury et al. 2009). Many educational efforts focusing on education of clinicians caring for children with ASD have been put into place by federal agencies (Centers for Disease Control and Prevention, Learn the Signs Act Early http://cdc.gov/ncbddd/actearly/index.html), professional societies (http://medicalhomeinfo.org/about/cocwd/autism.aspx), and not-for-profit advocacy organizations (e.g. First Signs https://firstsigns.org/treatment/index.htm). The CDC has developed a training curriculum that can be used in graduate medical education to prepare clinicians to screen and manage children and youth with ASD (http://cdc.gov/ncbddd/actearly/act/class.html) While the demand by practitioners for continuing professional education about ASD continues to grow, there is a need for an evidence base to support the development of health care algorithms for children with ASD in order to guide clinical care in the medical home.

Financial Barriers to Provision of the Medical Home

Whether a primary care provider practices in a solo private practice or a large multispecialty managed care organization, the provision of health care is a business. It is a unique business, in that health care in our society is a right and the responsibilities undertaken by the health care provider have ethical, as well as legal ramifications. Medicine is a business that provides a product—health and wellness, to the patient, and documentation to the insurer who purchases the care. Like any other business, there is a production line that supports the health care interaction that must serve to efficiently produce the product in a manner that covers the costs of the operation. This operation is increasingly complex and includes electronic medical records, malpractice insurance, and the operations costs common to other businesses such as office staff, rent, and salaries. Reimbursement depends on the clinician documenting for the insurers standard diagnoses (International Classification of Diseases for diagnoses, 9th edition) and billing codes that support relative value units (RVUs). These standardize comparable units of service based on time spent with a patient and the difficulty of decision making required by the visit. RVUs have been modified to attempt to be fair to pediatricians and other primary care providers with rates that compensate the cognitive, nonsurgical work that they do. The long hours and relatively low compensation for primary care are among the reasons that fewer medical students are entering primary care fields (Ebell et al. 2008).

Two barriers to provision of the Medical Home are the time required to provide more comprehensive care and the lack of compensation. For children in general, illness visits to pediatricians last on average 8 min and visits to address behavioral concerns last 20 min. Physician remuneration for behavioral visits is less per minute than medical visits (Meadows et al. 2011). Advocacy with insurers will be necessary for the Medical Home to be implemented in a financially viable fashion.

With the recommendation to perform developmental screening (AAP et al. 2007; Council on Children with Disabilities 2006; Johnson and Myers 2007), billing codes that allowed billing for screening tests distributed and scored by office staff were identified. This code does not cover the clinician time required for interpretation and counseling regarding results. Clinician time needs to be captured in the basic evaluation/management service (EM) billing by the provider during a face-to-face visit (Wegner and Macias et al. 2009). Billing codes for prolonged visit time that could cover non face-to-face time for case coordination are rarely reimbursed by insurers. Care coordination could also be enhanced if insurers reimbursed the clinician or designated staff using the appropriate service codes for prolonged visits, telephone management, and consultation with other medical, behavioral, and educational providers involved in the care of the child with ASD. Non face-to-face care coordination codes for Care Plan Oversight would allow for review of laboratory results, reading therapy reports, and communication with educational and behavioral professionals (Wegner and Macias 2009). Care coordination has been covered in some practice settings by grants and special arrangements with insurers. Care coordination coverage would be an important component of a Medical Home covered by universal health insurance. Until that time, providers will benefit from additional instruction on how to submit billing to maximize reimbursement for Medical Home services with current insurance limitations, in addition to use of other funding sources when available.

The next sections will discuss the specific activities expected within the Medical Home related to screening for ASD and other developmental disorders. Mandates, evidence, and reimbursement present conflicting impacts on implementation of these goals.

Developmental Screening for ASD in the Medical Home

Infants and toddlers in the United States are seen by their primary care provider 11 times for well child care by their third birthday. The pediatrician is typically seen as a trusted advisor on issues related to infant health and development and is in a unique position to be able to assess young children for many areas of risk during a time frame in which developmental intervention is possible. Since 2006, the American Academy of Pediatrics recommends that in addition to general developmental screening at 6, 9, and 12 months of age, that autism specific screening should be incorporated into the visits at 18 and 24 months of age with subsequent surveillance (Council on Children with Disabilities 2006; Johnson and Myers 2007). While developmental and behavioral screening is an important focus, several other age-specific topic areas to be covered as part of anticipatory guidance in well child visits include maternal depression, spousal abuse, nutrition, car seat use, helmet use, sleep hygiene, developmental stimulation through play and reading, and obesity prevention, amongst others (http://brightfutures.org/). All of these topics are important in the provision of a Medical Home. At any given well child visit, less than half of Bright Futures topic areas were observed to be addressed for CSHCN, although visits were over 1/3 longer than those for children without SHCN (Norlin et al. 2011).

