Barriers to service and unmet need among autistic adolescents and young adults

Abstract

Many of the nearly six million autistic adolescents and adults in the United States require support to navigate daily life. Family members often provide the first line of support for autistic youth by providing care and coordinating services. Although considerable research has examined the perspectives of family members caring for young autistic children, comparatively less has focused on those caring for transition-age youth who often struggle to access needed services as they leave child-serving systems of care. This study examined caregiver-reported barriers to service for 174 adolescents and young adults on the spectrum (ages 16–30) and the association between such barriers and unmet service needs. Exploratory factor analysis suggested two service barrier domains: access (e.g., cost) and quality (e.g., providers not trained). Regression models indicated that caregivers whose youth were diagnosed at older ages perceived both greater access and quality barriers. Male caregivers reported fewer access barriers, and those who perceived greater caregiver burdens (daily life disruptions, financial difficulties, and worries) reported more access barriers. Caregivers whose youth lived with them reported fewer quality barriers. Greater access–but not quality–service barriers predicted greater unmet service needs. Findings have implications for service delivery to autistic youth and specific directions for future research.

Lay Abstract

Prior studies have described the roadblocks, or barriers, to needed services experienced by families with young autistic children, but less research has focused on those faced by autistic adolescents and young adults. In this study, we wished to understand the barriers to service experienced by autistic adolescents and young adults and their families. We surveyed 174 caregivers of autistic youth between 16 to 30 years old. We found that caregivers who felt more caregiving burden had more difficulty accessing services for their youth. Specifically, caregivers who felt more strongly that their daily lives had been disrupted, felt more financial strain, and worried more about their youth well-being experienced more roadblocks to getting services for the youth. Male caregivers also reported fewer difficulties related to service access. Importantly, the older the youth was when they had been diagnosed with autism, the more service barriers their caregivers reported. We did not see any differences in the level of barriers experienced by youth who lived in urban versus suburban settings, or between white and non-white families. However, when youth lived with their caregivers (rather than, for example, in a group home), fewer quality-related barriers to services were reported. Finally, greater access (but not quality) barriers were linked to youth having more unmet service needs. These findings can help to reduce the barriers to service experienced by autistic adolescents and young adults and their families.

Keywords

autism spectrum disorders family caregivers unmet service needs obstacles

Autism spectrum disorder (hereafter “autism”) is a lifelong neurodevelopmental disability, largely diagnosed during childhood, and commonly characterized by social and communication difficulties and the presence of restricted and repetitive behaviors (American Psychiatric Association, 2013). Autism commonly co-occurs with other developmental or neuropsychiatric conditions such as intellectual disability, epilepsy, sleep disorders, and anxiety (Al-Beltagi, 2021). Recent estimates suggest that over 0.5 million adolescents and nearly 5.5 million adults in the United States have an autism diagnosis (Dietz et al., 2020; Kogan et al., 2018). Youth on the autism spectrum often face unique difficulties and require aid with developmental tasks such as navigating peer relationships, seeking and securing employment, and moving toward independent living (Bennett et al., 2018). Compared to their older counterparts, transition-age autistic adolescents and young adults experience greater difficulty with daily living and interpersonal skills (Matthews et al., 2015). Many autistic youth require assistance in learning autonomy and responsibility, maneuvering social situations, and finding community spaces of belonging (Kersten et al., 2020).

The complex and enduring needs associated with autism render ongoing service use critical for transition-age youth. Less than 20% of young adults with autism live independently (Roux et al., 2015). In addition to serving as caregivers, family members must often function as service coordinators (Roux et al., 2015). Given their prominent roles in supporting individuals throughout the life course, caregivers are a critical source of knowledge regarding the service needs and barriers faced by autistic youth. Families often face a volatile service landscape due to the “service cliff” encountered when services provided under the Individuals with Disabilities Education Act cease (Ishler et al., 2022; Shattuck et al., 2011).

Research has rarely moved beyond documenting the service loss that often accompanies high school exit to explore the specific barriers experienced by autistic youth when attempting to access and use services. Studies have focused largely on the experiences of young children, with considerable attention to obstacles to diagnosis and early intervention (Smith-Young et al., 2018). Most existing studies have been qualitative in nature. Several studies have examined the lived experiences of autistic youth as they navigate adult service systems (Anderson et al., 2018; Coleman-Fountain et al., 2020). Others have identified potential barriers from the perspective of service providers (Havlicek et al., 2016) or broad stakeholder groups that included researchers, providers, and family members (Kuo et al., 2018). A growing body of work documents the barriers encountered in transitioning from pediatric to adult health care (see Doherty et al., 2020).

Quantitative research on barriers to service is somewhat limited, especially those faced by autistic adolescents and young adults. Service barriers have been a central focus of just a handful of studies (Dudley et al., 2019; Pickard & Ingersoll, 2016; Platos & Pisula, 2019; Vogan et al., 2017). A few additional studies (Jose et al., 2021; Lai & Weiss, 2017) have examined barriers as a focal predictor of service receipt. Descriptive data on barriers to service can also be extracted from a number of studies which—while focused primarily on service use and unmet service needs—also measured and reported barriers to service (e.g., Kuhlthau et al., 2016; Taylor & Henninger, 2015). Existing data point to some common barriers experienced by autistic youth and their families. For example, caregivers generally affirm services as being unavailable or too far away (Dudley et al., 2019; Taylor & Henninger, 2015). Information about services is reportedly difficult to locate (Havlicek et al., 2016; Vogan et al., 2017), and services are often described as costly, difficult to navigate, hard to qualify for, and/or not offered at convenient times (Jose et al., 2021; Pickard & Ingersoll, 2016; Platos & Pisula, 2019; Taylor & Henninger, 2015). Caregivers commonly report insufficient provider knowledge of autism (Kuhlthau et al., 2016) and some families relate negative experiences with providers (Vogan et al., 2017).

Few studies have explored whether barriers to service for autistic youth are associated with specific sociodemographic, clinical, and contextual characteristics. Much existing work emanates from international research teams. While generally informative, such work provides limited insight into the experiences of families in the United States, where the policy and service environment may result in different levels of social and economic integration for persons with disabilities. In addition, studies often include autistic individuals from a broad age range, which overshadows unique experiences of the transition-age youth and their families. It is challenging to draw firm conclusions from this literature; in addition to being small in number, studies vary in the number and types of barriers measured, sample composition, and whether analyses were bivariate or multivariate. In addition to the paucity of empirical research exploring service barriers for autistic youth, the theoretical literature surrounding barriers to services is lacking. Existing conceptual frameworks of service use (e.g., Andersen, 1995) do not delineate what predicts barriers to service use or articulate their relationship to service use or unmet need.

Prior research suggests that characteristics such as race, youth age, and socioeconomic status (SES) are associated with barriers to service. For example, non-White families were more likely to report being unsure about where to find employment, health, and other services (Dudley et al., 2019). A Canadian study found barriers to service to be highest among autistic adolescents and young adults and lowest among preschool- and elementary-school-aged children (Lai & Weiss, 2017). Yet, a study in Poland found that families with younger children and those with lower income were most likely to identify cost as a barrier (Platos & Pisula, 2019). Families from lower SES backgrounds in the United States also report resource-related challenges, such as a lack of transportation and inconvenient service hours (Pickard & Ingersoll, 2016). Lower SES caregivers were also more likely to need information about services and insurance and to report barriers related to child age and symptom severity.

The literature regarding clinical characteristics related to service barriers is scarce. Among autistic adults, having a co-occurring medical problem was associated with more barriers to health care services (Vogan et al., 2017). Studies have also rarely examined the role of broader contextual variables, such as the characteristics of the areas in which families live. However, in one prior study, caregivers in large cities were less likely to report that autism services were unavailable or too far away than were those in medium or small cities (Platos & Pisula, 2019).

