Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Nov 11.
Published in final edited form as: J Autism Dev Disord. 2021 Apr;51(4):1054–1066. doi: 10.1007/s10803-020-04586-1

Delay to early intensive behavioral intervention and educational outcomes for a medicaid-enrolled cohort of children with autism

Adele F Dimian 1, Frank J Symons 2, Jason J Wolff 3
PMCID: PMC12599662  NIHMSID: NIHMS2119941  PMID: 32642958

Abstract

Increased prevalence of autism spectrum disorder (ASD) has underscored the need for early intervention services. Early Intensive Behavioral Intervention (EIBI) is among the most common evidence-based approaches, however, stakeholders report significant waitlists. The effects of these delays to intervention are unknown. The purpose of this study was to evaluate the effects of delay to EIBI for preschool aged children with ASD on later educational outcomes. Medicaid records from Minnesota (2008–2010) were used to evaluate a cohort diagnosed with ASD and their later educational outcomes from 2010 to 2014 (n= 667) using generalized estimating equations. Approximately 70% of children experienced a delay to EIBI and children that experienced less delay and started EIBI at a younger age had better educational outcomes.

Keywords: autism, waitlists, early intensive behavioral intervention

The latest prevalence estimates of autism spectrum disorder (ASD) indicate that the neurodevelopmental disability is estimated to impact 1 in 59 children (8 -years of age) nationwide and roughly 1% world-wide (Baio et al., 2018). With the increase in diagnoses of ASD in the past decade, it is critical that children and families have access to high quality services early on (Chasson, Harris, & Neely, 2007). There is currently no standard treatment recommended for all individuals with ASD; however, the extant literature indicates that early intensive behavioral intervention (EIBI) can have ameliorating effects and is associated with positive long-term outcomes for many children diagnosed with the disorder (Sallows & Graupner, 2005; Reichow, 2012; Makrygianni, Gena, Katoudi, & Galanis, 2018). Addressing areas of impairment and increasing adaptive behavior early on for children with ASD is imperative to increase independence in daily living and quality of life across the lifespan. Early intervention is also cost-effective and health care expenditures to cover services early on has expanded within the United States to meet the needs of those with ASD (Chasson, Harris, & Neely, 2007).

states vary greatly in system structure and administration; eligibility criteria; interagency coordination; and service delivery

Coverage of EIBI, an efficacious early intervention modality based on the principles of applied behavior analysis (Lovaas, 1987; Makrygianni et al., 2018; Reichow, 2012), has increased in an effort to meet the needs of families and children with ASD. States differ in eligibility criteria, system structure, early intervention service delivery, and administration (Stahmer, Dababnah, & Rieth, 2019). In general, early intervention services are afforded and federally mandated by the Individuals with Disabilities Education Act (IDEA, 2004) in the United States through Part C (birth to two) and Part B (three to 21 years old). Approximately 89% of children 3 to 5 years old served under IDEA receive services in regular early childhood programs or a separate special education classroom environment and 8.5% receive their services at a service provider or within the home (U.S. Department of Education, 2020). Similarly, publicly funded Medicaid and private health insurance in many States provide coverage of EIBI and community based service providers/ agencies that typically provide in home or center based clinical services. There is therefore considerable variability in how families access and navigate the service landscape and funding sources after a diagnosis is given, but the research to date suggests that the intensity of early intervention is paramount (e.g., Granpeesheh, Dixon, Tarbox, Kaplan, & Wilke, 2009). Recommendations have been clear that at least 25 hours of intensive intervention is needed each week for at minimum 12 months to promote positive outcomes (National Research Council, 2001). Estimates unfortunately suggest that this is not always the case and that there are numerous barriers to service that families face (Stahmer, 2007; Wise, Little, Holliman, Wise, & Wang, 2010).

Disparities in access to services have been well documented in terms of racial differences (Mandell et al., 2010), age of ASD diagnoses for Medicaid-enrolled children (Liptak et al., 2008; Mandell et al., 2010), and geographic barriers to accessing services (Murphy & Ruble, 2012; Thomas, Ellis, McLaurin, Daniels, & Morrissey, 2007). For instance, there are reports that among children who receive a diagnosis of ASD, that 31% experience problems getting referrals, 14% experience delayed care, 43% have unmet care coordination needs, and 28% have trouble using services across the United States (Thomas et al., 2012). In regards to demographics, lower income families, racial minorities, and those with lower levels of education tend to report problems with accessing early intervention services at higher rates than more educated or higher income families (Bailey et al., 2004). Taken altogether, families with a child with ASD are more likely to have problems accessing services in general compared to caregivers of children with special health care needs or other types of developmental delays (Krauss, Gulley, Sciegaj, & Wells, 2003; Ruble, Heflinger, Renfrew, & Saunders, 2005; Siklos & Kerns, 2007) and often report unmet therapy needs (Chiri & Warfield, 2012).

With the acceleration in ASD prevalence and increase in demand for early intervention services, there are reported waiting periods that families encounter for both an initial ASD evaluation/ medical diagnosis of ASD in clinical settings and a time lag for starting intervention services by stakeholders (Hewitt et al., 2012; Yingling, Hock, & Bell, 2018). There are few estimates of current wait times to start community-based ASD services in general and EIBI, specifically. In one state estimate of a Medicaid-funded EIBI program similar to the current study, children waited on average for 3 years from diagnosis to treatment onset (Yingling et al., 2018). The associated effects of wait times on later outcomes has received limited attention, however, previous research has shown that later positive outcomes are associated with starting EIBI younger (Flanagan, Perry, & Freeman, 2012; Perry et al., 2011; Rogers et al., 2012; Smith, Klorman, & Mruzek, 2015; Vivanti & Dissanayake, 2016). The specific outcomes observed for younger children with ASD who received EIBI (i.e., <4 years) had better cognitive outcomes (Flanagan et al., 2012), adaptive behavior (Smith et al., 2015), achieved average functioning (Perry et al., 2011), lower autism symptoms (Smith et al., 2015) and had superior verbal language gains (Vivanti & Dissanayake, 2016) at follow up.

There are several potential factors that could contribute to waitlists and delay to starting ASD services and EIBI. For example, there are widespread shortages of qualified service providers, such as Board Certified Behavior Analysts (BCBA), Speech-Language Pathologies (SLP), and occupational therapists that contribute to the waitlist for services (Wise et al., 2010). Over 80% of states reported shortages of ASD-related personnel (Wise et al., 2010). Given that prevalence estimates of ASD are trending upward, the strain on resources and providers will likely continue to grow. Long waits for ASD services are therefore a significant concern of both parents, and healthcare providers and educators, especially given the evidence on the efficacy of early intervention services. Timely access to high quality early intervention services may improve quality of life for both children and families and given the developmental window early on, early diagnosis and treatment is imperative.