Provider Satisfaction with Screening for Autism

Prior to 2006, pediatricians were expected to monitor the development of their patients informally through surveillance with use of screening tools only if concerns were identified (Dworkin 1989). Survey data indicated that few pediatricians followed the recommendation that informal surveillance be supplemented with objective screening (Dosreis et al. 2006). With revision of the AAP recommendations in 2006 to provide developmental screening at routine intervals, pediatricians have been encouraged to incorporate validated approaches to developmental screening as part of well child visits (Council on Children with Disabilities 2006). Developmental screening is mandated in the early periodic screening, diagnosis and treatment program (EPSDT) that provides the basis of well child care for children insured through medical assistance (http://mchb.hrsa.gov/epsdt/). Efforts to educate pediatricians about screening have resulted in a gradual incorporation of routine screening into primary care with 47.7% of pediatricians reporting use of a valid screening test in 2009 compared to 23% in 2002 (AAP Department of Research 2011). Survey data prior to the 2006 AAP developmental screening recommendations indicated that pediatricians did not screen for autism because they needed more education about screening tools (62%), referred children of concern to specialists (47%), or did not have enough time for screening (32%) (Dosreis et al. 2006).

However reasonable it seems to include standardized instruments for screening for ASD into well child care, the barriers include the accuracy of existing validated instruments (Al-Qabandi et al. 2011), the time necessary for screening, the cost of screening, the comfort of the care provider in referrals and management of children with ASD prior to diagnostic referral, and referral resources for diagnosis and treatment in the community (Dosreis et al. 2006; Gura et al. 2011). Surveillance alone is inaccurate in early identification of autism. Pediatricians flagged 4 of 21 children diagnosed with ASD in one large study where 4,797 children were screened for ASD in primary care settings (Robins 2008). While the positive predictive value of the Modified Checklist for Autism in Toddlers with follow up interview was disappointing at 57%, the inaccuracy of surveillance alone points to the need for a structured approach to screening for a disorder with a population prevalence of 1:110 (Autism and developmental disabilities monitoring network surveillance year 2006 principal investigators (CDC) 2009). There continues to be controversy regarding the recommendation for universal screening on the basis of concerns that surround the accuracy of screening, limited evidence on the improvement in outcome from early diagnosis and therapy, unnecessary worry to children who are not diagnosed with ASD despite positive screening, and limited resources to meet the demand for evaluation and treatment in the community that would need to absorb potential false positives (Campos-Outcault et al. 2011; Al-Qabandi et al. 2011).

Parent Satisfaction and Screening for Autism

Families of children diagnosed with ASD are vocal in their dissatisfaction that developmental and behavioral concerns raised with primary care providers resulted in reassurance or monitoring rather than in referral for assessment and intervention (Carbone et al. 2009). Routine screening furthers the collaboration of families and their primary care provider in monitoring and advancing the development of the child. Formal screening provides a mechanism to ensure that all families have the opportunity to discuss developmental and behavioral concerns with their health care provider. While the overwhelming majority of families correctly identify that their child has a developmental or behavioral need, some parents, especially families who are not proficient in English, will not express these concerns unless specifically asked (Glascoe et al. 1997). An important area for research will be development of improved screening tests. Given the prevalence of ASD and the potential for developmental and behavioral improvement with early intervention (Dawson and Burner 2011), a low cost, validated, and standardized population level approach to screening in the Medical Home is desirable (Pinto-Martin et al. 2008).

Toward Screening for Autism in Primary Care

Screening for autism has been a major focus for professional education by federal and state governments and the American Academy of Pediatrics. For screening programs to be effective, several components need further attention and research.

Valid Screening Test

Sensitive and specific screening tools for ASD for use in primary care sites need to be developed that are both cost effective and easy to implement and interpret. While the screening tests used for general developmental delay are not sensitive to the symptoms of ASD (Pinto-Martin et al. 2008), children who are identified through ASD screening are likely to be at high risk for other developmental disorders on further testing (Robins et al. 2008; Pierce et al. 2011). The MCHAT is the most commonly used screening measure at this time. It is possible that measures such as the Communication and Symbolic Behavior Scales Developmental Profile Infant-Toddler Checklist might have utility in identification of ASD and other developmental disorders such as language delay at very young ages (Pierce et al. 2011).

Education

A significant barrier to developmental screening by pediatricians was reported to be the lack of familiarity with the tests used and confidence in how to manage positive screens (Dosreis et al. 2006). Since publication of the AAP developmental screening guidelines in 2006 (Council on Children with Disabilities 2006) and the diagnosis and management of autism clinical report in 2007 (Johnson et al. 2007), the AAP, CDC, states, and other organizations that advocate for early diagnosis of autism have all contributed educational programming to teach primary care providers how and when to use existing screening tests. Multiple strategies for professional education are necessary to effect practice change. Allen et al. (2010) reported on a systematic effort to educate pediatricians in Illinois in screening for ASD. The rate of screening increased to 85%. While screening rates can be increased with concerted effort, the impact that screening has on early identification, early diagnosis, access to services, and outcome needs careful study.

