Medical Home Care and Educational Services for Children and Youth on the Autism Spectrum: A Scoping Review

The research questions were: (1) What is known about medical home care and its association with educational services for CYAS? and (2) What factors are associated with medical home care and educational services in CYAS?

The first author (SR) received assistance from a health sciences reference librarian with experience in scoping review methodology to conduct the literature search. Publications were limited to those published between 2006 through December 2022. The 2006 selection was guided by the AAP's release of an algorithm for developmental monitoring in primary care, targeting early detection and EI referral and establish rules to enhance monitoring and screening of children and youth with disabilities in medical home care. The algorithm includes autism-specific screening at 18–24 months to facilitate early detection and referral of children at risk or diagnosed with autism for detailed evaluations and potential early intervention (EI) (AAP; 2006; AAP, Council on Children with Disabilities, Section on Developmental Behavioral Pediatrics, Bright Futures Steering Committee, 2006).

We searched key terms related to autism, medical home care, and educational services. These key terms were refined and broadened per scoping review guidelines (Arksey & O’Malley, 2005; Aromataris & Munn, 2020; Levac et al., 2010; Peters et al., 2021). ‘Autism’ was broadened to include ‘ASD,’ ‘autistic,’ and ‘Asperger.’ ‘Medical home care’ was broadened to include ‘primary care.’ ‘Educational services’ was broadened to include ‘education,’ ‘IEP,’ ‘IFSP,’ ‘individualized education plan,’ ‘individualized family services plan,’ ‘early intervention,’ and ‘educational.’

Five database searches yielded 365 publications: PubMed (n = 122), CINAHL (n = 104), PsycINFO (n = 87), ERIC (n = 17), and Education Research Complete (n = 36). Ten additional publications were identified through backward and forward reference and citation searching of publications retrieved from a grey literature search (i.e., Google Scholar). The first author (SR) screened titles and abstracts to assess whether publications addressed autism, medical home care, or educational services, including key and/or broadening terms. Publications were included if they met these criteria: (1) children or youth diagnosed with autism spectrum disorder, ASD, Asperger’s syndrome, autistic disorder, or autism based on clinical or parent-reported data, (2) between the ages of 1–17 years old, (3) studies that are available in English language, (4) conducted in the US, (5) published between 2006 to December 2022, and (6) could include policy and position statements. Our study targeted children and youth on the autism spectrum ages 1 to 17 years old, in accordance with national surveys like the National Survey of Children's Health that typically use caregiver-reported data on children ages 0–17 and their families. We excluded publications that focused solely on health and/or medical services or educational services in CYAS, but not both, because they would not be examining how these two service systems work together. The search of the five databases and grey literature resulted in 365 publications screening, and after duplicates were removed, 225 publications remained for title and abstract screening completed by the first author (SR). Both authors piloted inclusion criteria by independently reviewing (Aromataris & Munn, 2020) 10 of the 62 publications. Since no discrepancies were noted during the pilot phase, and no adjustments were made to the inclusion criteria, both authors proceeded to independently perform full text reviews of the remaining 52 publications. The first author (SR) independently reviewed 15 publications. During full text review, 44 of 62 publications were excluded. The first and third authors (SR, ZS) met to discuss and resolve conflicts on the rationale for excluding (see Table 3 in Appendix) 10 of these publications. A total of 18 publications met inclusion criteria based on title and abstract screening and full text review (see PRISMA flowchart (Fig. 1)).

For Aim 1, the first author (SR) extracted relevant characteristics for included publications: author(s), publication year, study purpose, study design, respondent type, sample characteristics (e.g., child sex, child age, child race/ethnicity, caregiver education level, child insurance type) medical home care component(s) addressed, type(s) of educational service, type of association (positive, negative, or no association) between medical home care and type(s) of educational services.

For Aim 2, two authors (SR, DB) collated and created a summary of factors attenuating the relationship between medical home care and educational services, according to the Andersen Behavioral Model of Health Services Use (BMHSU) (Andersen, 1968). Predisposing factors are individual or family traits that can affect their likelihood to use services. These factors include demographics (e.g., age or ethnicity), and sociocultural elements (e.g., beliefs and health knowledge). Enabling factors are conditions that affect the service use. They can include service availability and accessibility, financing, and the existence of social support or health care infrastructure. Need factors are recognized or assessed health issues that require the service use. In collating the content from the included publications, we identified patterns and explanations that described the factors attenuating the relationship between medical home care and educational services. The first and fourth authors (SR, DB) began with creating themes to broadly capture the type of factors most cited in the available evidence and sorted the various factors under each of the four themes (see Table 2). After sorting specific factors according to type, the first author (SR) categorized each predisposing, enabling, and need factor using the Andersen BHMSU. The first and fourth authors (SR, DB) then discussed the sorting process, and discrepancies noted in the “Provider Training, Knowledge, Awareness, and Skills” and “Organizational Resources and Workflows” were discussed until we reached consensus.

Included publications were summarized using descriptive statistics (frequency counts and percentages) to provide an overview of available evidence on medical home care in relationship to educational services, considering differences in predisposing, enabling, and need factors.

Table 1 illustrates that the 18 included publications across multiple formats. Eight original publications used cross-sectional (n = 4; 22%), prospective cohort (n = 1; 5%), mixed-method (n = 1; 5%), randomized controlled trial (n = 1; 5%), or qualitative (n = 1; 5%) study designs. One original research publication utilized secondary analyses of a population-based data source (i.e., National Survey of Children with Special Health Care Needs) (Sobotka et al., 2016).

Original publications that focused on children and youth samples who were predominantly male (Carbone et al., 2010, 2016; Dick et al., 2021; Sobotka et al., 2016), with age ranges spanning from 2 to 35 years old (Carbone et al., 2010, 2016; Dick et al., 2021; Williams et al., 2012). Most caregivers were White, non-Hispanic (Carbone et al., 2010; Dick et al., 2021; Sobotka et al., 2016; Williams et al., 2012). Most CYAS held private insurance and belonged to households where English was the primary language (Carbone et al., 2010). Moreover, their caregivers had an education level that extended beyond high school. While the study by Williams et al. (2012) had a higher age limit compared to ours, most of our age categories were included within their selection parameters.

In the original research publications involving health care providers, these respondents included physicians or primary care providers (e.g., pediatricians, family medicine physicians, nurse practitioners) (Carbone et al., 2016; Dang et al., 2017; Edwards et al., 2021; Sohl et al., 2022; Williams et al., 2012), and caregivers of CYAS (Carbone et al., 2010; Dang et al., 2017; Dick et al., 2021; Sobotka et al., 2016). In those original research publications involving educational service professionals, respondents included EI providers (Ibañez et al., 2019), early child educators or teachers (Buranova et al., 2022; Dang et al., 2017), school psychologists (McClain et al., 2020a, 2020b), and skilled therapists (e.g., speech and language pathologists, occupational therapists) (Buranova et al., 2022)).

In 18 of the included publications, there is an implicit or explicit focus on one or more components of medical home care. Per the NSCH definition of medical home care, these publications captured the components of experiencing a personal doctor or nurse (n = 2; 11%) (Sobotka et al., 2016; Williams et al., 2012), usual source of sick care (n = 2; 11%) (Sobotka et al., 2016; Williams et al., 2012), family-centered care (n = 10; 56%) (Aloisio et al., 2020; Buranova et al., 2022; Carbone et al., 2010; Dick et al., 2021; Duby, 2007; Hyman & Johnson, 2020; Shahidullah et al., 2020; Sobotka et al., 2016; Williams et al., 2012) referrals (n = 16; 89%) (Aloisio et al., 2020; Buranova et al., 2022; AAP, Council on Children with Disabilities et al., 2006; Dick et al., 2021; Duby, 2007; Edwards et al., 2021; Hyman & Johnson, 2020; Shahidullah et al., 2018; Williams et al., 2012) and effective care coordination (n = 13; 82%) (Aloisio et al., 2020; Carbone et al., 2010; AAP, Council on Children with Disabilities et al., 2006; Dang et al., 2017; Duby, 2007; Ellerbeck et al., 2015; Lobar, 2016; McClain et al., 2020a; Sobotka et al., 2016; Shahidullah et al., 2018, 2020; Williams et al., 2012) (See Table 1).

Similarly, all 18 included publications focus on one or both educational service type(s). Seven of the 18 (39%) included publications examined educational services inclusive of both EI and special education, whereas 7 publications solely examined EI services (39%), and four publications solely examined special education services (22%) (see Table 1). Nine of the 18 included publications addressed multiple medical home care components and both EI and special education service types (Aloisio et al., 2020; Buranova et al., 2022; Carbone et al., 2010; AAP, Council on Children with Disabilities et al., 2006; Duby, 2007; Edwards et al., 2021; Shahidullah et al., 2020; Sobotka et al., 2016; Sohl et al., 2022) (See Table 1).

As shown in Table 1, five studies have highlighted associations between the medical home care components and educational services. Among these studies, one demonstrated a negative association, while another indicated a positive one.

As shown in Table 2, there was 1 predisposing factor and 12 enabling factors each described as attenuating the association between medical home care component(s) and educational services for CYAS and their families (Andersen et al., 2013). Prior publications have not cited need factors as attenuating or intensifying the relationship between medical home care component(s) and educational services for this population.

Select child and family demographic characteristics (Andersen et al., 2013) that have so far been included for their potential role in predisposing families to inequitable access to medical home care component(s) and educational services include: (1) primary household language (Sobotka et al., 2016), (2) child race/ethnicity (Dick et al., 2021; McClain et al., 2020a; Sobotka et al., 2016), (3) caregiver education level (Dick et al., 2021; McClain et al., 2020a), and (4) geographic region (Buranova et al., 2022; Edwards et al., 2021). Caregivers who were non-Hispanic White mothers with higher education levels, and older children with private insurance, were significantly less likely to report receiving help as they needed with care coordination, inclusive of doctors’ communication with their child’s school, EI program, or other educational setting (Sobotka et al., 2016). Additionally, families who reside in rural areas lack access to adequate services due to the limited number of providers available to aid children, youth, and families (Buranova et al., 2022; Edwards et al., 2021).

Three levels of enabling factors were included in prior publications for their hypothesized role in attenuating the relationship between medical home care component(s) and educational services for CYAS and their families (McClain et al., 2020a; von Lengerke et al., 2013). At the provider level, lack of adherence to AAP guidelines on autism surveillance and screening (AAP, Council on Children with Disabilities et al., 2006; Hyman & Johnson, 2020; Shahidullah et al., 2018; von Lengerke et al., 2013), hesitancy to discuss early autism concerns and responsiveness to caregivers’ concerns about their children’s development (AAP, Council on Children with Disabilities et al., 2006; Carbone et al., 2016; Edwards et al., 2021) lack knowledge of autism and complex needs, lack of knowledge on how to direct caregivers to educational services (e.g., referral, service eligibility criteria) were regularities cited as attenuating factors (Carbone et al., 2010; Dick et al., 2021; Sobotka et al., 2016; Shahidullah et al., 2018; Lobar, 2016; von Lengerke et al., 2013). At the program level, limited guidelines on how medical home care and educational service should coordinate (Shahidullah et al., 2018), transitions that are specific to the educational setting (i.e., EI, school, post-secondary) (Carbone et al., 2010; Hyman & Johnson, 2020), insufficient provider availability to meet the volume of service needs, lack of medical home care providers’ time to manage the coordination, long diagnostic and service wait times, and reimbursement for providers’ time expended on coordination were factors that limited organizational capacity to facilitate coordination with educational settings (Carbone et al., 2010; Dick et al., 2021; Shahidullah et al., 2018; Sobotka et al., 2016; Williams et al., 2012). At the service system level, differences in organizational or structural characteristics (i.e., funding, intervention, and documentation) and various local and state laws related to service eligibility (Hyman & Johnson, 2020; McClain et al., 2020b), and provision acted as barriers to medical home care and educational services (Aloisio et al., 2020; Carbone et al., 2010; Shahidullah et al., 2018). Privacy and confidentiality policies specific to medical home care (i.e., HIPAA) and to educational settings (i.e., FERPA) were cited as attenuating factors to cross-sector care (McClain et al., 2020a).

Our first aim was to characterize the association between medical home care and educational services for CYAS. Five original research studies have underscored the relationships between components of medical home care and educational services. Of these studies, one found a negative correlation, while another study showed a positive association (McClain et al., 2020a, 2020b; Sobotka et al., 2016).

This finding can be attributed to the small number of original research publications that we identified, which used a variety of methodologies (e.g., randomized control trial, mixed method approach, secondary analyses) to explore medical home care component(s) in relationship to educational services for CYAS (see Table 1). Alternatively, the five publications that estimated associations did not model all medical home care component(s) in relationship to educational services (see Table 1), preventing a comprehensive measurement of this association and how CYAS access educational services through the medical home care pathway.

In addition, the emphasis on the outcome of educational services, relative to EI, and less with special education, is another future direction. Given that many CYAS access special education services longer, compared to EI (Aloisio et al., 2020; AAP, Council on Children with Disabilities et al., 2006; Carbone et al., 2010; Dick et al., 2021; Duby, 2007; Ibañez et al., 2019; Sobotka et al., 2016), and effective care coordination (n = 13; 82%) (Aloisio et al., 2020; Carbone et al., 2010; AAP, Council on Children with Disabilities et al., 2006; Dang et al., 2017; Duby, 2007; Ellerbeck et al., 2015; Lobar, 2016; McClain et al., 2020a, 2020b; Sobotka et al., 2016; Shahidullah et al., 2018, 2020; Williams et al., 2012) which is a more limited time service, addressing this imbalance will provide a better understanding of how CYAS access special education to determine the extent to which their service needs are being addressed.

There have been several clinical and practice reports published alongside original research studies (cross-sectional, prospective cohort, mixed-method, randomized control trial, qualitative), but only one publication involved secondary analyses of population-based data. We recently determined a negative association between medical home care and educational services for CYAS based on population-based data (Rizk et al., 2023). Studies like this can benefit from data sources that include a robust definition to reflect AAP's intent on how medical home care should be provided (AAP, 2001), for consistent operationalization of the medical home care concept (Child and Adolescent Health Measurement Initiative, 2009). In addition to NSCH and other relevant population-based data sources, other more localized data sources (e.g., electronic health record data) may also be useful to confirm, disconfirm, and/or extend knowledge about the association between medical home care with educational services. Furthermore, it should be noted that the AAP definition is a medical definition and therefore has limitations when used to examine access and use to other services, including educational services.

We used the Andersen BMHSU (Andersen, 1995; Andersen et al., 2013; Babitsch et al., 2012) to explore the available evidence on predisposing, enabling, and need factors attenuating the access educational services in relation to healthcare utilization (i.e., medical home care). We found that enabling factors were the most salient in examining the association. In contrast, we found limited evidence on the role of predisposing factors, and no evidence on the role of need factors, as attenuating or intensifying the association between medical home care and educational services.

Enabling factors (Andersen et al., 2013) captured reflect the distinguished and unique features of medical home care and educational services, in terms of their service intent, design, and delivery. Based on the Andersen BMHSU, we identified three categories of enabling factors: (1) provider training, knowledge, awareness, and skills; (2) organizational resources as impacting provider workflows; and (3) health and educational policies (see Table 2).

A first type of enabling factor relates to provider training, knowledge, awareness, and skills to discuss autism and its associated comorbidities (e.g., attention difficulties, anxiety, sensory issues, sleep disorders) (Aloisio & Huron, 2020; Williams et al., 2012). In Williams et al.'s study, more than half of the sampled physicians had worked more than 15 years in practice. There were many physicians who have never attended a Continuing Medical Education on autism. Without ongoing training, physicians will be unable to meet even the basic needs of children, youth, and their families. Edwards et al. (2021) examined provider hesitancy to discuss early autism concerns and responsiveness to caregivers’ concerns about their children’s development. Current evidence suggests that medical home care providers can increase caregiver awareness of educational services, respond to caregiver concerns, and adhere to recommended autism screening timelines. (Carbone et al., 2010; AAP, Council on Children with Disabilities et al., 2006; Duby, 2007; Ellerbeck et al., 2015; Hyman & Johnson, 2020; Ibañez et al., 2019; Sobotka et al., 2016; Williams et al., 2012).

A second type of enabling factor relates to several organizational resources as impacting workflow, including lack of provider time, long diagnostic and service wait times, and workforce capacity with insufficient provider availability (Aloisio et al., 2020; Buranova et al., 2022; Carbone et al., 2010; Dick et al., 2021; Duby, 2007; Edwards et al., 2021; Ellerbeck et al., 2015; Hyman & Johnson, 2020; McClain et al., 2020b; Sobotka et al., 2016; Shahidullah et al., 2018). The AAP (2006) recommends well-child visits every month until approximately 2 years of age, then yearly or as needed after that, whereas EI providers often see families weekly (Dick et al., 2021). Carbone et al. (2010) observed a shortage of physicians who feel qualified to care for CYAS and other developmental impairments and would benefit from more resources (e.g., family groups and schools) to support medical home care. For CYAS, resource limitations may lead to infrequent and brief interactions with primary care providers. A significant majority of primary care providers (85%) recognized that time constraints are the principal challenge when caring for this demographic. A smaller group (17%) noted the need for devoting more time for patient interaction and advocating for administrative accommodations, such as specialized appointment slots for CYAS. Simultaneously, providers understand that spending more time with CYAS and their families to ensure adequate care necessitates appropriate compensation. To this end, 8% of providers emphasize the potential benefits of improved reimbursement strategies and a comprehensive understanding of billing and coding systems. This would ensure that the additional time they spend with CYAS and their families is adequately compensated (Mazurek, et al., 2020).