Educational Outcomes for Children with ASD

In general, there is limited research on the predictors of educational outcomes and academic achievement for individuals with ASD. Previous research suggests a strong correlation between IQ and academic achievement, response to intervention, and academic progress for individuals with ASD (Keen et al., 2016; Kim et al., 2018; Mayes-Dickerson & Calhoun, 2003). Under-achievement of students with ASD is reported compared to neuro-typical peers (Ashburner, Ziviani, & Rodger, 2010). In one study that characterized academic functions and predictors of achievement among twenty-six children with ASD, multiple regression analyses indicated weaknesses in reading comprehension relative to word reading (Miller et al., 2017). Mathematic skills were better for the group, but math reasoning was lower. A review of academic achievement among students with ASD further showed that reading achievement was proportionate with IQ (Keen et al., 2016; Kim, Mayes-Dickerson & Calhoun, 2003). Reading comprehension skills frequently appear to be impaired among some students with ASD (Miller et al., 2017; Minshew, Goldstein, Taylor, & Siegel, 1994; Nation et al., 2006; Troyb et al., 2014). Similar results were found in terms of mathematics where students with higher ability were either average or below average on mathematics performance (Estes et al., 2011; Keen et al., 2016; Mayes-Dickerson & Calhoun, 2003; Tryob et al., 2014).

With regards to instructional placements, more children with ASD receiving EIBI than children receiving treatment as usual/eclectic treatments were placed in a general education classroom with or without support (Cohen et al., 2006; Kim et al., 2018; Lovaas, 1987; Magiati et al., 2007; McEachin et al., 1993; Remington et al., 2007; Sheinkopf & Siegel, 1998; Smith et al., 2000). Special education and self-contained classrooms were the second most common reported instructional placements for children with ASD who received EIBI (e.g., Remington et al., 2007). Additionally, younger age at EIBI and higher IQ at intake was predictive of being placed in a general education classroom (Harris & Handleman, 2000). In summary, there is limited literature on the academic achievement and school-based performance of children with ASD.

Study Aims

The overall goal of this study was to characterize the effects of delay in EIBI service on educational outcomes in elementary school for a state-wide cohort of Medicaid-enrolled children diagnosed with ASD. We investigated the extent to which a time lag between being given a diagnosis of ASD to starting intensive services was associated with later instructional placements, an educational diagnosis of ASD, special education hours, and standardized testing scores. We hypothesized that children with ASD that experienced less delay to service would have better educational outcomes overall.

Method

The current study was a population-based (state-wide) observational study based on cross-linked data from the Minnesota Department of Human Services (DHS) and Minnesota Department of Education (MDE) administrative records. Data from Medicaid- enrolled families were used to identify a cohort of approximately 3 to 5 year olds with a diagnosis of ASD. The ICD-9-CM billing code 299.0 for autistic disorder was used to identify EIBI service recipients anytime between January 1st, 2008 and December 31st, 2010. During this time within the state of Minnesota, Medicaid was one of the only entities that covered EIBI services. We matched the data records from MDE and DHS using Link Plus and probabilistic matching of first name, last name, middle name, and date of birth. The match rate was 94.5%. and all data were de-identified for privacy and confidentiality reasons after linking was complete (see Figure 1 for cohort development). All children included in the cohort received EIBI services before entering elementary school and had a diagnosis of ASD (n=667).

Figure 1.

Figure 1.

Open in a new tab

Sample inclusion and exclusion.

Data Sources

The Medicaid Management Information System (MMIS) is Minnesota’s automated system for payment of medical claims and capitation payment for Minnesota Health Care Programs (MHCP). The specific billing code used to identify the cohort examined was ICD-9 CM 299.0 for autistic disorder. EIBI billing claims were identified for individualized skills training and family skills training (H2014 UA/HR). Individualized/ family skills training are general billing codes used for direct service by EIBI service providers. The list of providers from the dataset was cross referenced with a list of EIBI eligible service providers. Comorbid intellectual disability and communication disorder codes were also recorded, as well as billing claims from speech, occupational, and physical therapy.

The Minnesota Automated Reporting Student System (MARSS) was used to evaluate the educational outcomes and includes administrative data from school districts on every student enrolled. Minnesota uses MARSS data to report on compliance with state and federal educational mandates such as IDEA. The data system contains information on students’ attendance, special education service, primary disability diagnosis, district numbers, and eligibility for free or reduced lunch/meals. We utilized data from school years 2010–2014.

The Minnesota Comprehensive Assessments (MCA) are state academic achievement tests that help districts measure student academic progress towards standards specified under the Elementary and Secondary Education Act. At 3rd grade, students take the reading and mathematics sections of the test. In 5th grade, students are also required to take a science section. This study focused only on the reading and mathematics scores because only a subgroup of students participated in the science section. Students receiving special education services for cognitive disabilities are exempt from the MCA and can take an adapted version. The MCA-II was used by Minnesota school districts in school year 2011. The version was updated the following year and the MCA data for school years 2012 to 2014 were from the MCA-III.

Independent Variables

Delay to EIBI.

The primary independent variable was delay to start EIBI services from the date of diagnosis to the start of EIBI date. The first billing date in the MMIS dataset with a diagnosis (billing code 299.00) was used as the date of diagnosis. To confirm the date of diagnosis, the MMIS dataset was examined for the diagnostic billing code 12 months prior. If an earlier date of diagnosis was found, outside the date range of January 1st, 2008 to December 31st, 2010, that was the date used for the date of diagnosis. Date of ASD diagnosis was then subtracted from the first billing code date for individualized skills training and/or family skills training (H2014 UA/HR) to yield number of months until services started. If EIBI services were received before an ASD diagnosis was given (i.e., for developmental delay), the data were coded as a zero.

Average hours of EIBI service per week.

The average hours of EIBI service per week was derived by taking each unit billed for (e.g., 1 unit = 15 min) and taking the sum of the units for each participant and multiplying them by 15 to get total minutes of EIBI services for the date range examined. Total minutes billed were then divided by 60 to get total number of hours of service.

Age, gender and race.

Age of diagnosis, gender, and race were based on billing claims in the MMIS dataset. Approximately 18% of the cohort was female and the racial groups identified were American Indian (2.4%), Asian (5.1%), Hispanic (6.1%), Black (14.7%), and White (71.7%; Table 3).

Table 3.