Although screening tests themselves may be flawed, the diagnosis of autism made at 2 years of age by expert clinicians is stable through childhood (Charman et al. 2005). A small percentage of children diagnosed at 2 years of age do not fulfill diagnostic criteria for autism at ages 4 and 9 years. These children, however, very often have alternate DSM IV behavioral or cognitive diagnoses (Lord et al. 2006). It is unknown whether children with diagnoses other than ASD have had their outcome modified by the interventions they experienced or if the symptoms necessary to meet diagnostic criteria for ASD at younger ages and lower language levels were not specific for autism.

Financial and Time Barriers to Screening

Barriers to implementation of screening include the cost and the time necessary to incorporate screening into current well child care (Golnik et al. 2009). Models have been presented that allow for use of quality improvement methodology to implement screening in primary care (Gura et al. 2011; Miller et al. 2011). The MCHAT has improvement in sensitivity and positive predictive value with the addition of a structured interview done by the clinician. This adds time to the scheduled well child visit or requires that the family return for another visit entirely to complete the interview. Gura et al. (2011) reported that the reimbursement for the office staff to administer the MCHAT questionnaire and enter it into the electronic record took 5 min per patient and was covered by the insurance reimbursement for the screening code (26110). They did not report, however, on the reimbursement for the E/M code for the clinician time for a follow-up interview. In order for screening to be routinely performed there will need to adequate reimbursement for the process. Families will need to expect that in order for the clinician to have adequate time to screen a child for developmental or behavioral differences, as well as provide any necessary referrals and counseling, it may require a return visit. Collaborative care with families would include dialogue around developmental and behavioral concerns.

On Beyond Screening: Diagnosis and Referral for Services

While the AAP algorithm recommends screening for developmental delays and referral to the local Early Intervention program when delays are identified, practitioners refer only 61% of children who fail screening (King et al. 2010). Although federal law mandates evaluation through the Early Intervention program of children less than 3 years of age referred for assessment of developmental delays, there are limited resources available for both diagnosis and intervention for autism (Wise et al. 2010). Primary care practitioners refer children to Developmental and Behavioral Pediatricians for diagnostic evaluation of ASD (Dosreis et al. 2006) as well as specialists in psychology, child and adolescent psychiatry, and child neurology. The AAP clinical report of 2007 (Johnson et al. 2007) recommends multidisciplinary team assessment for autism diagnosis. In some communities this requires multiple appointments through a tertiary care center, in others an expert team utilizes cognitive and language testing collected through the Early Intervention or school based assessment system. In either case, the wait for diagnostic assessment by a subspecialist may be prolonged. Parents often cite delays in diagnosis as one source of great dissatisfaction with their primary care practitioner (PCP) (Keenan et al. 2010). Delay between screening and diagnosis places the PCP and family in an uncomfortable limbo together. The child might receive speech and language, or motor therapies, and/or educational intervention for tested delays until the diagnosis of an ASD can be made and services provided in a more disorder specific fashion. Programs to improve the capacity for developmental diagnosis in primary care may help decrease delays in ASD diagnosis (Kobak et al. 2011). While pediatricians have training in child development, the differentiation of subtle or atypical presentations of ASD from other developmental disorders in young children such as intellectual disabilities, language disorders, and other behavioral health diagnoses, typicially requires additional evaluation. Federal support through the Leadership Education in Neurodevelopmental Disabilities (LEND) and Developmental and Behavioral Pediatrics programs is aimed at increasing capacity for training clinicians in relevant fields.

The AAP recommends that pediatricians caring for CSHCN refer children to Early Intervention or school services for their intervention needs, arrange for appropriate etiologic evaluations, provide health monitoring to prevent secondary disabilities, and collaborate with the other agencies and professionals involved with the child’s care (Committee on Children with Disabilities 2001). Associated medical conditions that may impact the provision of intervention services include seizures, sleep disorders, feeding disorders, and constipation among others (Carbone et al. 2009). Improved communication between the medical care providers, the family, and the providers of educational and behavioral interventions are necessary to enhance successful outcomes. Once children are referred for intervention through educational systems, there will need to be a simple and cost efficient system for bidirectional information sharing. Optimal outcome, whatever that may be, for an individual child with ASD requires a continuum of collaboration between health care providers, behavioral therapists, the educational team and the family (Fig. 1).

Fig. 1
figure 1

Continuum of care for Young Children with ASD

Full size image

Conclusion

Primary care providers are faced with many obstacles when caring for CSHCN, including autism. It is important that providers are well-equipped with the skills needed to first recognize and then subsequently manage children with ASD. Early identification of autism spectrum disorders permits early educational interventions that help to foster the best possible outcomes. A brief and validated screening tool specific for autism spectrum disorders for general use in all children at key well-child visits or when concerns are raised would further this aim. Given the workload and time constraints faced by primary care providers, reimbursement for the time necessary to administer and address the findings of a screening tool is critical to meet the public health challenge for early identification. Availability of providers to make confirmatory diagnoses and educational and behavioral professionals to provide intervention requires commitment of funding for services and for training the professionals needed. The family and primary care provider have shared goals for collaboration in the management of the child with ASD. However, for true care coordination to be feasible within the Medical Home, support for confidential and efficient communication between providers, in addition to financial support for the time required to address care coordination within the business of medicine, needs to be a focus of joint advocacy by parents and providers.