Despite the conceptual closeness of—and the seemingly logical connection between—experiencing barriers to service and having unmet service needs, few studies have explicitly tested this association among individuals with autism. There are two notable exceptions. First, Lai and Weiss (2017) found a positive relationship between the number of service barriers and the number of unmet priority service needs. However, this relationship was significant only among elementary-school-aged children with autism, not among older youth or adults. Second, a more recent Canadian study found that autistic adults who encountered barriers to accessing services had more unmet health and social service needs than did those who encountered no barriers (Jose et al., 2021).

As part of a larger, cross-sectional study of families providing care to autistic adolescents and young adults, the current investigation specifically sought to expand the knowledge base regarding service barriers encountered by autistic youth. The overall aim was to examine barriers experienced in accessing services for transition-age autistic youth and the degree to which those barriers are related to their unmet service needs. We asked three specific research questions, each addressing one or more gaps in the current literature:

Research Question 1: What service barriers do caregivers to adolescents and young adults with autism report, and can these barriers be grouped in an empirically meaningful way? To add to the limited quantitative data regarding the barriers encountered by autistic youth, we crafted a broad list of service barriers informed by a diverse group of prior studies (see Measures). In contrast to existing studies that have constructed only tallies of obstacles (e.g., Lai & Weiss, 2017; Vogan et al., 2017), we measured the extent to which each obstacle had been experienced as a barrier to getting or using services. To further extend the base of knowledge regarding barriers, we assessed the structure and performance of this new measure. Of specific interest was the extent to which responses to these barrier items would be related empirically, and therefore might suggest the existence of multiple underlying “types” of barriers. Given the exploratory nature of this inquiry, we made no a priori hypotheses regarding the existence of any interrelationships or the number of item groups.

Research Question 2: What caregiver, youth, and contextual characteristics are associated with barriers to service? To bolster knowledge from prior descriptive studies, we examined the relationships between service barriers and the demographic and background characteristics of autistic youth and their families. The specific characteristics considered, along with associated hypotheses, were derived largely from studies in broader age groups. Based on prior findings (Pickard & Ingersoll, 2016; Platos & Pisula, 2019), we hypothesized that families with lower income would report greater barriers to service. Given general autism literature showing that non-White children tend to be diagnosed later (e.g., Mandell et al., 2002) and be less connected to service systems (Magana et al., 2012), along with findings suggesting that they may experience specific obstacles (Dudley et al., 2019), we hypothesized that racial/ethnic minority families would report greater service barriers. Given findings of greater perceived availability of services in more densely populated areas (Platos & Pisula, 2019), we hypothesized that youth in highly urban areas would experience fewer service barriers. We also broadened consideration of caregiver characteristics to include perceived burdens associated with caregiving; we know of no prior studies that have quantitatively measured the associations between caregiver burden and service barriers.

Research Question 3: How are barriers to service related to unmet service needs of autistic youth? To fill the sizable gap in the existing knowledge base, we examined whether and how barriers to service were related to the unmet service needs of adolescents and young adults with autism. Given the paucity of existing research, this examination was exploratory and we made no a priori hypotheses regarding the size or strength of this relationship.

Method

Sample and procedure

This study included 174 family caregivers of autistic adolescents or young adults in the northern portion of the rust-belt state of Ohio. Eligible family members were required to (1) be the primary caregiver, (2) to an adolescent or young adult age 16–30, and (3) who had received an autism spectrum disorder diagnosis (e.g., autism, Asperger’s, PDD-NOS) by an education or health professional. A lower age of 16 was selected based on federal requirements for the inclusion of transition services in a student’s individual educational plan. A generous upper age of 30 was selected to encourage broad participation and in recognition that: (1) individuals with disabilities can attend school into their early 20s in many states and (2) lags in service access can result from delays in Medicaid eligibility evaluations after age 18 and after the loss of coverage under a parent’s health insurance plan at age 26.

Family members were recruited via ads posted in autism service agencies, hospitals, schools, and public libraries. Graduate-level research assistants conducted in-person, semistructured interviews at the university, a public library, or the caregiver’s home. Interviews were completed with 83.7% of caregivers who met the inclusion criteria. Interviews averaged 95 min and caregivers received a $25 gift card. Recruitment and data collection took place over 18 months, concluding in November, 2018. The study was approved by the university institutional review board.

Sample demographics are displayed in Table 1. Caregivers were primarily mothers (91.4%), with a mean age of 54 years. Most were married or partnered. Caregivers were fairly well-educated and most were employed. Over half had an annual household income of $75,000 or more, yet nearly one-fifth reported income below $40,000. Autistic youth were predominantly male, with an average age of 21. Over one-quarter had an intellectual disability (ID). In 37 families (21.3%), the caregiver and/or the youth were from a historically marginalized racial/ethnic minority group; most were African American, which reflects the demographics of the region. Additional caregiver, youth, and contextual characteristics are reviewed in the Results; more details about the sample and procedures can also be found in Ishler et al. (2022).

Table 1.

Sample characteristics (N = 174).

	n	%	M	SD	Range
Caregiver characteristics
Male	15	8.6
Age			54.22	6.75	35–72
Historically marginalized minority race/ethnicity^a	33	19.0
Married/partnered	133	76.4
Has 4-year college degree or higher	105	60.3
Currently employed	120	69.0
Annual HH income (1–7), M (SD)			4.66	1.93	1–7
1. <$20,000	11	6.3
2. $20,000–$39,999	22	12.6
3. $40,000–$59,999	21	12.1
4. $60,000–74,999	21	12.1
5. $75,000–$99,999	22	12.6
6. $100,000–$150,000	40	23.0
7. >$150,000	37	21.3
Caregiving burden
Daily disruptions (0–16)			7.10	3.89	0–16
Impacts (0–5)			2.70	1.38	0–5
Financial (3–15)			8.71	4.08	3–15
Worry (0–32)			20.13	5.90	2–32
Autistic youth characteristics
Male	125	71.8
Age			20.93	3.66	16–30
Historically marginalized minority race/ethnicity^a	36	20.7
Intellectual disability	47	27.0
Age at diagnosis			6.25	4.98	1–24
Still in high school	71	40.8
Adaptive functioning (0–200)			56.09	30.8	0–123
Contextual characteristics
Youth lives with caregiver	142	81.6
Youth covered by Medicare/Medicaid	100	57.5
Youth has a Medicaid waiver	59	33.9
Youth lives in highly urban county	120	69.0

Note. HH: household; SD: standard deviation.

Historically marginalized minority race/ethnicity was defined as Black/African American, Hispanic, Native American/Alaskan Native, or multiracial. The caregiver and/or youth in 37 families (21.3%) identified as one of these historically marginalized racial/ethnic groups.

Measures

Outcomes

Barriers

We developed a list of 15 barriers to services based on a review of prior studies (e.g., Lai & Weiss, 2017; NLTS2, 2002). Items were added from studies of health care service barriers (e.g., Kuhlthau et al., 2016). Sample items included cost, location, and scheduling conflicts (see Table 2 for the list). Consultation with community informants (family caregivers, autistic youth, service providers, agency leaders) led us to adapt several items to better reflect the experiences of local families. For example, feedback obtained during pilot testing prompted us to alter the idiomatic wording of “on waitlist” (Lai & Weiss, 2017) to the more commonly used “waiting list.” This item was further refined to “waiting list for services or waiting list for a waiver for services,” after learning that youth approved for a service might be unable to receive it because of a waiting list for Medicaid waiver slots. Caregivers were asked to indicate the degree to which each was experienced as a barrier to getting or using services during the prior six months. Response options were 0 (no barrier), 1 (minor barrier), 2 (moderate barrier), or 3 (major barrier). Because assessing the structure and performance of this measure was a study aim, psychometric details are presented in the Results.