Generalized Estimating Equation Estimates for Educational Outcomes

Outcomes
Predictors General Education+ (n=667) ASD Diagnosis+ (n=667) Special education service hours++ (n=605) MCA Math Scores++ (n=257) MCA Reading Scores++ (n=256)
Delay to EIBI in Mos 0.97 *** (0.95,0.98) 1.03** (1.01,1.04) −0.01** (0.01) −0.007** (0.003) −0.003 (0.003)
Diagnosis Age in Yrs 0.70*** (0.59,0.82) 0.76** (0.63,0.92) −0.12* (0.06) −0.09** (0.03) −0.03 (0.04)
Average EIBI hrs per wk 0.99 (0.98,1.01) 1.04*** (1.02,1.06) −0.01** (0.004) −0.004 (0.003) −0.004 (0.003)
Male 1.10 (0.75,1.60) 2.30*** (1.52,3.47) −0.34** (0.13) 0.09 (0.08) −0.06 (0.08)
Intellectual Disability 0.32*** (0.23,0.44) 0.95 (0.66,1.39) 0.19 (0.10) −0.17* (0.07) −0.20** (0.09)
Race- White (ref)
 Asian 0.76 (0.38,1.52) 2.17 (0.88,5.36) −0.14 (0.21) 0.11 (0.14) −0.07 (0.15)
 Hispanic 1.20 (0.66,2.19) 0.72 (0.40, 1.31) −0.25 (0.16) −0.05 (0.13) −0.08 (0.14)
 Black 0.45*** (0.30,0.69) 0.66 (0.41, 1.06) 0.26* (0.12) −0.47*** (0.11) −0.45** (0.12)
 American Indian 0.69 (0.25,1.92) 0.47 (0.18, 1.21) 0.08 (0.38) −0.04 (0.16) −0.06 (0.15)
Non-metro residence 0.98 (0.72,1.34) 0.68* (0.47,0.97) −0.30 ** (0.10) −0.04 (0.06) −0.02 (0.07)
Speech services 0.68* (0.49,0.95) 1.25 (0.86,1.83) −0.17 (0.13) −0.10 (0.07) −0.12 (0.07)
OT services 1.04 (0.76,1.44) 1.05 (0.72, 1.53) −0.11 (0.13) 0.07 (0.06) 0.18** (0.07)
PT services 1.00 (0.64,1.57) 0.45** (0.27,0.74) 0.15 (0.14) 0.05 (0.08) 0.05 (0.10)
Free/reduced lunch 1.07 (0.78,1.47) 0.59** (0.39,0.88) 0.04 (0.11) −0.10 (0.07) −0.08 (0.08)
School Year- 2010 (ref)
 Year 2 (2011) 0.84 (0.69,1.02) 1.02 (0.88,1.17) −0.16 (0.10)
 Year 3 (2012) 0.67 (0.53,0.83) 1.11 (0.94,1.31) −0.10 (0.10) −0.13 (0.10) −0.11 (0.07)
 Year 4 (2013) 0.60 (0.48,0.76) 1.13 (0.94,1.37) −0.56*** (0.11) −0.23* (0.11) −0.38*** (0.08)
 Year 5 (2014) 0.58 (0.46,0.74) 1.09 (0.89,1.35) −0.40*** (0.12) −0.29** (0.11) −0.31*** (0.08)
Open in a new tab

Note.

*

p<.05

**

p<.01

***

p <.001 based on Type- III Wald chi-square test

+

indicates adjusted odds ratios with 95% confidence interval and

++

indicates unstandardized partial regression coefficient representing differences in educational outcomes associated with the predictors (b, SE). Reference school years for the MCA was 2011.

Residence.

County of residence was identified based on the MMIS claims data. County data were then recoded into a binary variable for residence in the seven county metro area (n=460) or non-metro areas (n=207) within the state of Minnesota. The metro area is made up of 7 of the 87 counties in Minnesota including: Hennepin, Anoka, Carver, Scott, Dakota, Washington, and Ramsey counties.

Comorbid services.

Participation in other rehabilitative services was also examined. Billing claims from the MMIS dataset for occupational therapy, physical therapy, and speech language therapy were recoded into a binary variable. The billing codes included 92507 GN, 97110 GO, 97110 GP and 97530 GP. Overall, approximately 52% of the cohort received speech services, 43% received occupational therapy, and 12% received physical therapy.

Intellectual disability status.

ICD-9-CM billing codes for mild, moderate, severe, profound, and unspecified intellectual disability from the MMIS dataset (i.e., time 1) were also examined (317, 318, 318.1, 318.2, and 319). Mild intellectual disability is classified as an intellectual quotient (IQ) of 50 to 70. Moderate consists of an IQ between 35 and 49. Severe classification is an IQ of 20–34 and profound is less than 20. Comorbid codes for language disorders (315.3) and developmental delay (315.8) were also included. For the descriptive analysis of the cohort, each classification was used as a categorical variable. In the statistical analyses conducted, a binary variable was created for intellectual disability status (0= no intellectual disability, 1=mild, moderate, severe, profound, and unspecified intellectual disability) to aggregate the data.

Free or reduced lunch receipt.

The only economic indicator available for the cohort was free or reduced lunch receipt from the MARSS dataset. The U.S. Department of Agriculture sets the annual eligibility criteria for the National School Meal Program based on family income and size. Families qualified for reduced price meals if their income was 131–185% of the poverty level. Incomes at or below 130% of the poverty level qualified for free price meals. At entry to elementary school, free or reduced lunch receipt was coded as a binary variable and used in the descriptive and statistical data analyses conducted. Approximately 65% of the cohort received free or reduced lunch in school.

Dependent variables

The dependent variables included an educational diagnosis of ASD (yes or no), the instructional placement (general education, special education resource room, separate classroom, or a separate school for special education), special education service hours, and MCA participation/scale scores for reading, math, and science subscales.

Instructional placement setting.

Instructional placement setting is an educational placement setting examined at each year of follow up. A placement in general education included students who received the majority of their special education and related services in a regular class. General education placement consisted of children with disabilities receiving special education and related services outside the regular classroom for less than 21% of the school day. Resource room placement included students who received special education and related services outside the regular classroom for 21% to 60% of the school day. A separate class placement (i.e., self-contained classroom) consisted of children with disabilities receiving special education and related services outside the regular classroom for more than 60% of the school day. Finally, a separate school placement included students with disabilities receiving special education and related services for greater than 50% of the school day in separate facilities in a public or private facility.

ASD diagnosis in school.

Students that were assessed and identified by the school as needing/ receiving special education services, or had a signed Individualized Education Plan (IEP), Individual Family Service Plan (IFSP), or Individual Learning Plan (IILP), had a primary disability reported in the MARSS dataset. There are 13 possible disability categories that a student could qualify under for special education services from 6 to 21-years old and 14 categories for early childhood special education, which included a developmental delay diagnoses (used for birth to 6 years old only). A binary variable was created for each school year evaluated based on if a primary educational diagnosis of ASD (yes, no) was given in elementary school. An educational diagnosis was included to evaluate if the student continued to require autism specific supports and services when they entered elementary school.

MCA scores.