Differences in health and educational policy are a third type of enabling factor. These policy differences may be related to service eligibility (e.g., DSM-V versus IDEA) (Ellerbeck et al., 2015; Hyman & Johnson, 2020; Lobar, 2016), such as differences in autism medical diagnostic criteria and autism educational criteria used to determine eligibility for special education services (McClain et al., 2020a, 2020b; Hyman & Johnson, 2020). Financing barriers include reimbursement for coordination efforts and longer health care visit durations (Carbone et al., 2010), where the time required by primary care providers to coordinate care and consult with other services may exceed what insurance reimburses (Dick et al., 2021). Furthermore, educational services are supplied at no cost to families (Hyman & Johnson, 2020)), as opposed to health-care costs that rely on third-party payers (e.g., health-care systems) (Ellerbeck et al., 2015). Both service systems have their own unique documentation processes. For example, electronic health records are utilized in the health care system (Aloisio & Huron, 2020; Ellerbeck et al., 2015), whereas educational service plans are outlined in IFSPs or IEPs (Ellerbeck et al., 2015; Hyman & Johnson, 2020).

Research on predisposing and need factors is needed when examining the association between medical and educational services for CYAS (Babitsch et al., 2012; Andersen et al., 2013). The absence of need factors can be explained by the fact that medical home providers vary in their training, and this may hinder their ability to accurately capture important need factors, such as condition severity (Duby, 2007) or address many of the comorbid conditions (e.g., attention deficits, anxiety, sensory issues, sleep disorders), associated with autism (Aloisio & Huron, 2020; Williams et al., 2012). Similarly, we contend that need factors, while important, were muted in these studies possibly because “need” is implied with CYAS. These children and youth may have other additional health and social needs that should be monitored and addressed through better cross-sector care. Most of the factors in this review that have been shown to attenuate the relationship between medical home care and educational service use are also factors that if addressed can increase cross-sector care between medical home care and educational services.

There is a possibility that the relationship between medical home care and educational services is indirect, such as by means of medical home care referrals to allied health services provided in school rather than in hospitals or clinics. Referring patients and families to specialists based on positive or elevated screening scores is within a primary care provider’s scope of practice (AAP, 2001), and this practice is further endorsed by IDEA Subpart D that defines primary referral sources to educational services as including hospitals, physicians and other clinics and health care providers (U.S. Department of Education, 2017) (p. 14, lines 14–22). If future studies measured all medical home care components (e.g.., referral), we may test more specified hypotheses about its association with educational services. For example, it would be possible to examine potential indirect links between medical home care referral to educational services by another mitigating factor (e.g., use of specialized therapies).

This review has four key limitations. First, it is possible that we have missed publications that did not mention medical home care or educational services in their title or abstract. Second, data extraction from included publications was performed independently by the first author (SR), which could have resulted in incorrect data classification. Third, results are limited to children and youth ages 1–17 years old and do not generalize to transition-aged youth on the autism spectrum who may be continuing to receive medical home primary care through their pediatrician as part of their transition to adult health care services. Fourth, the Andersen BMHSU has traditionally been used to measure access and use of health care services based on predisposing enabling, and need factors (Andersen et al., 2013). We explored how it could be applied to other services, such as educational services, as part of our review. Our review included 18 publications, of which five studies have underscored the relationships between components of medical home care and educational services. Of these five studies, one found a negative correlation, while another study showed a positive correlation. This scoping review provides new insights into the topic of medical home care as it pertains to educational services. The results pinpoint three areas that warrant further investigation: (1) limited evidence on the association between medical home care component(s) and educational services; (2) limited use of relevant population-based data sources on this topic; and (3) need to consider a broader range of factors associated with medical home care and educational services. Each of these areas is discussed as they guide future research directions. In addition to the present study, subsequent studies using population-based data sources will further examine medical home care on a granular level to identify the component(s) of medical home care that are closely related to educational services as well as factors attenuating their association for CYAS.