Table 2.

Barriers to service: item descriptives and exploratory factor analysis results.

Item	Mean	SD	Factor Loadings^a
			Factor 1	Factor 2
			(“Access”)	(“Quality”)
Lack of information (about services or how to get them)	1.56	1.20	0.49	0.15
Services not available	1.32	1.32	0.47	0.24
Waiting list (for services or a waiver)	1.25	1.31	0.38	0.08
Location	1.11	1.15	0.68	0.06
Providers not trained	1.09	1.17	0.06	0.62
Poor service quality	1.05	1.21	0.01	0.88
Lack of time	1.03	1.05	0.58	0.06
Cost	1.03	1.22	0.73	0.18
Youth not eligible	1.01	1.25	0.34	0.23
Scheduling conflicts	0.91	1.08	0.57	0.02
(Lack of) transportation	0.81	1.14	0.41	0.13
Provider does not accept insurance	0.73	1.11	0.56	0.01
Language/communication^b	0.45	0.90
Physical accessibility^c	0.23	0.72
Other^c	0.20	0.73

Note. N = 168. Caregivers rated the extent to which each was a barrier during the past 6 months (0 = No barrier, 1 = Minor, 2 = Moderate, 3 = Major); Extraction method = Principal Axis Factoring; Rotation method = Oblimin with Kaiser Normalization; SD: standard deviation.

Pattern matrix loadings > 0.34 in boldface type.

In preliminary analyses, this item loaded on Factor 2; however, it was excluded from the final analysis because it compromised the internal consistency of the factor-derived scale.

Items excluded from the final factor analysis because of low endorsement and low communalities.

Unmet service needs

Caregivers were asked about services received by the youth in the past 6 months. A list of 15 services was developed from prior research (NLTS2, 2002; Taylor & Henninger, 2015): Behavioral interventions, case management, day programming, educational, employment/vocational training or supports, life skills education and training, medical, mental health, personal care, residential supports, social skills training/supports, speech/language therapy, supportive/complementary therapies, transition planning, and transportation. Caregivers indicated if the service had been received (1), was needed but not received (2), or was not received but not needed (3). Unmet service need was measured as the total number of services needed but not received (i.e., those rated 2) out of 15.

Predictors

Caregiver characteristics

Demographic and background characteristics

Caregivers reported their gender, age, and marital status. Those who identified as Black/African American, Hispanic, Native American/Alaskan Native, or multi-racial were coded as belonging to a historically marginalized racial/ethnic minority group. Caregivers currently working full- or part-time were coded as employed. Caregivers’ highest degree was coded into two categories reflecting completion of a 4-year college degree or higher. Caregivers selected one of 12 categories that best captured their annual household income; values were then collapsed into seven categories ranging from less than $5000 to $150,000 or more. Values for four cases were imputed using stochastic regression.

Caregiving burden

Caregivers completed four scales adapted from the Family Experiences Interview Schedule (Tessler & Gamache, 1995) to assess the perceived burden of caring for a youth on the autism spectrum: disruptions to daily life, significant life impacts, financial difficulties, and worry.

The daily disruptions scale included four items inquiring how often caregivers experienced disruptions to their daily activities (e.g., school/work/family obligations) and household routines in the past six months due to the youth autism. Item response options ranged from never (0) to constantly or almost constantly (4). Scores could range from 0 to 16, with higher scores indicating greater disruption. Internal consistency reliability in this sample was good (α = 0.84).

The impacts scale included five yes/no questions asking caregivers whether caring for the youth had resulted in any significant life changes (e.g., quitting a job, reducing social life). Scores could range from 0 to 5, with higher scores indicating greater life impacts. Scale reliability was acceptable (α = 0.70).

The financial burden scale consisted of three questions assessing the impact of the youth condition on the family’s financial resources (e.g., “As a result of (youth) needs or condition, the money available for ‘extras’ has been reduced”). Response options ranged from 1 (not at all) to 5 (very much). Scores could range from 3 to 15, with higher scores indicating a greater financial burden. Internal consistency reliability was high (α = 0.92).

The worry scale included eight items asking caregivers to rate the extent to which, in the past 6 months, they worried about the youth safety, romantic/sexual relationships, social life, help and treatment, physical health, living arrangements, finances, and future well-being. Responses ranged from never (0) to constantly or almost constantly (4). Scores could range from 0 to 32, with higher scores indicating greater worry. Scale reliability in this sample was acceptable (α = 0.76).

Youth characteristics

Demographic and background characteristics

Youth gender, age, and racial/ethnic identity were coded from caregiver reports. Given the substantial overlap between the racial/ethnic identification of caregivers and autistic youth, a variable was created to indicate if either was a member of a historically marginalized minority racial/ethnic group. Caregivers reported the age at which the youth first received an autism diagnosis. They also indicated if the youth was still attending high school. Finally, caregivers were asked if the youth had ever been diagnosed with ID.

Adaptive functioning

Caregivers completed the Adaptive Behavior subscale of the Scales of Independent Behavior-Revised Short Form (SIB-R SF; Bruininks et al., 1996), consisting of 40 items asking about youth ability to complete certain personal, community living, communication, motor, and social interaction tasks without help or supervision. Items were rated on a four-point scale ranging from 0 (never or rarely, even if asked) to 3 (does very well or always or almost always, without being asked). The scale yields an age-based standard score, ranging from 0 to 200 (M = 100, SD = 15). Higher scores indicate greater functional ability. The SIB-R SF has good psychometric properties including internal consistency, interrater, and test–retest reliability (Bruininks et al., 1996).

Contextual characteristics

Youth were coded as residing with the caregiver or elsewhere (e.g., with another relative, in a group home). Caregivers reported whether the youth was currently covered by Medicaid or Medicare, and whether they had a Medicaid waiver. Youth county of residence was coded as highly urbanized or not (U.S. Office of Management and Budget, 2010).

Community involvement

Local family caregivers, autistic youth, service providers, and agency leaders served as community informants, contributing to multiple aspects of the study. We consulted with the founding co-directors of a local autism advocacy organization (each of whom parents an autistic young adult) in the overall design of the study. We solicited input from family members and service providers throughout the development of the interview protocol. We pilot-tested the interview with eight family caregivers whose feedback led us to revise several measures and procedures (e.g., see “Barriers” above). Preliminary findings were presented to an audience of roughly 100 family members, autistic individuals, and providers at the annual meeting of the local advocacy organization. Finally, all family caregivers who participated in the study were invited to attend a “feedback session” designed to provide feedback on, and assist in the interpretation of, study findings; 19 caregivers (10.9%) attended one of these sessions.

Data analysis

Descriptive data for the 15 service barrier items were used to examine Research Question 1. Because this was the first use of this measure, we conducted an exploratory factor analysis (EFA). We used principal axis factoring with oblique rotation to allow extracted factors to be correlated. The EFA results are detailed in the Results. Briefly, the analysis suggested a two-factor solution: Barriers related to access (e.g., location, cost) and quality (e.g., poor quality services, providers not trained). Subsequent analyses used these two scales. Six caregivers were missing one to two barrier items. The EFA was based on complete cases (N = 168), but scale scores were created for the entire sample by calculating an average item score across all available items (0 = no barrier to 3 = major barrier).

Research Questions 2 and 3 were addressed using ordinary least squares (OLS) regression. For Research Question 2, each barrier scale was regressed on a set of predictors including caregiver and youth characteristics and contextual factors. Predictors were the same for both models and were selected on the basis of prior empirical work and the presence of a significant bivariate correlation with one or both barrier scales. Youth age and high-school enrollment were collinear. Neither variable had a significant bivariate relationship with the barrier scales. High-school enrollment was used in regression models because of conceptual interest surrounding the transition from high school (e.g., Shattuck et al., 2011) and prior empirical evidence that high-school enrollment is related to service use (e.g., Ishler et al., 2022). We also conducted sensitivity analyses, substituting age for high school enrollment in models predicting barriers. Research Question 3 was addressed by simultaneously regressing the total number of unmet needs on the two barrier scales. Regression coefficients were used to assess the contribution of individual predictors and squared semipart correlation coefficients (sr²) were used to indicate effect size. Diagnostics revealed no serious violations of key assumptions, including linearity, (absence of) influential outliers, (lack of) multicollinearity, and homoscedasticity and normality of residuals.

Results

Additional descriptive data

Descriptive data for other substantive caregivers, youth, and contextual characteristics are also shown in Table 1. On average, caregivers reported moderate levels of caregiver burden. The highest levels of burden were observed on the worry scale. Although most youth were diagnosed with autism at an early age (M = 6.25), nearly 15% were diagnosed after age 12. Youth adaptive functioning averaged 56.09 (SD = 30.80) on the 200-point SIB-R SF, well below the age-based mean of 100 (Bruininks et al., 1996). Most youth received some government support: Half received SSI or SSDI, and more than half were covered by Medicaid or Medicare. One-third had a Medicaid waiver. Most youth lived with the caregiver and nearly 70% lived in a highly urbanized county.

Barriers

Descriptive data for the 15 barrier items, along with the EFA results, are presented in Table 2. Lack of information about services or how to get them was the highest-rated barrier, with an average rating between minor and moderate. Other barriers rated between minor and moderate included services not being available, a waiting list for services (or for a waiver), and the location of services. Two items (physical accessibility and other) had low rates of endorsement and low communalities and were thus excluded from further analysis. The language/communication barrier item had a good loading on Factor 2 in the initial EFA (0.37); however, it detracted significantly from the internal consistency of the derived scale and was excluded from further analysis. The remaining 12 items were retained for the final EFA. The initial eigenvalues and scree plot suggested that two factors should be extracted. Four factors had initial eigenvalues greater than 1 (4.232, 1.274, 1.046, and 1.011); however, only the first two explained sizable proportions of variance in the data (35.3% and 10.6%, respectively). Table 2 shows the factor loadings (pattern matrix) from the oblique rotation of these two extracted factors. Factor 1 (labeled Access Barriers) included 10 items, with the cost of services having the highest loading. Factor 2 (labeled Quality Barriers) included two items, with poor service quality having the highest loading.

Because the EFA revealed an interfactor correlation in the moderate-to-strong range (r = 0.48), we also ran an EFA constrained to a single factor. All 12 items loaded acceptably (>0.40) on one factor. However, the two-factor solution also yielded good factor loadings and was regarded as more optimal because it: (1) explained more cumulative variance after extraction (36.9% vs 29.5%); (2) generated postextraction item communalities that were generally higher and closer to the initial values; (3) resulted in fewer sizable residuals (>0.05) when the solution-reproduced correlations were compared to the original inter-item correlations (33% vs 46%); and (4) yielded higher mean interitem correlations in the two factor-derived subscales (r = 0.31 for access, r = 0.58 for quality) than in the total scale (r = 0.29).

The average score on access barriers was 1.08 (SD = 0.72). The average score on quality barriers was 1.06 (SD = 1.05). Although these averages correspond to barrier ratings of “minor,” substantial variability was observed on both scales, as indicated by sizable standard deviations and observed scores spread across the range of possible scores. Both scales also demonstrated acceptable estimates of internal consistency (alpha = 0.82 for access and 0.73 for quality).

Table 3 presents the results of the OLS regression model for each barrier scale. The overall model for access barriers was statistically significant, with predictors explaining roughly 24% of the variability in scores. Compared to females, male caregivers reported significantly less access barriers. Age at autism diagnosis was significantly related to access barriers, with each 5-year increment associated with an elevation in average item score of 0.10. Neither minority race/ethnicity nor household income was related to access barriers. Daily disruptions, financial difficulties, and worries about the autistic youth current and future well-being were each independently associated with greater access barriers. No significant relationships were found for other caregiver or youth characteristics. No contextual factors were significantly related to access barriers; most notably, no support was found for a relationship between access barriers and residing in a highly urban area. Overall, effect sizes for significant predictors were fairly small (sr² < 0.05).

Table 3.

Access and quality barriers regressed on caregiver and youth characteristics and contextual factors.

	Access barriers					Quality barriers
Predictors	b	SE	Beta	p	sr²	b	SE	Beta	p	sr²
(constant)	0.09	0.34		0.782		0.69	0.53		0.195
CG and youth characteristics
CG is male	−0.40	0.17	−0.16	0.023	0.02	−0.41	0.27	−0.11	0.134	0.01
CG has 4-year college degree or higher	−0.05	0.11	−0.04	0.634	0.00	−0.02	0.18	−0.01	0.929	0.00
HH income	0.01	0.03	0.03	0.708	0.00	0.00	0.05	0.01	0.949	0.00
Minority race/ethnicity^a	0.07	0.14	0.04	0.621	0.00	−0.26	0.22	−0.10	0.233	0.01
Youth still in high school	0.04	0.11	0.03	0.738	0.00	0.02	0.17	0.01	0.928	0.00
Youth has ID	−0.11	0.12	−0.07	0.331	0.00	0.08	0.18	0.03	0.655	0.00
Youth age at ASD diagnosis	0.02	0.01	0.15	0.042	0.02	0.05	0.02	0.24	0.003	0.05
Youth adaptive functioning	0.00	0.00	−0.04	0.613	0.00	−0.01	0.00	−0.15	0.104	0.01
CG burden: daily disruptions	0.05	0.02	0.25	0.011	0.03	0.02	0.03	0.09	0.408	0.00
CG burden: impacts	−0.06	0.04	−0.12	0.149	0.01	0.04	0.07	0.05	0.548	0.00
CG burden: finance	0.04	0.01	0.23	0.005	0.04	0.00	0.02	0.02	0.832	0.00
CG burden: worry	0.02	0.01	0.19	0.024	0.02	0.02	0.02	0.13	0.145	0.01
Contextual factors
Youth lives with CG	−0.10	0.13	−0.05	0.458	0.00	−0.52	0.20	−0.19	0.012	0.03
Youth has Medicaid waiver	−0.03	0.13	−0.02	0.831	0.00	0.27	0.20	0.12	0.183	0.01
Youth lives in highly urban county	0.02	0.11	0.01	0.881	0.00	−0.08	0.17	−0.03	0.654	0.00
Model adjusted R²	0.24					0.15

Note. CG = caregiver; HH = household; ID = intellectual disability; sr² = squared part correlation (effect size). Separate OLS regression models run for the average item score on each barrier scale (0 = No barrier to 3 = Major barrier). Statistically significant results (p < 0.05) in boldface type.

Coded if either CG or youth identified as a member of a historically marginalized minority racial/ethnic group (Black/African American, Hispanic, Native American/Alaskan Native, or multiracial).

For quality barriers, the overall model was statistically significant and predictors explained roughly 15% of the variability in scores. No caregiver characteristics were significantly associated with quality barriers, and there was no evidence to support the hypothesized associations involving minority race/ethnicity or household income. None of the caregiver burden scales was associated with quality barriers. Age at autism diagnosis was positively related to quality barriers; each 5-year increment was associated with an average quality barrier item score 0.25 points higher. No support was found for a relationship between the urbanicity of residence and quality barriers. However, caregivers living with the autistic youth reported significantly fewer quality barriers than did those not living with the youth. Effect sizes for significant predictors were fairly small (sr² < 0.06).

High-school enrollment was not significantly related to either access or quality barriers. When age was substituted for high-school enrollment in the regression models, age was not a significant predictor either. The results for other predictors were also unchanged.

Unmet service need

Autistic youth received an average of 6.1 (SD = 3.2) out of 15 services in the past 6 months. All but three youth had received at least one service, with the majority (78.2%) receiving four or more. Despite relatively high service use, caregivers identified an average of 3.2 (SD = 2.6) services that were not received, but needed. The number of unmet service needs ranged from 0 to 11. Most common needed services included social skills training/supports (43.1%), supportive/complementary therapies (35.6%), transition planning (33.3%), life skills education and training (33.3%), and employment supports/vocational training (26.4%).

Table 4 presents the results of the OLS regression model in which the total number of unmet service needs was regressed on both barrier scales. Although the overall model was statistically significant, only the access barriers scale was significantly related to unmet needs. Each one-point increment on the access barriers scale (e.g., moving from a rating of “mild” to “moderate”) was associated with roughly one additional unmet service need (b = 1.2). Access barriers explained 10% of the variability in unmet needs.

Table 4.

Number of unmet service needs regressed on service barriers.

Predictors	b	SE	Beta	p	sr²
(constant)	1.93	0.34		<0.001
Access barriers	1.22	0.29	0.34	<0.001	0.10
Quality barriers	−0.06	0.19	−0.03	0.740	0.00
Model adjusted R²	0.10

Note. SE: standard error. sr² = squared part correlation (effect size). OLS regression used. Statistically significant results (p < 0.05) in boldface type.

Discussion

This study examined caregiver-reported barriers to service experienced by autistic adolescents and young adults and the degree to which these barriers were related to their unmet service needs. With respect to Research Question 1, our analyses suggested the presence of two related, but distinct, types of barriers: access and quality. That is, caregivers tended to think about the obstacles encountered in gaining access to services (e.g., cost, location, availability of services), as well as the challenges posed by the quality of the services being provided (e.g., provider training, service quality). Both barrier scales displayed substantial variability and acceptable internal consistency in this sample. We are not aware of any prior studies that have examined the factor structure of a measure of barriers to service in autism.

Service barriers are most often measured using a checklist of yes/no items, and previous studies have varied widely in the number of items used (e.g., 5 in Platos & Pisula, 2019; 13 in Vogan et al., 2017). Thus, the extent to which we can compare the number of barriers endorsed by caregivers in this study to those reported in prior research is limited. Lack of information about services or how to get them was the highest-rated barrier in this sample. Difficulty obtaining information about services has been identified as a barrier to: use of health and social services among families with school-age autistic children (Pickard & Ingersoll, 2016); use of adult-transition services by autistic youth (Havlicek et al., 2016); and health care service use among autistic adults (Vogan et al., 2017). Other higher-rated barriers in this study included services not being available, a waiting list for services (or for a waiver), and the location of services. Barriers related to service availability and location are included in most existing barrier measures (e.g., NLTS2, 2002; Platos & Pisula, 2019) and have been commonly reported among autistic youth and adults (e.g., Dudley et al., 2019; Jose et al., 2021; Taylor & Henninger, 2015). Although waiting lists are often reported in qualitative interviews with caregivers and providers of services to autistic youth (e.g., Anderson et al., 2018; Pickard & Ingersoll, 2016), only Lai and Weiss (2017) included a “waitlist” item in their measure. Similarly, just a few prior studies have included lack of training among service providers (e.g., Kuhlthau et al., 2016) or concerns about service quality (e.g., NLTS2, 2002; Taylor & Henninger, 2015).

In examining the caregiver, youth, and contextual characteristics related to service barriers (Research Question 2), we found no evidence in support of the hypothesized relationships between service barriers and race/ethnicity or household income. The relatively small numbers of minority racial/ethnic families and those in the lower income strata may have hampered our ability to detect small, but potentially meaningful differences. However, it is important to consider the constellation of variables included in prior tests of these relationships. Of three studies that have examined family SES in relation to service barriers, only one simultaneously controlled for race/ethnicity: Dudley et al. (2019) found that SES (measured as parent education) was unrelated to barriers, but non-White families were more likely to report being unsure about where to find autism services. Despite using disparate measures of SES (parent education in Pickard & Ingersoll, 2016; household income in Platos & Pisula, 2019), SES was inversely related to service barriers in the other two studies. Based on prior studies, we included household income, historically marginalized minority racial/ethnic identity, and parent education in our models. None of these emerged as a significant predictor of either access or quality barriers and no collinearity was observed among predictors. Notably, none of these predictors had a significant bivariate relationship with either type of barrier.

Only one variable was related to both barrier types: Increased age at autism diagnosis was associated with greater access and greater quality barriers. No prior barrier studies have included age at autism diagnosis as a predictor. The current findings suggest that delays in diagnosis may result in more tenuous connections to services. Moreover, these delays may make it more difficult for families to secure quality services.

It is noteworthy that neither enrollment in high school nor age was related to either access or quality barriers. Although there is ample evidence of the negative impact of high-school exit on service use in autism (e.g., Laxman et al., 2019; Shattuck et al., 2011), we know of no prior research examining high-school enrollment as a predictor of caregiver- or consumer-reported barriers to service. Three prior studies have examined age (Lai & Weiss, 2017; Platos & Pisula, 2019; Vogan et al., 2017). Samples in those studies had very broad age ranges. Both Platos and Pisula (2019) and Lai and Weiss (2017) found a relationship between age and barriers. However, differences were evident only when autistic adolescents/young adults were compared to younger age groups, and the direction of differences was inconsistent. In Platos and Pisula (2019), those in the youngest group (age 12–14) were more likely than adolescents/young adults (age 18–24) to report one barrier: services being too costly. Yet, Lai and Weiss (2017) found that high school, young adult, and adult groups reported more barriers than did those in preschool and elementary school. It is possible that the age range in the current sample accounts for the lack of a relationship between age (or high-school enrollment) and barriers.

The most novel study finding concerns the relationship between caregiver burden and access barriers. Caregiver-reported disruptions to daily life, financial difficulties, and worry about the autistic youth current and future well-being were each independently associated with greater access barriers. Since caregivers completed both the barriers and burden measures, it is possible that the observed relationship reflects some underlying shared methods variance. However, given the variability in the relationships between access barriers and measures of burden (e.g., impact was unrelated to access) and the lack of any significant relationships between burden and quality barriers, shared method variance is unlikely to be the main source of the observed associations. To our knowledge, no prior studies have quantified the association between caregiver burden and barriers to service. Although they did not measure barriers directly, one study of caregivers to transition-age autistic youth in the U.K. found a sizable correlation between the number of unmet service needs and caregiver burden (Cadman et al., 2012). Although the directionality of the relationships between the strains of caregiving and barriers to service access is unclear, our findings suggest that caregiver concerns and perceptions of burden are important to consider when examining service barriers.

Male caregivers in this study reported fewer access barriers than did their female counterparts. Caution is warranted in attempting to generalize this finding because fathers comprised just 9% of the sample. Representation of male caregivers has varied widely in prior studies that have examined barriers to service for autistic youth—for example, 3% in Pickard and Ingersoll (2016), 12% in Platos and Pisula (2019), and 28% in Dudley et al. (2019). None of these studies reported differences in barriers by caregiver gender. Studies of caregivers to aging and/or frail adults have found that females tend to provide more hands-on, personal care (Pinquart & Sorensen, 2006). If fathers in the current study were more involved in the practical logistics of arranging care for the autistic youth, they might report fewer access-related barriers. Although differing caregiver responsibilities may well influence perceived barriers to service, it is also possible that more complicated relationships exist—for example, interactions between caregiver gender and the gender and needs of the autistic youth. The present finding underscores the need to further explore the multitude of factors that may influence the perception and experience of barriers to service, as well as the importance of ensuring greater representation of caregiving fathers in research.

Contrary to our hypothesis, the urbanicity of the youth residence was unrelated to service barriers. Platos and Pisula (2019) observed such a relationship in Poland. Further testing of this hypothesis is warranted in U.S. samples. Although urbanicity was unrelated to service barriers, we uncovered an association involving another contextual factor–living arrangement. Caregivers who were living with the autistic youth reported fewer quality barriers. Compared to those whose youth reside in other settings, co-resident caregivers may be more involved in selecting service providers and/or better able to supervise service delivery in the home. Dudley et al. (2019) found that autistic adults living with a family experienced more service barriers than did those in supported-living facilities; however, barriers in that study (services being unavailable or too far away, feeling unsure about where to find services) were more closely aligned with our measure of access barriers. Nonetheless, such findings imply that service barriers may differ depending on the living arrangement of the autistic youth. As more youth on the autism spectrum transition to adulthood, the impact of living arrangement on barriers to service requires increased attention.

With respect to Research Question 3: While access barriers were significantly related to unmet need, quality barriers were not. The relationship between unmet needs and access barriers is generally consistent with the findings of Jose et al. (2021) and Lai and Weiss (2017). However, the lack of a relationship between unmet needs and quality barriers invites deeper contemplation. One potential explanation is that different types of barriers may be experienced at different points in the service use trajectory. For example, access barriers may be important in determining whether or not services are initially received, while quality barriers may be more related to decisions regarding continuation of (or, conversely, attrition from) services and to satisfaction with received services.

Strengths and limitations

This study had several notable strengths, including the use of a service barrier measure that is more comprehensive than those used in most prior studies (e.g., Lai & Weiss, 2017; Platos & Pisula, 2019). For example, our measure included more structural or systemic barriers. In addition, we employed an ordinal response scale, allowing for greater precision in assessing barriers. This is potentially useful because it can help to distinguish obstacles that some might regard as “annoyances” from those with a more notable impact on service receipt. Family caregivers and service providers contributed to the development, testing, and revision of the measure. Our analyses suggested the presence of two related, but distinct, types of barriers—those related to access and those related to quality.

Our study also examined the relationships between barriers and selected caregiver, youth, and contextual characteristics, and uncovered some unique predictors of access and quality barriers. Most notably, caregiver burden—which has not been included in prior research—was associated with access, but not quality, barriers. Finally, this study adds to the very limited body of research that has examined barriers in relation to unmet needs (e.g., Jose et al., 2021; Lai & Weiss, 2017). Our findings suggest that unmet needs may be more influenced by certain types of barriers (access), and not others (quality).

Study findings should be considered in light of several weaknesses. First, the barriers measure requires enhancement and additional validation. The current two-item quality barriers scale is weak and should be bolstered with additional items. One potential item could be a lack of evidence-based services, such as transition planning interventions (e.g., Ruble et al., 2018). Others to consider are negative interactions with (e.g., Lai & Weiss, 2017) and lack of trust in (e.g., Vogan et al., 2017) service providers. The language/communication barrier item may have greater relevance in samples that include non-native English speakers; alternatively, the item could be revised to more clearly reflect the communication challenges faced by autistic persons when attempting to use health and social services (e.g., Nicolaidis et al., 2015). Larger samples are needed to more fully examine the dimensionality of the barriers measure and to better clarify the empirical and conceptual nature of the barriers to service construct. There is important methodological, empirical, and conceptual work that still needs to be done regarding the barriers to service construct.

Second, our sample had somewhat limited diversity in regard to race/ethnicity and SES. Although power analyses suggested that our overall sample size, the proportion of families of minority race/ethnicity, and the proportion of caregivers with less than a bachelor’s degree were adequate to detect moderately-sized differences in barriers, the relatively small numbers of racial/ethnic minority families and those from lower income strata would have rendered it difficult to detect smaller-sized differences. Despite concentrated recruitment efforts, we had lower numbers of racial/ethnic minority families and families with lower income in our study than we had hoped. Given their relatively small numbers in our sample, our findings may not fully capture the experiences of racial/ethnic minority families and/or those with lower SES. In addition, we recognize that our characterization of certain racial and ethnic identities as “historically marginalized” is both geographically and historically limited.

Third, although we did not recruit families based on service use per se, most youth were receiving some services. Thus, it may be useful to specifically examine access and quality barriers in families whose youth are not receiving services. Fourth, given scarce prior research, we did not include any covariates in analyses relating barriers to unmet service needs. For example, prior literature suggests that race/ethnicity, SES, ID, age, and/or high-school enrollment are related to service use and/or unmet needs (e.g., Ishler et al., 2022; Lai & Weiss, 2017; Taylor & Henninger, 2015). Finally, the cross-sectional nature of the study limits our ability to draw causal connections among variables.

Practice and policy implications

Two specific findings carry important implications for practitioners and policy-makers who wish to support autistic youth and their caregivers. First, the associations between later age at autism diagnosis and greater access and quality barriers underscore the value of early diagnosis and timely connections to services. The American Academy of Pediatrics recommends screening for autism at 18- and 24-month well-child visits, with simultaneous referrals to an evaluation specialist and early intervention services for children with a positive screen (Johnson & Myers, 2007). Yet, there are marked differences in screening and referral practices. For example, family medicine physicians are much less likely than pediatricians to use autism screening tools (Carbone et al., 2020). Although delays in diagnosis are often attributed to a lack of screening, even in systems with high screening rates, referrals for diagnostic evaluation can be delayed or nonexistent (Monteiro et al., 2019). Racial/ethnic disparities have also been documented: Hispanic children are less likely than white children to be screened (Carbone et al., 2020); and Hispanic and African American children are diagnosed with autism, on average, more than a year later than white children (Mandell et al., 2002). Practitioners in primary care and early education settings may need periodic training in screening, referral, and follow-up practices. Routine monitoring may help to ensure that at-risk youngsters receive timely diagnostic evaluations and to accelerate the connection of diagnosed children and their families to services. It is important to acknowledge that delays in diagnosis could also reflect individual differences in symptom presentation and severity, family factors such as health literacy and fear of stigma, and structural constraints related to service funding and availability. Consequently, efforts to reduce delays in diagnosis are more likely to be successful if they attend to multiple levels of influence and are supported by relevant practices and policies.

Second, regardless of the causal direction between caregiver perceptions of burden and barriers to service, additional supports for families providing care to youth with autism are warranted. Although caregivers express the need for guidance when attempting to secure services for autistic youth (e.g., Hodgetts et al., 2013), it may be especially critical to support those experiencing higher levels of caregiving burden. Autism “navigators” have been observed to be helpful in linking newly diagnosed children and their families to services (Roth et al., 2016). Extending navigation services to transition-age autistic youth could help to reduce access barriers. Trained service navigators could provide support to all caregivers, and provide more intensive support to those experiencing greater disruptions and strains. In addition, services can be costly without government assistance and this may contribute to financial burden and strain on families. Policy initiatives that include expanding access to Medicaid, increasing the number of autistic individuals eligible for Medicaid waivers, and providing additional funding to help cover out-of-pocket expenses could reduce potential service barriers, as well as bolster the well-being and capacity of family caregivers.

Directions for future research

The current study provides valuable information about the service barriers faced by transition-age autistic youth. Our finding suggesting two distinct types of service barriers—access and quality—requires replication in other samples. It is noteworthy that several characteristics that we expected to be related to service barriers on the basis of prior research (e.g., SES, race/ethnicity, urbanicity) were not significantly associated with either access or quality barriers. Although the lack of findings could be at least partially attributed to differences in study samples and measures, it is important that future research includes larger samples with more diverse SES and a greater representation of racial/ethnic minority families. Also, samples drawn from diverse geographical locations may help to elucidate the relationship between urbanicity and barriers to services. Additional research is needed to confirm several novel findings, especially those related to age at diagnosis and caregiver burden.

It is surprising that so few studies have examined barriers in relation to the unmet service needs of autistic youth. One provocative study finding—that only access (and not quality) barriers appear to be related to unmet need—requires replication. Future inquiries should assess the role of barriers in predicting unmet service needs, over and above the role of demographic and background characteristics. Eilenberg et al. (2019) recently concluded that, in light of substantial evidence of disparities in service receipt and transition outcomes among racial/ethnic and low-income autistic youth, researchers should work to delineate the potential mechanisms that contribute to such disparities. Thus, future studies should examine service barriers as possible mediators (or moderators) of the relationships between race/ethnicity, SES, and other background characteristics and service use and unmet need. Greater conceptual and operational clarity is also needed to distinguish service use, unmet needs, and barriers, as these terms are often conflated in the literature.

However, additional efforts to identify specific relationships and patterns will be of limited value without a more complete understanding of the pathways and trajectories of service use among autistic youth. It is imperative that future research be guided by theoretical and conceptual models that specify how barriers impact service use (and unmet service needs), as well as where and when these effects occur. Existing studies of autism service use are largely atheoretical. Several recent studies have drawn on Andersen’s (1995) behavioral model of health care utilization (e.g., Ishler et al., 2022; Platos & Pisula, 2019); however, this model has three major shortcomings when applied to the study of barriers. First, it is unclear whether barriers should be regarded as impeding factors that are conceptually parallel to “enabling” factors. Second, it is unclear whether unmet needs and service utilization are two sides of the same conceptual coin. Third, the model does not readily accommodate the dynamic and temporal nature of service use–especially among individuals with lifelong service needs. An enhanced understanding of the trajectories of service use by autistic youth will require not only longitudinal research, but also carefully crafted studies that utilize both qualitative and quantitative approaches. Future studies should aim to pinpoint where and when barriers to service are encountered, with the goal of informing policy changes and developing more targeted interventions to support families.

Footnotes

Acknowledgements

The authors are grateful to all of the families who participated in this study and to those who assisted in the review and interpretation of findings. Special thanks to Milestones Autism Resources for help with study design, recruitment, and dissemination. They also wish to thank the editor and reviewers for their thoughtful comments and suggestions.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by a Mt. Sinai Catalytic Autism Research Award through the International Center for Autism Research & Education, Case Western Reserve University and a Research Development & Training Grant Award from the Jack, Joseph and Morton Mandel School of Applied Social Sciences, Case Western Reserve University (PI: D.E.B.).

ORCID iDs

Karen J Ishler

Kristen A Berg

References

Al-Beltagi

(2021). Autism medical comorbidities. World Journal of Clinical Pediatrics, 10(3), 15–28. https://doi.org/10.5409/wjcp.v10.i3.15

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed., text rev., DSM-V-TR). American Psychiatric Publishing.

Andersen

R. M.

(1995). Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36(1), 1–10.

Anderson

Lupfer

Shattuck

P. T.

(2018). Barriers to receipt of services for young adults with autism. Pediatrics, 141(Suppl. 4), S300–S305. https://doi.org/10.1542/peds.2016-4300G

Bennett

A. E.

Miller

J. S.

Stollon

Prasad

Blum

N. J.

(2018). Autism spectrum disorder and transition-aged youth. Current Psychiatry Reports, 20(11), 1–9. https://doi.org/10.1007/s11920-018-0967-y

Bruininks

R. H.

Woodcock

R. W.

Weatherman

R. F.

Hill

B. K.

(1996). Scales of Independent Behavior-Revised comprehensive manual. Riverside Publishing.

Cadman

Eklund

Howley

Hayward

Clarke

Findon

Xenitidis

Murphy

Asherson

Glaser

(2012). Caregiver burden as people with autism spectrum disorder and attention-deficit/hyperactivity disorder transition into adolescence and adulthood in the United Kingdom. Journal of the American Academy of Child & Adolescent Psychiatry, 51(9), 879–888. https://doi.org/10.1016/j.jaac.2012.06.017

Carbone

P. S.

Campbell

Wilkes

Stoddard

G. J.

Huynh

Young

P. C.

Gabrielsen

T. P.

(2020). Primary care autism screening and later autism diagnosis. Pediatrics, 146(2), Article e20192314. https://doi.org/10.1542/peds.2019-2314

Coleman-Fountain

Buckley

Beresford

(2020). Improving mental health in autistic young adults: A qualitative study exploring help-seeking barriers in UK primary care. British Journal of General Practice, 70(694), e356–e363. https://doi.org/10.3399/bjgp20X709421

10.

Dietz

P. M.

Rose

C. E.

McArthur

Maenner

(2020). National and state estimates of adults with autism spectrum disorder. Journal of Autism and Developmental Disorders, 50(12), 4258–4266. https://doi.org/10.1007/s10803-020-04494-4

11.

Doherty

A. J.

Atherton

Boland

Hastings

Hives

Hood

. . .Chauhan

(2020). Barriers and facilitators to primary health care for people with intellectual disabilities and/or autism: An integrative review. BJGP Open, 4, Article bjgpopen20X101030. https://doi.org/10.3399/bjgpopen20X101030

12.

Dudley

K. M.

Klinger

M. R.

Meyer

Powell

Klinger

L. G.

(2019). Understanding service usage and needs for adults with ASD: The importance of living situation. Journal of Autism and Developmental Disorders, 49(2), 556–568. https://doi.org/10.1007/s10803-018-3729-0

13.

Eilenberg

J. S.

Paff

Harrison

A. J.

Long

K. A.

(2019). Disparities based on race, ethnicity, and socioeconomic status over the transition to adulthood among adolescents and young adults on the autism spectrum: A systematic review. Current Psychiatry Reports, 21(5), 1–16. https://doi.org/10.1007/s11920-019-1016-1

14.

Havlicek

Bilaver

Beldon

(2016). Barriers and facilitators of the transition to adulthood for foster youth with autism spectrum disorder: Perspectives of service providers in Illinois. Children and Youth Services Review, 60, 119–128. https://doi.org/10.1016/j.childyouth.2015.11.025

15.

Hodgetts

Nicholas

Zwaigenbaum

McConnell

(2013). Parents’ and professionals’ perceptions of family-centered care for children with autism spectrum disorder across service sectors. Social Science & Medicine, 96, 138–146. https://doi.org/10.1177/1362361314543531

16.

Ishler

K. J.

Biegel

D. E.

Wang

Olgac

Lytle

Miner

Edguer

Kaplan

(2022). Service use among transition-age youth with autism spectrum disorder. Journal of Autism and Developmental Disorders, 52, 1051–1065. https://doi.org/10.1007/s10803-021-04999-6

17.

Johnson

C. P.

Myers

S. M.

(2007). American Academy of Pediatrics Council on Children with Disabilities. Identification and evaluation of children with autism spectrum disorders. Pediatrics, 120(5), 1183–1215. https://doi.org/10.1542/peds.2007-2361

18.

Jose

George-Zwicker

Bouma

Tardif

Pugsley

Belanger

Robichaud

Gaudet

(2021). The associations between clinical, social, financial factors and unmet needs of autistic adults: Results from an observational study. Autism in Adulthood, 3(3), 266–274. https://doi.org/10.1089/aut.2020.0027

19.

Kersten

Coxon

Lee

Wilson

N. J.

(2020). “In their own time”: Parents gently push their autistic youth towards independent community mobility and participation. Journal of Autism and Developmental Disorders, 50(8), 2806–2818. https://doi.org/10.1007/s10803-020-04384-9

20.

Kogan

M. D.

Vladutiu

C. J.

Schieve

L. A.

Ghandour

R. M.

Blumberg

S. J.

Zablotsky

Perrin

J. M.

Shattuck

Kuhlthau

K. A.

Harwood

R. L.

M. C.

(2018). The prevalence of parent-reported autism spectrum disorder among US children. Pediatrics, 142(6), Article e20174161. https://doi.org/10.1542/peds.2017-4161

21.

Kuhlthau

K. A.

Delahaye

Erickson-Warfield

Shui

Crossman

van der Weerd

(2016). Health care transition services for youth with autism spectrum disorders: Perspectives of caregivers. Pediatrics, 137(Suppl. 2), S158–S166. https://doi.org/10.1542/peds.2015-2851N

22.

Kuo

A. A.

Crapnell

Lau

Anderson

K. A.

Shattuck

(2018). Stakeholder perspectives on research and practice in autism and transition. Pediatrics, 141(Suppl. 4), S293–S299. https://doi.org/10.1542/peds.2016-4300F

23.

Lai

J. K.

Weiss

J. A.

(2017). Priority service needs and receipt across the lifespan for individuals with autism spectrum disorder. Autism Research, 10(8), 1436–1447. https://doi.org/10.1002/aur.1786

24.

Laxman

D. J.

Taylor

J. L.

DaWalt

L. S.

Greenberg

J. S.

Mailick

M. R.

(2019). Loss in services precedes high school exit for teens with autism spectrum disorder: A longitudinal study. Autism Research, 12(6), 911–921. https://doi.org/10.1002/aur.2113

25.

Magana

Parish

S. L.

Rose

R. A.

Timberlake

Swaine

J. G.

(2012). Racial and ethnic disparities in quality of health care among children with autism and other developmental disabilities. Intellectual and Developmental Disabilities, 50(4), 287–299. https://doi.org/10.1352/1934-9556-50.4.287

26.

Mandell

D. S.

Listerud

Levy

S. E.

Pinto-Martin

J. A.

(2002). Race differences in the age at diagnosis among Medicaid-eligible children with autism. Journal of the American Academy of Child & Adolescent Psychiatry, 41(12), 1447–1453. https://doi.org/10.1097/01.CHI.0000024863.60748.53

27.

Matthews

N. L.

Smith

C. J.

Pollard

Ober-Reynolds

Kirwan

Malligo

(2015). Adaptive functioning in autism spectrum disorder during the transition to adulthood. Journal of Autism and Developmental Disorders, 45(8), 2349–2360. https://doi.org/10.1007/s10803-015-2400-2

28.

Monteiro

S. A.

Dempsey

Berry

L. N.

Voigt

R. G.

Goin-Kochel

R. P.

(2019). Screening and referral practices for autism spectrum disorder in primary pediatric care. Pediatrics, 144(4), Article e20183326. https://doi.org/10.1542/peds.2018-3326

29.

Nicolaidis

Raymaker

D. M.

Ashkenazy

McDonald

K. E.

Dern

Baggs

A. E. V.

Kapp

S. K.

Weiner

Boisclar

W. C.

(2015). “Respect the way I need to communicate with you”: Healthcare experiences of adults on the autism spectrum. Autism, 19(7), 824–831. https://doi.org/10.1177/1362361315576221

30.

NLTS2. (2002). National Longitudinal Transition Study-2, Wave 1 (Parent) interview. National Center for Special Education Research, SRI International. https://nlts2.sri.com/data_dictionary/downloads/Instruments/NLTS2_W1_Parent.pdf

31.

Pickard

K. E.

Ingersoll

B. R.

(2016). Quality versus quantity: The role of socioeconomic status on parent-reported service knowledge, service use, unmet service needs, and barriers to service use. Autism, 20(1), 106–115. https://doi.org/10.1177/1362361315569745

32.

Pinquart

Sorensen

(2006). Gender differences in caregiver stressors, social resources, and health: An updated meta-analysis. The Journals of Gerontology: Series B, 61(1), P33–P45. https://doi.org/10.1093/geronb/61.1.P33

33.

Platos

Pisula

(2019). Service use, unmet needs, and barriers to services among adolescents and young adults with autism spectrum disorder in Poland. BMC Health Services Research, 19(1), Article 587. https://doi.org/10.1186/s12913-019-4432-3

34.

Roth

B. M.

Kralovic

Roizen

N. J.

Spannagel

S. C.

Minich

Knapp

(2016). Impact of autism navigator on access to services. Journal of Developmental & Behavioral Pediatrics, 37(3), 188–195. https://doi.org/10.1097/DBP.0000000000000261

35.

Roux

A. M.

Shattuck

P. T.

Rast

J. E.

Rava

J. A.

Anderson

K. A.

(2015). National autism indicators report: Transition into young adulthood. Life Course Outcomes Research Program, A. J. Drexel Autism Institute, Drexel University. https://drexel.edu/autismoutcomes/publications-and-reports/publications/National-Autism-Indicators-Report-Transition-to-Adulthood

36.

Ruble

L. A.

McGrew

J. H.

Toland

Dalrymple

Adams

Snell-Rood

(2018). Randomized control trial of COMPASS for improving transition outcomes of students with autism spectrum disorder. Journal of Autism and Developmental Disorders, 48(10), 3586–3595. https://doi.org/10.1007/s10803-018-3623-9

37.

Shattuck

P. T.

Wagner

Narendorf

Sterzing

Hensley

(2011). Post-high school service use among young adults with an autism spectrum disorder. Archives of Pediatrics & Adolescent Medicine, 165(2), 141–146. https://doi.org/10.1001/archpediatrics.2010.279

38.

Smith-Young

Murray

Swab

(2018). Parents’ and guardians’ experiences of barriers and facilitators in accessing autism spectrum disorder diagnostic services for their children: A systematic review protocol of qualitative evidence. JBI Evidence Synthesis, 16(5), 1141–1146. https://doi.org/10.11124/JBISRIR-2017-003437

39.

Taylor

J. L.

Henninger

N. A.

(2015). Frequency and correlates of service access among youth with autism transitioning to adulthood. Journal of Autism and Developmental Disorders, 45(1), 179–191. https://doi.org/10.1007/s10803-014-2203-x

40.

Tessler

Gamache

(1995). Toolkit for evaluating family experiences with severe mental illness. The Evaluation Center at the Human Services Research Institute.

41.

U.S. Office of Management and Budget. (2010). 2010 Standards for delineating metropolitan and micropolitan statistical areas. https://www.govinfo.gov/content/pkg/FR-2010-06-28/pdf/2010-15605.pdf

42.

Vogan

Lake

J. K.

Tint

Weiss

J. A.

Lunsky

(2017). Tracking health care service use and the experiences of adults with autism spectrum disorder without intellectual disability: A longitudinal study of service rates, barriers and satisfaction. Disability and Health Journal, 10(2), 264–270. https://doi.org/10.1016/j.dhjo.2016.11.002