There were three subscales on the MCA achievement test: reading, mathematics, and science. Scale scores were utilized as a continuous variable in the statistical analyses performed.

Special education service hours.

Special education service hours are used in generating tuition billing for special education in the MARSS dataset. Licensed special educational teacher administered direct and indirect special education services were included. Data on one-to-one para professional hours and hours for programs were not available. Special education service hours were restricted in the statistical analyses by removing any zero hours from the distribution. We included special education service hours as a proxy for intensity of services needed in school (i.e., if more special education service hours were allocated to a student, that would indicate they needed more support than another, in comparison).

Statistical analyses

We used SPSS version 22 to conduct all analyses. All statistical tests were two-sided with an alpha level of 0.05. To investigate the relationship between delay in months to start EIBI services and educational outcomes at 4–6 years of follow up (at each school year time point), Generalized Estimating Equations (GEE) regression analyses were conducted. GEE is a marginal (population-average) approach used to model longitudinal data generated from repeated measures (Zeger & Liang, 1986). GEEs are an extension of generalized linear modelling and take into account the correlation among outcomes measured repeatedly over time (Ballinger, 2004).

Intellectual disability status, gender, race, county/region of residence, comorbid service receipt (SLT, OT, or PT), free/reduced lunch receipt, and school year were included as covariates/predictors in each GEE model calculated to account for heterogeneity among the cohort. The assumptions were met for each GEE model conducted (sufficient sample size, and observations independent) and multicollinearity of the predictor variables was assessed. Variance inflation factors (VIF) indicated acceptable levels of multicollinearity (<10) and bivariate correlations were all <.80. To test the sensitivity of the results to the correlation structure, we conducted robustness checks with alternative specifications. The results of the robustness checks yielded similar estimates across unstructured, autoregressive (AR(1)), and exchangeable working correlational matrices. An unstructured correlation structure does not have constraints across observations and are estimated from the data without restriction. Autoregressive differs in that the correlation over time diminishes exponentially with time. Lastly, exchangeable correlation structure co-varies equally across all observations (see Zorn, 2001 and Garson, 2013 for more information on correlation structures). Quasi-likelihood under independence criterion (QIC) is a goodness-of-fit measure and is an adaptation of AIC for repeated measures. The QIC coefficient was used to select the best working correlation structure for each model implemented (lower values indicate better fit).

GEE for a binomial distribution with a logistic regression were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for a general education placement (yes, no) and an ASD diagnosis (yes, no). GEE linear regression with a Gamma distribution was used to estimate coefficients and 95% CIs for special education service hours and MCA-reading or math scale scores. Based on the QIC values, unstructured correlational matrices were used for general education placement and math scale score outcomes. Autoregressive correlation structure was used for ASD diagnosis, special education hours and reading scale score outcomes. Robust standard errors were used in all models to account for correlation across the school years. Unstandardized beta- weight coefficients, odds ratios, 95% confidence intervals, Type- III Wald chi-square tests for the null hypothesis (test of model effects) and Wald chi-square test for each parameter (predictor) were estimated. Finally, post-hoc power analyses indicated that the observed power was sufficient for all statistical tests conducted. Missing data were excluded from the analyses.

Results

Cohort Demographics

The cohort of 667 children analyzed in this study was 82.2% male, and 71.7% White (Table 1). Within the cohort, approximately 15% of the children were Black, 6% Hispanic, 5% Asian, and 2% American Indian. A majority of the children were diagnosed with ASD at age 4 (35.4%; range 2 to 6 years old) and 92% had a comorbid intellectual disability, language disorder, or developmental delay between 3 and 5 years of age. There were no statistically significant differences between age of ASD diagnosis and race, language disorder, or developmental delay. Language disorder was the most common (39%) comorbid impairment followed by unspecified intellectual disability (20.4%). There Upon entry into elementary school (i.e., between 2010 and 2014), 64.5% of the cohort qualified for free or reduced priced lunch, 1% were homeless, and 6% had limited English proficiency.

Table 1.

Demographics

Characteristics % of sample (n=667)
Gendera,b
 Female 17.8%
 Male 82.2%
Race/ethnicitya,b
 American Indian 2.4%
 Asian 5.1%
 Hispanic 6.1%
 Black 14.7%
 White 71.7%
Age of ASD diagnosisa
 2- years old 3.4%
 3- years old 29.7%
 4 -years old 35.4%
 5 -years old 26.5%
 6-years old 4.9%
Free/Reduced Price Lunch receipt 64.5%
Homelessb 1.4%
Limited English Proficiencyb 5.9%
Comorbid Disabilitya
 Language disorder 39.0%
 Developmental delay 15.4%
 Mild intellectual disability (IQ 50–70) 7.5%
 Moderate intellectual disability (IQ 35–49) 5.8%
 Severe intellectual disability (IQ 20–34) 3.4%
 Profound intellectual disability (IQ <20) 0.7%
 Unspecified Intellectual disability 20.4%
Average hours of EIBI per weeka 19.04 (12.00)*
Average delay to EIBI in monthsa 8.99 (10.63)*
Received Speech and Language Therapya 51.9%
Received Occupational Therapya 42.9%
Received Physical Therapya 12.4%
Open in a new tab

Note.

*

indicates estimate is represented by a mean (sd).

a

indicates the variable is from the MMIS dataset and

b

indicates the variable is from the MARSS dataset.

Delay to EIBI, Age of Diagnosis, and Services Received

All children included in the cohort received EIBI services between 2008 and 2010. The average delay to begin EIBI was 8.99 months (range= 0 to 45 months; SD= 10.63 months). The average hours of EIBI per week received by the cohort was 20.23 hours (SD=14.68 hours). Table 2 shows the average delay to EIBI in months and the average hours per week of EIBI by gender, racial group, residence, and comorbid disability. Males experienced a slightly greater average delay to EIBI between diagnosis and the start of EIBI (M= 9.14 months; SD=10.65 months; range=0–44 months) than females (M=8.29 months; SD=10.48 months; range=0–45 months); this difference was not statistically significant. Average delay to EIBI varied by race and there was a statistically significant difference between children who resided in the metro and non-metro areas. Children in the metro area had a larger average delay to EIBI compared to children who resided in the non-metro area. Children with severe ID experienced the longest delay with an average delay to start EIBI of 16.22 months.

Table 2.

Descriptives

Demographics Average delay to EIBI in months (SD) Average age of ASD diagnosis (SD) Average hours per week of EIBI (SD) SLT Receipt OT Receipt PT Receipt
Gender
Female (n=119) 8.29 (10.48) 3.87 (0.94) 19.41 (12.42) 49% 47% 13%
Male (n=548) 9.14 (10.65) 3.99 (0.89) 18.96 (11.92) 53% 55% 12%
Racial group
American Indian(n= 16) 5.13 (5.85) 4.46 (0.81) 10.44** (6.42) 50% 25% 13%
Asian (n=34) 8.15 (8.96) 4.07 (0.84) 23.12 (11.05) 44% 50% 15%
Hispanic (n= 41) 7.59 (8.76) 4.00 (0.85) 15.76 (12.40) 59% 44% 10%
Black (n=98) 10.96 (11.84) 3.82 (0.87) 17.81 (12.39) 65% 50% 9%
White (n=478) 8.90 (10.67) 3.97 (0.91) 19.57 (11.92) 49% 41% 13%
Residence
Metro (n=460) 9.89* (11.24) 3.97 (0.90) 19.36* (12.45) 59%*** 48%*** 13%
Non-metro (n=207) 7.00 (8.78) 3.97 (0.90) 18.33 (10.94) 37% 31% 11%
Comorbid Disability
Language Disorder (n=260) 10.17* (11.17) 3.90 (0.90) 17.34* (11.85) 84%*** 63%*** 16%**
Developmental delay (n=103) 9.09 (10.84) 3.92 (0.93) 17.77 (11.33) 74%*** 59%*** 23%***
Mild ID (n=50) 11.92* (12.00) 4.33* (0.80) 17.86 (10.47) 66%* 52% 16%
Moderate ID (n=39) 12.92* (12.62) 4.08 (0.76) 18.88 (12.73) 54% 51% 8%
Severe ID (n=23) 16.22* (12.82) 3.85 (0.77) 19.98 (12.81) 65% 57% 30%***
Profound ID (n=5) 11.20 (10.85) 4.33 (0.44) 14.93 (8.07) 60% 80% 20%
Unspecified ID (n=136) 11.75* (11.82) 3.78* (0.82) 19.49 (12.00) 65%*** 54%*** 21%***
Open in a new tab

Note. SLT= speech language therapy, OT= occupational therapy, PT= physical therapy

*

p<.05

**

p<.01

***

p<.001 based on pairwise Chi-square or independent t-test analyses.

Over 34% of the cohort received an ASD diagnosis at age 4. The average age of ASD diagnosis was similar across gender, racial groups, and comorbid disabilities. There was no difference observed by residence, however, children who were American Indian were diagnosed later and received the least number of hours of EIBI, on average. The average hours per week of EIBI was 19 hours. Over half of the cohort received SLT along with EIBI and approximately 43% of the sample received OT. A smaller proportion of children in the non-metro areas received SLT, OT, and PT in comparison to those who resided in the metro areas. Results also indicated that only 12% of children in the cohort received physical therapy with the highest proportion among children with severe intellectual disabilities.

GEE Results

General education placement.

A GEE logistic regression model was conducted to determine the main effect of delay to EIBI on general education instructional placement at follow-up (i.e., school years 2010–2014) with age of diagnosis, gender, intellectual disability status, race, residence, other types of service receipt, free/reduced lunch receipt and school year as covariates (Table 3). Delay to EIBI, diagnosis age, race, intellectual disability status, speech services, and school year were statistically significant. Specifically, the odds of a general education placement decreased with every unit increase in delay to EIBI (OR=0.97; 95% CI=0.95–0.98; p <.001). The odds of general education placement also decreased if the student had a comorbid intellectual disability (OR=0.32; 95% CI=0.23–0.44; p <.001) and if the child was diagnosed at an older age (OR=0.70; 95% CI=0.59–0.82; p <.001). Race was statistically significant with students that were Black having lower odds than students that were White to receive a general education placement (OR= 0.45; 95% CI= 0.30, 0.69). Children that received speech therapy services had lower odds of a general education placement than children that did not receive speech therapy (OR=0.68; 95% CI = 0.49, 0.95; p=.03).

ASD diagnosis in school.

The odds of an ASD diagnosis in school increased if delay to EIBI was greater (OR=1.03; 95% CI=1.01–1.04; p =.001), the child was male (OR=2.30; 95% CI=1.52–3.47; p =.001), or was White (X2= 9.74; p =.05). Odds of an ASD diagnosis decreased if the child received a diagnosis at an older age (OR=0.76; 95% CI=0.639–0.92; p =.001) and resided in the non-metro area (OR=0.68; 95% CI=0.47–0.97; p=.03). There was a decrease in the odds of an ASD diagnosis in school at follow-up if children received physical therapy (OR=0.45; 95% CI=0.27–0.74; p =.002) or free/reduced lunch receipt (OR=0.59; 95% CI=0.39–0.88; p <.001).

Special education service hours.

Delay to EIBI was statistically significant with special education service hours, decreasing with greater delay to EIBI (B=−0.01; 95% CI= −0.02–0.003; p=.01). Average EIBI hours per week (B =−0.01; 95% CI= −0.02, −0.003; p=.01), age of diagnosis (B= −0.12; 95% CI= −0.24, −0.01; p= .04), gender (B=−0.34; 95% CI= −0.60; −0.08; p=.01), race (X2= 9.58; p=.05), residence (B=−0.30; 95% CI= −0.50, −0.10; p=.004), and school year were also statistically significant. Intellectual disability status approached significance, whereas receipt of other services and free/reduced lunch were not statistically significant.

MCA-Reading scale scores.

Delay to EIBI and average EIBI hours were not significantly associated with MCA- reading scale scores. Intellectual disability status, race, occupational therapy receipt, and school year were significantly associated with reading scale scores. Children with intellectual disabilities scored lower than children without intellectual disabilities (B= −0.20; 95% CI= −0.38, −0.03; p=.02). Additionally, children who were Black scored lower than children who were White (B= −0.45; 95% CI= −0.69, −0.21; p=.001). Children who received occupational therapy scored lower than those that did not (B= 0.18; 95% CI= 0.04–0.32; p=.01). Finally, school years 2013 and 2014 resulted in lower scores than the other years at follow-up.

MCA-Mathematics scale scores.

The main effect of delay to EIBI was statistically significant (B=−0.006; 95% CI= −0.01– 0.001; p=.03). Age of diagnosis, intellectual disability status, race, and school year were also significantly associated with MCA mathematics scores. Children that took the test and had comorbid intellectual disabilities scored lower than children without intellectual disabilities (B= −0.17; 95% CI= −0.30, −0.03, p=0.02). Children who were Black scored lower on average than students who were White (B= −0.47; 95% CI=−0.68, −0.25, p<.001) and school years 2013 and 2014 resulted in lower scores compared to 2011 (X2= 23.78; p<.001).

Discussion

The current study aimed to assess the impact of delays to EIBI services on educational outcomes for young children with ASD enrolled in Medicaid. The average delay to start EIBI was 9 months on average, a shorter estimate than others previously reported (e.g., Yingling et al., 2018). The average delay to EIBI across the state of Minnesota was nine months and over 30% waited a year or longer. The educational outcomes examined in this study at follow- up included instructional setting placement (e.g., general education), a primary educational diagnosis of ASD, special education service hours, participation and scores on standardized achievement tests. Over 94% of the sample examined qualified for special education services, and 70% qualified for a primary diagnosis of ASD in elementary school. Findings from the GEE models calculated for each educational outcome examined for school years 2010 to 2014 suggest that the main effect of delay to EIBI (in months) was significant. Specifically, the odds of receiving a general education placement and participating in the reading or math standardized achievement tests was decreased if the child experienced a longer delay to start EIBI. Additionally, the odds of receiving a primary educational diagnosis of ASD was also increased in associate with longer delay to EIBI. Overall, the results suggest that the time between when a diagnosis is made and the start of intensive services is important in optimizing outcomes for children with ASD.

This study extends previous findings that children diagnosed younger and starting EIBI services earlier tend to have better outcomes (Baker-Ericzen et al., 2007; Bibby et al., 2002; Flanagan et al., 2012; Granpeesheh et al., 2009; Harris & Handleman, 2000; MacDonald et al., 2014; Perry et al., 2011; Smith et al., 2015; Virués-Ortega & Rodriguez, 2013). The most common instructional placement setting was general education, however, a racial disparity was evident in the outcomes examined within Minnesota between Black and White students and of the eligible students, only half participated in standardized testing. Previous research has demonstrated similar disparities in other disability categories with Black students receiving more restricted instructional placements (e.g., Skiba, Poloni-Staudinger, Gallini, Simmons, & Feggins-Azziz, 2006).

Average hours of EIBI per week were used as a potential proxy for a dosage variable and has been used in other studies to analyze EIBI dose-response (e.g., Klintwall et al., 2015; Virués-Ortega, 2010). More hours of service in general suggest more severe autism symptoms (i.e., children with more severe ASD may need more service hours/higher treatment intensity). For the outcomes in which the average hours of EIBI per week and delay to EIBI was significant, the pattern (and direction) was the same. For instance, children receiving more EIBI hours per week on average, and who experienced more time lag to EIBI, had increased odds of an educational ASD diagnosis.

Comorbid service receipt varied by residence and comorbid disability status with SLT receipt during preschool years decreasing the odds of a later general education placement and of an educational diagnosis of ASD if PT services were needed. Children with a comorbid intellectual disability were also less likely to be in a general education placement. In contrast to previous research, within this cohort, there was little difference in terms of age of ASD diagnosis across gender, racial groups, and residence with most children receiving a diagnosis at 4 years old. Data were limited by birthdate to create the cohort evaluated and could explain in part these results. Recent data from the Centers for Disease Control and Prevention indicated that within Minnesota among 8 -year olds identified with a diagnosis of ASD, half of the children were diagnosed after 4 years of age (Baio et al., 2018). The odds of receiving an ASD diagnosis upon entry into elementary school was two-fold for males compared to females and reflects the increased risk of males being identified with ASD (Baio et al., 2018). Finally, children that also received free or reduced lunch in school were less likely to receive an educational diagnosis of ASD, indicating that there may be a difference in identification and special education services allocated by socio-economic group.

EIBI is considered to be efficacious and meta-analyses show repeatedly that children that receive EIBI make gains in adaptive behavior, communication, and IQ (e.g., Makrygianni et al., 2018; Reichow, 2012). The seminal study by Lovaas (1987) that serves as the foundational model for EIBI reported gains for children with ASD that resulted in children not needing any additional supports when they entered school. In the present study, we attempted to evaluate if a time lag to starting intensive services affected educational outcomes that are typically reported in administrative datasets, such as instructional placement. There has been much attention on the delay to diagnosis, which is needed to access services in some states, but we also need to focus on the time lag to starting early intervention services that are community -based to support a wraparound of service delivery across settings. Recent child count data from Minnesota, for instance, indicates that children with ASD are the third highest incident disability category served in schools statewide under Part C and Part B (Minnesota Department of Education, 2019). There is no evidence to suggest that ASD prevalence will decrease, therefore, we must examine what system wide changes are needed now to minimize access barriers to service receipt early on. Finally, the American Academy of Pediatrics clinical guidelines for ASD emphasize the importance of referrals to early intervention services as young as possible (Hyman, Levy, & Meyers, 2019). Cutting down on wait times for both diagnosis and service initiation should be prioritized by policy makers going forward.

Limitations

This study was observational and the educational outcomes were limited to those available in administrative databases. Other indicators for academic achievement were not available, such as the individualized education plans for the students receiving special education services. Instructional placement data are used as an indicator of state compliance with IDEA, but stakeholders can request that their child receive their educational programming in a particular setting. It is unknown if that was the case with any of the students in this study. Further, school-based early intervention services were not accounted for because there was no education data available for school years 2008 and 2009. Given that early childhood special education (ECSE) is the most common service provider for the age group evaluated it could be assumed that most children in the sample were also receiving ECSE due to IDEA (i.e., under federal law, early intervention services end at age 3 and are then determined by the child’s school). The sample is exclusively from one state and therefore generalizability may be limited. No information on caregiver’s education, income, or socioeconomic status, other than qualification for free of reduced meals, was available. Quality of EIBI service was not assessed or known. While there are standards of practice for EIBI service providers, there may be variability in terms of curriculum or type of EIBI service provided. Finally, educational diagnoses and labels of ASD do not use the same criteria as a medical diagnosis of ASD and may vary by school district and by State. We did not have data available on if the children continued to qualify and receive Medicaid-funded ASD services, which typically requires a medical diagnosis. Although this is a limitation, of interest to us in this study was if the children that started EIBI sooner continued to need ASD specific supports once they entered elementary school.

Future Research

One potential area for further research would be alternative service delivery strategies to decrease delays to intervention. A telehealth (internet based video-conferencing) model of service could potentially bridge the gap in service delivery time after a diagnosis. Implications of this type of research include more efficient allocation of services for families and children with ASD throughout Minnesota. For example, Vismara and colleagues (2016) used a randomized controlled trial to compare parent training using the Early Start Denver Model (i.e., EIBI) delivered via telehealth and community treatment as usual early intervention program. More parent satisfaction and implementation fidelity was observed after 12-weeks for the telehealth ESDM group (Vismara et al., 2016). Further, telehealth can reduce costs for providing services and could potentially eliminate any geographic barriers to autism-related services. In general, more research is needed to investigate how to deliver EIBI via telehealth and test whether such procedures are a viable means of reducing wait times for diagnostic assessment and intervention.

Future research should also extend the findings on the cost-effectiveness of timely access to EIBI. Piccininni and colleagues (2017) reported for example that in Canada the economic effect of eliminating wait times to EIBI was associated with lifetime savings of CDN$267,000 per individual when they compared estimates to current wait times. Although the cost of treatment is expensive, it can ultimately save society money in the long-term. With prevalence estimates at 1 in 54 for ASD diagnoses (Maenner et al., 2020), the burden on service providers is growing and may result in more children and families spending more time on waitlists for intensive services. A cost-benefit analysis of a delay to service to provide further evidence for why timely early intervention for children with ASD is critical.

In summary, the current study aimed to evaluate how delays to EIBI relate to educational outcomes for young children with ASD. The overall goal in providing services in a timely manner to children with ASD is to facilitate long-term positive outcomes. More support and research is warranted, but early intensive intervention is vital for children with ASD. The results of this study support the exigency of early diagnosis and timely, effective evidence based intervention and supports early on.

Contributor Information

Adele F. Dimian, Institute on Community Integration, University of Minnesota, Minneapolis, MN, USA

Frank J. Symons, Dept. of Educational Psychology, University of Minnesota, Minneapolis, MN, USA

Jason J. Wolff, Dept. of Educational Psychology, University of Minnesota, Minneapolis, MN, USA

References

  1. Bailey D, Hebbeler K, Scarborough A, Spiker D, & Mallik S (2004). First experiences with early intervention: a national perspective. Pediatrics, 113, 887–896. [DOI] [PubMed] [Google Scholar]
  2. Baio J, Wiggins L, Christensen DL, Maenner MJ, Daniels J, Warren Z, ... & Durkin MS (2018). Prevalence of autism spectrum disorder among children aged 8 years autism and developmental disabilities monitoring network, 11 sites, United States, 2014. MMWR Surveillance Summaries, 67(6), 1–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Chasson GS, Harris GE, & Neely WJ (2007). Cost comparison of early intensive behavioral intervention and special education for children with autism. Journal of Child and Family Studies, 16(3), 401–413. [Google Scholar]
  4. Chiri G & Warfield ME (2012). Unmet need and problems accessing core health care services for children with autism spectrum disorder. Maternal and Child Health Journal, 16, 1081–1091. [DOI] [PubMed] [Google Scholar]
  5. Estes A, Rivera V, Bryan M, Cali P, & Dawson G (2011). Discrepancies between academic achievement and intellectual ability in higher-functioning school-aged children with autism spectrum disorder. Journal of autism and developmental disorders, 41(8), 1044–1052. [DOI] [PubMed] [Google Scholar]
  6. Flanagan HE, Perry A, & Freeman NL (2012). Effectiveness of large-scale community based intensive behavioral intervention: A waitlist comparison study exploring outcomes and predictors. Research in Autism Spectrum Disorders, 6(2), 673–682. [Google Scholar]
  7. Granpeesheh D, Dixon DR, Tarbox J, Kaplan AM, & Wilke AE (2009). The effects of age and treatment intensity on behavioral intervention outcomes for children with autism spectrum disorders. Research in Autism Spectrum Disorders, 3(4), 1014–1022. [Google Scholar]
  8. Hewitt A, Timmons J, Nord D, Hall-Lande J, Moore T, Kliest B, et al. (2012). A report on early intervention services for Minnesota’s children with autism spectrum disorders. Minneapolis, MN: University of Minnesota, Institute on Community Integration, Research and Training Center on Community Living. [Google Scholar]
  9. Hyman SL, Levy SE, & Myers SM (2019). Identification, Evaluation, and Management of Children With Autism Spectrum Disorder. Pediatrics, 144 (6), 1–64. [DOI] [PubMed] [Google Scholar]
  10. Individuals with Disabilities Education Improvement Act, H.R. 1350, Pub. L. No. P.L. 108–446 (2004). [Google Scholar]
  11. Jarbrink K, & Knapp M (2001). The economic impact of autism in Britain. Autism, 5, 7–22. [DOI] [PubMed] [Google Scholar]
  12. Kim SH, Bal VH, & Lord C (2018). Longitudinal follow-up of academic achievement in children with autism from age 2 to 18. Journal of Child Psychology and Psychiatry, 59(3), 258–267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Krauss MW, Gulley S, Sciegaj M, & Wells N (2003). Access to specialty medical care for children with mental retardation, autism, and other special health care needs. Mental Retardation,41(5), 329–339. [DOI] [PubMed] [Google Scholar]
  14. Lovaas OI, (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55, 3–9. [DOI] [PubMed] [Google Scholar]
  15. Maenner MJ, Shaw KA, Baio J, Washington A, Patrick M, DiRienzo M….&Dietz PM (2020). Prevalence of autism spectrum disorder among children aged 8 years. Surveillance Summaries, 69(4), 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Makrygianni MK, & Reed P (2010). A meta-analytic review of the effectiveness of behavioural early intervention programs for children with autism spectrum disorders. Research in Autism Spectrum Disorders, 4, 577–593. [Google Scholar]
  17. Makrygianni MK, Gena A, Katoudi S, & Galanis P (2018). The effectiveness of applied behavior analytic interventions for children with Autism Spectrum Disorder: A meta analytic study. Research in Autism Spectrum Disorders, 51, 18–31. [Google Scholar]
  18. Mandell DS, Wiggins LD, Carpenter LA, Daniels J, DiGuiseppi C, Durkin MS, ... & Kirby RS (2009). Racial/ethnic disparities in the identification of children with autism spectrum disorders. American Journal of Public Health, 99(3), 493–498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. McIntyre LL, & Zemantic PK (2017). Examining services for young children with autism spectrum disorder: parent satisfaction and predictors of service utilization. Early childhood education journal, 45(6), 727–734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Minnesota Department of Education (2019). Child count. Retrieved from https://public.education.mn.gov/MDEAnalytics/DataTopic.jsp?TOPICID=455 [Google Scholar]
  21. Myers SM, & Johnson CP (2007). Management of children with autism spectrum disorders. Pediatrics, 120(5), 1162–1182. [DOI] [PubMed] [Google Scholar]
  22. Murphy MA, & Ruble LA (2012). A comparative study of rurality and urbanicity on access to and satisfaction with services for children with autism spectrum disorders. Rural Special Education Quarterly, 31(3), 3–11. [Google Scholar]
  23. National Autism Center (2015). Findings and conclusions: National standards project, phase 2. Randolph, MA: Author [Google Scholar]
  24. National Research Council. Committee on Educational Interventions for Children with Autism. In: Lord C, McGee JP, editors. Educating children with autism. National Academy Press; Washington, DC: 2001. Division of Behavioral and Social Sciences and Education [Google Scholar]
  25. Perry A, Cummings A, Geier JD, Freeman NL, Hughes S, Managhan T, ... & Williams J (2011). Predictors of outcome for children receiving intensive behavioral intervention in a large, community-based program. Research in autism spectrum disorders, 5(1), 592–603. [Google Scholar]
  26. Reichow B, & Wolery M (2009). Comprehensive synthesis of early intensive behavioral interventions for young children with autism based on the UCLA Young Autism Project model. Journal of Autism and Developmental Disorders, 39, 23–41. [DOI] [PubMed] [Google Scholar]
  27. Reichow B (2012). Overview of meta-analyses on early intensive behavioral interventions for young children with autism spectrum disorders. Journal of Autism and Developmental Disorders, 42, 512–520. [DOI] [PubMed] [Google Scholar]
  28. Reichow B, Barton EE, Boyd BA, Hume K. (2014). Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD): A systematic review. Campbell Systematic Reviews. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rogers SJ, Estes A, Lord C, Vismara L, Winter J, Fitzpatrick A, & Dawson G (2012). Effects of a brief Early Start Denver Model (ESDM)–based parent intervention on toddlers at risk for Autism Spectrum Disorders: a randomized controlled trial. Journal of the American Academy of Child and Adolescent Psychiatry, 51, 1052–1065. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Ruble LA, Heflinger CA, Renfrew JW, & Saunders RC (2005). Access and service use by children with autism spectrum disorders in Medicaid managed care. Journal of Autism and Developmental Disorders, 35(1), 3–13. [DOI] [PubMed] [Google Scholar]
  31. Sallows GO, & Graupner TD (2005). Intensive behavioral treatment for children with autism: Four-year outcome and predictors. American Journal on Mental Retardation, 110, 6, 417–438. [DOI] [PubMed] [Google Scholar]
  32. Siklos S, & Kerns KA (2007). Assessing the diagnostic experiences of a small sample of parents of children with autism spectrum disorders. Research in Developmental Disabilities, 28(1), 9–22. [DOI] [PubMed] [Google Scholar]
  33. Skiba RJ, Poloni-Staudinger L, Gallini S, Simmons AB, & Feggins-Azziz R (2006). Disparate access: The disproportionality of African American students with disabilities across educational environments. Exceptional Children, 72(4), 411–424. [Google Scholar]
  34. Smith T, Groen AD, & Wynn JW (2000). Randomized trial of intensive early intervention for children with pervasive developmental disorder. American Journal on Mental Retardation, 105, 4, 269–285. [DOI] [PubMed] [Google Scholar]
  35. Smith T, Klorman R, & Mruzek DW (2015). Predicting outcome of community-based early intensive behavioral intervention for children with autism. Journal of Abnormal Child Psychology, 43(7), 1271–1282. [DOI] [PubMed] [Google Scholar]
  36. Spreckley M, & Boyd R (2009). Efficacy of applied behavioral intervention in preschool children with autism for improving cognitive, language, and adaptive behavior: A systematic review and meta-analysis. The Journal of Pediatrics, 154, 338–344. [DOI] [PubMed] [Google Scholar]
  37. Stahmer AC (2007). The basic structure of community early intervention programs for children with autism: Provider descriptions. Journal of Autism and Developmental Disorders, 37(7), 1344–1354. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Stahmer AC, Dababnah S, & Rieth SR (2019). Considerations in implementing evidence based early autism spectrum disorder interventions in community settings. Pediatric medicine,2, 18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Thomas KC, Ellis AR, McLaurin C, Daniels J, & Morrissey JP (2007). Access to care for autism-related services. Journal for Autism and Developmental Disorders, 37, 1902–1912 [DOI] [PubMed] [Google Scholar]
  40. Thomas P, Zahorodny W, Peng B, Kim S, Jani N, Halperin W, & Brimacombe M (2012). The association of autism diagnosis with socioeconomic status. Autism, 16(2), 1–13. [DOI] [PubMed] [Google Scholar]
  41. Tregnago MK, & Cheak-Zamora NC (2012). Systematic review of disparities in health care for individuals with autism spectrum disorders in the United States. Research in Autism Spectrum Disorders, 6(3), 1023–1031. [Google Scholar]
  42. U.S. Department of Education, Office of Special Education and Rehabilitative Services, Office of Special Education Programs (2020). 41st Annual Report to Congress on the Implementation of the Individuals with Disabilities Education Act, 2019. Retrieved from https://www2.ed.gov/about/reports/annual/osep/2019/parts-b-c/41st-arc-for-idea.pdf [Google Scholar]
  43. Virués-Ortega J (2010). Applied behavior analytic intervention for autism in early childhood: Meta-analysis, meta-regression and dose-response meta-analysis of multiple outcomes. Clinical Psychology Review, 30, 387–399 [DOI] [PubMed] [Google Scholar]
  44. Vivanti G, Dissanayake C, & Victorian ASELCC Team. (2016). Outcome for children receiving the early start denver model before and after 48 months. Journal of Autism and Developmental Disorders, 46(7), 2441–2449. [DOI] [PubMed] [Google Scholar]
  45. Warren Z, McPheeters ML, Sathe N, Foss-Feig JH, Glasser A, & Veenstra VanderWeele J (2011). A systematic review of early intensive intervention. Pediatrics, 127, e1303–e1311. [DOI] [PubMed] [Google Scholar]
  46. Wiggins LD, Baio J, & Rice C (2006). Examination of the time between first evaluation and first autism spectrum diagnosis in a population- based sample. Journal Developmental and Behavioral Pediatrics, 27 (2), S79–S87. [DOI] [PubMed] [Google Scholar]
  47. Wise MD, Little AA, Holliman JB, Wise PH, & Wang CJ (2010). Can state early intervention programs meet the increased demand of children suspected of having autism spectrum disorders? Journal of Developmental & Behavioral Pediatrics, 31(6), 469–476. [DOI] [PubMed] [Google Scholar]
  48. Yingling ME, Hock RM, & Bell BA (2018). Time-lag between diagnosis of Autism spectrum disorder and onset of publicly-funded early intensive behavioral intervention: Do race–ethnicity and neighborhood matter? Journal of autism and developmental disorders, 48(2), 561–571. [DOI] [PubMed] [Google Scholar]
  49. Yingling ME, Bell BA, & Hock RM (2019). Treatment utilization trajectories among children with autism spectrum Disorder: Differences by race-ethnicity and neighborhood. Journal of Autism and Developmental Disorders, 1–11. [DOI] [PubMed] [Google Scholar]
  50. Yingling ME, Hock RM, Cohen AP, & McCaslin EM (2018). Parent perceived challenges to treatment utilization in a publicly funded early intensive behavioral intervention program for children with autism spectrum disorder. International Journal of Developmental Disabilities, 64(4–5), 271–281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zuckerman KE, Lindly OJ, Sinche BK (2015). Parental concerns, provider response, and timeliness of autism spectrum disorder diagnosis. Journal of Pediatrics, 166,1431–1439. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES