Missed Early Intervention Opportunities for Children With Autism Spectrum Disorder

ABSTRACT

Early intervention (EI) is essential for improving developmental outcomes in children with autism spectrum disorder (ASD). However, participation during the critical neurodevelopmental (0–3 years) period remains a challenge. To identify those factors associated with the participation of preschool children with ASD in EI before age 2, this study uses recent cross‐sectional data from the 2021 to 2023 National Survey of Children's Health (NSCH). Binary logistical regression analysis was conducted to identify factors associated with EI receipt before age 2, including age of ASD diagnosis, socioeconomic status (SES), race, ASD severity, biological sex, birth weight, and diagnosis of early indicators of future developmental delay (DD). The findings revealed that only 15% of preschool children with ASD received EI before age 2. Significant predictors of timely participation in Part C EI included an ASD diagnosis before age 2, higher SES, and lower birth weight. Findings also revealed that 15.5% of children who did not receive timely EI had severe ASD symptoms, highlighting the critical need to improve EI participation for these children. Potential solutions discussed include expanding the definition of DD, increasing the number of states recognizing low birth weight as a Part C EI diagnosis, reducing barriers to Part C EI participation for disadvantaged families, and developing more effective tools to detect ASD and DD earlier in development.

Summary.

• In this study, a very small proportion of preschool children with ASD had received early intervention (EI) services by age 2.

• Children with an ASD diagnosis before age 2, higher family income, and lower birth weight were more likely to participate in EI by age 2.

• These results highlight the need to increase EI participation for children with ASD, particularly for disadvantaged families.

1. Introduction

Autism spectrum disorder (ASD) is one of the most common neurodevelopmental disorders in the United States, with an estimated prevalence of 1 in 36 children (Maenner et al. 2023). Early diagnosis and specialized intervention are crucial, as they significantly improve outcomes for children and their families (Dawson et al. 2010; Fuller and Kaiser 2020; Rogers et al. 2019). In particular, early intervention (EI) before age 2 offers the best opportunity to modify developmental trajectories, as peak brain growth occurs during this period (Bethlehem et al. 2022) (Figure 1).

FIGURE 1.

Hypothesized motor development trajectories: The neurotypical trajectory (blue dashed line), the slowed trajectory in children with ASD (red line), and the potential impact of intensive early intervention (EI) (green line) in modifying ASD‐related delays. Based on brain developmental trajectories (Bethlehem et al. 2022), the optimal window for EI is suggested to be between birth and age 2.

A significant public health challenge is the early identification of children with ASD and the prompt delivery of intervention within the optimal EI window. While reliable diagnosis of ASD is possible after age 18–24 months (Dawson et al. 2023), the average age of ASD diagnosis is 5 years (van't Hof et al. 2021). Later diagnosis of ASD (after 18 months) has been associated with more severe developmental delays (DDs) in children with ASD, supporting the benefits of early detection and timely receipt of EI (Miller et al. 2021). Since genetic testing alone cannot establish an ASD diagnosis, clinical in‐person observation by a team of trained medical professionals is required, resulting in prolonged wait times and delayed intervention (Hayes et al. 2022). Further, while an early ASD diagnosis may provide eligibility for federally funded EI services (Part C of the Individuals with Disabilities Education Act [IDEA]), many states require that the child shows significant DDs to be eligible for EI services. Diagnosed conditions with a high likelihood of future DD (e.g., Down syndrome, deafness, and blindness) may also qualify a child for EI; however, these qualifying conditions differ widely by state (Barger et al. 2019) and often the presence of “significant” DD is a requirement to be eligible for Part C EI services. Therefore, most children with ASD in Part C EI services likely qualify due to an early ASD diagnosis (< 24 months), a genetic condition at birth, low birth weight, congenital vision or hearing impairment, or significant DDs (e.g., speech or motor) identified by parents or a primary care physician (Worley et al. 2011).

When a toddler is suspected of DD, a referral is made by the parent or the child's primary care physician to request an assessment by the Individual Family Service Plan (IFSP) team. The multidisciplinary IFSP team assesses the child's development across multiple domains (e.g., motor, adaptive, social–emotional, communication, cognitive). If the child shows “significant” DD in any domain, meeting the state‐specific cut‐off score (e.g., 25%–50% delay in one or more developmental areas), they are eligible for an IFSP, a special education plan that is developed by the IFSP team with the family (Figure 2). The IFSP is a “living document” that is updated every 6 months and includes a plan to reach specific goals for the child created by the IFSP team with the families. The delivery of EI services may involve a parent‐coaching model (i.e., the therapist teaches the parent strategies to meet IFSP goals) or direct intervention provided by the therapist to the child in their natural environment (e.g., at‐home or in‐daycare).

FIGURE 2.

Schematic showing the process to assess eligibility for federally funded early intervention (EI) services.

There are multiple reasons why a child with undiagnosed autism and their family would not participate in Part C EI. These possible reasons may include:

1. The child was assessed by the IFSP team and determined to be eligible, but the family refused services.

2. The child was assessed by the IFSP team, but is not determined to be eligible for Part C.

   • DDs are not severe enough OR the child does not have a qualifying condition.

3. The child was not assessed by the IFSP team.

Understanding how many young children with ASD miss out on EI opportunities during the most sensitive period of neurodevelopment, and factors underlying this missed opportunity is crucial. Additionally, identifying the factors underlying Part C EI participation before age 2 may help identify policy changes to address these shortfalls. The need for improving EI participation for children with ASD is highlighted by the finding that only 50% of children with ASD receive EI services before age 3, with predictors of EI service receipt including more severe ASD symptoms, non‐Hispanic White race, and higher socioeconomic status (SES) (Mendez et al. 2024; Portillo et al. 2024; Shenouda et al. 2022). These disparities suggest that disadvantaged groups may have lower EI participation, leading to missed opportunities and worse long‐term outcomes. Identifying predictors of Part C EI participation before age 2 can help inform policy changes to improve access. Certain conditions and risk factors detected at or near birth—such as low birth weight (< 2500 g), deafness, blindness, and genetic disorders (e.g., Down syndrome)—may increase the likelihood of EI receipt, as these conditions, depending on the state, may qualify children for Part C EI services and lead to increased medical monitoring, facilitating earlier identification and intervention.

This study aims to identify factors influencing Part C EI participation before age 2 in a nationally representative sample of children with ASD using recent National Survey of Children's Health (NSCH, 2021–2023) data. Establishing a baseline for Part C EI participation and its key predictors will help inform future research on the impact of parent education, policy, screening, and assessment in improving EI access for children with ASD. In addition to estimating the proportion of children with ASD who receive Part C EI, this study examines the strongest predictors of timely participation, including age of ASD diagnosis, ASD severity, SES, race, biological sex, and early developmental risk factors such as low birth weight and other conditions detectable at birth (e.g., Down syndrome, deafness, and blindness). It is hypothesized that (1) only a small proportion of preschool children with ASD received Part C EI before age 2 and (2) participation before age 2 will be associated with earlier ASD diagnosis (< 2 years), higher SES, White race, greater ASD severity, low birth weight, and other early diagnostic indicators of future DD.

2. Methods

This study used data from the NSCH for the years 2021, 2022, and 2023 (Health Resources and Services Administration and Maternal and Child Health Bureau 2021, 2022, 2023). Data were extracted to create a subset of 478 preschool‐aged children (age 3–5 years) with a current diagnosis of ASD, a current DD, and a current special education plan (i.e., IFSP/IEP). Responses of “Yes” to all the following survey questions were used to select the cohort for analysis:

• Age of child (coded as SC_AGE_YEARS): “How old is this child? [0–17].”

• Autism diagnosis (coded as K2Q35A): “Has a doctor or other health care provider EVER told you that this child has Autism or ASD? Include diagnoses of Asperger's Disorder or Pervasive Developmental Disorder (PDD).” [Yes, No]

  • ○
    Current autism diagnosis (coded as K2Q35B): “If yes, does this child CURRENTLY have the condition?” [Yes, No]
• DD (K2Q36A): “Has a doctor, other health care provider, or educator EVER told you that this child has DD? Examples of educators are teachers and school nurses.” [Yes, No]

  • ○
    Current DD (coded as K2Q36B): “If yes, does this child CURRENTLY have the condition?” [Yes, No]
• Special Education Plan (coded as K6Q15): “Has this child EVER had a special education or EI plan? Children receiving these services often have an Individualized Family Service Plan or Individualized Education Plan.” [Yes, No]

  • ○
    Current Special Education Plan (coded as SESCURRSVC): “Is this child CURRENTLY receiving services under one of these plans?” [Yes, No]

The primary outcome variable was whether children had first received their first special education plan before the age of 2 years or later. The age of initial receipt of the special education plan was selected as a proxy measure of the timing of intervention receipt. Several predictor and control variables were included to examine factors associated with receiving timely EI services:

• Age of First Special Education Plan (coded as SESPLANYR): “How old was this child at the time of the FIRST plan?” [0–16 or older]. (Primary outcome variable)

• Age of Autism Diagnosis (coded as K2Q35A_1_YEARS): “How old was this child when a doctor or other health care provider FIRST told you that they had Autism, ASD, Asperger's Disorder or PDD?” [1–15]

• Birth Weight (coded as BIRTHWT): Re‐categorized as low birthweight (< 2500 g) or normal birthweight. (Reference = Normal)

• Autism Severity (coded as K2Q35C): “If yes [current ASD diagnosis], is it: 1=Mild, 2=Moderate, 3=Severe” (Reference = Mild)

• Family Poverty Ratio (coded as FPL): The NSCH survey asks respondents about household income. This number is not included as part of the public dataset but is included within the aggregate FPL variable. This variable was re‐categorized as ≤ 200% of the federal poverty level vs. > 200%. (Reference ≤ 200% FPL)

• Race (coded as SC_RACE_R): The NSCH survey asks respondents about the race of the child. This variable was re‐categorized as White or non‐White. (Reference = Non‐White)

• Sex (coded as SC_SEX): Categorized as male or female. (Reference = Female)

• Any Early Indicators of Future DD: Categorized as Yes or No. (Reference = No)

  • ○
    Down syndrome (coded as DOWNSYN): “Has a doctor or other health care provider EVER told you that this child has Down Syndrome?” [Yes, No]
  • ○
    Deafness (coded as K2Q43B): “Does this child have any of the following? Deafness or problems with hearing.” [Yes, No]
  • ○
    Blindness (coded as BLINDNESS): “Does this child have any of the following? Blindness or problems with seeing, even when wearing glasses.” [Yes, No]

Binary logistic regression was performed in RStudio (version 2024.04.2) using the glm() function with a binomial family to model the odds of receiving EI before age 2. Unadjusted odds ratios (ORs) were estimated using separate logistic models for each predictor, while adjusted odds ratios (AORs) and 95% confidence intervals were derived from the full model, which included ASD severity, Family Poverty Ratio, race, birth weight, sex, early diagnostic indicators of future DD, and ASD diagnosis before age 2. A likelihood ratio test compared the null model (intercept only) to the full model to assess model significance (Harris 2021). Only complete cases were analyzed, and participants with imputed data for Family Poverty Ratio, race, or sex were excluded.

3. Results

The study included NSCH 2021–2023 data from 478 children aged 3–5 years with a current ASD diagnosis, DD, and a special education plan (i.e., IEP). Of these, only 72 out of 478 preschool children with ASD (15%) had received EI services by age 2. The likelihood ratio test showed that adding ASD severity, FPL, race, birth weight, sex, early DD indicators, and ASD diagnosis age significantly improves model fit over the null model (χ ² (8) = 23.81, p = 0.0025), suggesting these factors collectively explain EI age classification. The binary logistic regression model revealed several significant predictors of receiving EI services by age 2 that included having an ASD diagnosis prior to age 2 (AOR = 3.25, 95% CI = [1.41, 7.15], p = 0.003), low birth weight (< 2500 g) (AOR = 2.03, 95% CI = [1.04, 3.83], p = 0.03), and higher SES (i.e., family poverty index > 200%) (AOR = 2.05, 95% CI = [1.12, 3.95], p = 0.02) (Table 1 and Figure 3). Trend‐level predictors included severe (vs. mild) ASD symptoms (AOR = 1.93, 95% CI = [0.87, 4.24], p = 0.09) and diagnosis of an early indicator of future DD (i.e., Down syndrome, deafness, blindness) (AOR = 2.08, 95% CI = [0.81, 4.88], p = 0.10). White race (AOR = 1.41, 95% CI = [0.78, 2.63], p = 0.26) and male sex (AOR = 0.89, 95% CI = [0.48, 1.70], p = 0.71) were not significant predictors of EI receipt prior to age 2. These findings highlight the influence of an ASD diagnosis before age 2, higher SES, and lower birth weight on the likelihood of participating in Part C EI before age 2.

TABLE 1.

Binary logistic regression results showing predictors of early intervention (EI) participation before age 2 among preschool‐aged children with ASD.

	Special education plan < 2 years old (n = 72)	Special education plan ≥ 2 years old (n = 406)	OR (95% CI)	AOR (95% CI)
			p	p
Autism diagnosis < 2 years old
No	61 (84.72%)	385 (94.83%)	Reference	Reference
Yes	11 (15.28%)	21 (5.17%)	3.30 (1.47–7.07) * p = 0.002	3.25 (1.41–7.15) * p = 0.003
Autism severity
Mild	20 (27.78%)	141 (34.73%)	Reference	Reference
Moderate	38 (52.78%)	202 (49.75%)	1.32 (0.74–2.41) p = 0.34	1.45 (0.80–2.71) p = 0.22
Severe	14 (19.44%)	63 (15.52%)	1.56 (0.73–3.28) p = 0.23	1.93 (0.87–4.24) † p = 0.09
Race
Non‐White	17 (23.61%)	136 (33.50%)	Reference	Reference
White	55 (76.39%)	270 (66.50%)	1.62 (0.92–2.99) p = 0.09	1.41 (0.78–2.63) p = 0.26
Birth weight
Normal (≥ 2500 g)	56 (77.78%)	350 (86.21%)	Reference	Reference
Low (< 2500 g)	16 (22.22%)	56 (13.79%)	1.78 (0.93–3.27) † p = 0.06	2.03 (1.04–3.83) * p = 0.03
Family poverty ratio
≤ 200% Poverty level	16 (22.22%)	143 (35.22%)	Reference	Reference
> 200% Poverty level	56 (77.78%)	263 (64.78%)	1.90 (1.07–3.53) * p = 0.03	2.05 (1.12–3.95) * p = 0.02
Sex
Female	16 (22.22%)	84 (20.69%)	Reference	Reference
Male	56 (77.78%)	322 (79.31%)	0.91 (0.50–1.71) p = 0.76	0.89 (0.48–1.70) p = 0.71
Early indicator of future DD
No	64 (88.89%)	385 (94.83%)	Reference	Reference
Yes	8 (11.11%) –Down syndrome (1) –Deafness (3) –Blindness (5)	21 (5.17%) –Down syndrome (2) –Deafness (14) –Blindness (6)	2.29 (0.92–5.21) † p = 0.05	2.08 (0.81–4.88) † p = 0.10

Note: Odds ratios (ORs) and adjusted odds ratios (AORs) with 95% confidence intervals (CIs) are reported. Bold indicates p < 0.05.

Abbreviation: DD = developmental delay.

^† Trend‐level predictor (p ≤ 0.1).

^*Significant predictor (p < 0.05).

FIGURE 3.

Adjusted odds ratios (AORs) of preschool children (aged 3–5 years) with ASD receiving early intervention services before 2 years of age and ranked by magnitude. **p < 0.005; *p < 0.05.

4. Discussion

This study examined factors influencing EI participation for preschool‐aged children with ASD. Despite the well‐documented benefits of EI (Dawson et al. 2010; Fuller and Kaiser 2020; Rogers et al. 2019), only 15% of preschoolers with ASD received EI services before age 2. The strongest predictor of timely EI participation was an ASD diagnosis before age 2, which increased the likelihood of participation more than threefold. However, given that ASD screening typically begins around 18 months, opportunities for dramatically increasing early EI participation remain constrained under current diagnostic timelines.

4.1. Timing of ASD Diagnosis and EI Participation

The timing of ASD diagnosis is a fundamental limitation to early EI participation. ASD symptoms often emerge gradually, and diagnosis before age 2 remains uncommon (Dawson et al. 2023; van't Hof et al. 2021). This study found that only 6.7% of preschoolers with ASD were diagnosed before age 2, reinforcing the challenge of significantly increasing EI rates within this age range. Therefore, most children receiving EI before age 2 qualified due to other developmental concerns rather than an ASD diagnosis, underscoring the importance of broadening DD eligibility criteria to support children at risk for ASD who may not yet meet full diagnostic criteria.

4.2. Socioeconomic and Structural Barriers to EI Participation

Consistent with prior research (Mendez et al. 2024; Portillo et al. 2024; Shenouda et al. 2022), this study found that children from higher‐income families were significantly more likely to receive EI before age 2. Lower‐income families may face challenges in navigating complex EI systems, attending required sessions, or accessing providers due to workforce shortages (Corr et al. 2016; U.S. Government Accountability Office 2023). While socioeconomic disparities impact service access, addressing these disparities alone is unlikely to substantially increase overall EI participation before age 2, given the constraints of current screening and eligibility frameworks.

4.3. Low Birth Weight as a Predictor of EI Participation

Low birth weight was also a significant predictor of EI participation before age 2, consistent with its association with increased neurodevelopmental risk (Scharf et al. 2016). In states where low birth weight qualifies for Part C EI, affected children may benefit from earlier identification and intervention. However, only 38% of states include low birth weight as a qualifying condition (Barger et al. 2019), and in this study, only 22% of children with ASD and low birth weight received EI before age 2. Expanding automatic EI eligibility for children with low birth weight may provide an opportunity to modestly increase early EI participation.

4.4. Policy Considerations

Efforts to increase EI participation before age 2 should consider practical policy changes rather than assuming that dramatically higher rates are achievable under current diagnostic and screening practices. Expanding eligibility to include broader DD risk factors, increasing coordination between primary care providers and EI programs, and addressing workforce shortages may help improve service access (U.S. Government Accountability Office 2023). However, major increases in EI participation before age 2 would likely require fundamental shifts in ASD screening timelines and eligibility criteria beyond those currently in place.

4.5. Limitations

The limitations of this study include the use of cross‐sectional data from the NSCH survey from 2021 to 2023, which restricts causal inferences about predictors of EI receipt among children with ASD. The NSCH includes only a limited set of conditions detectable at or near birth (e.g., Down syndrome, deafness, and blindness), potentially excluding other early‐diagnosed conditions (e.g., genetic syndromes with detectable features at birth) that may influence EI participation through increased medical monitoring. While the NSCH provides data on when special education plans (i.e., IFSP/IEP) were initiated, it lacks information on the quality, intensity, or type of intervention, which are critical for understanding outcomes among children with ASD. Parent‐reported data may also be prone to recall bias, particularly regarding the timing of EI initiation. The question on current ASD symptom severity may also be influenced by subjective reports, highlighting the need for a more objective evaluation of ASD severity, especially for children who did not participate in Part C EI before age 2. Additionally, the exclusion of children with imputed data for key variables (e.g., Family Poverty Ratio, race, and sex) improves estimate reliability but limits generalizability. Further, race was categorized as White vs. non‐White to maintain sufficient sample sizes for analysis, but this simplification may obscure nuanced racial differences in EI access. Finally, variability in state‐specific EI eligibility criteria further complicates findings, as these criteria influence which children qualify for Part C EI services.

4.6. Conclusion

This study provides a national estimate of Part C EI participation among preschool‐aged children with ASD, finding that only 15% received services before age 2. Key predictors of timely participation included an early ASD diagnosis, higher SES, and low birth weight. These findings highlight structural barriers to early EI access, including restrictive eligibility criteria and diagnostic timing constraints. While addressing socioeconomic disparities and improving service coordination could enhance EI access, substantially increasing EI participation before age 2 remains challenging given current screening and diagnostic frameworks. Future efforts should focus on refining eligibility policies and developing earlier screening tools while ensuring equitable access to services for all children with ASD.

Ethics Statement

This research used de‐identified, publicly available data from the National Survey of Children's Health (NSCH) and was exempt from IRB approval.

Consent

This study used de‐identified, publicly available data from the National Survey of Children's Health (NSCH).

Conflicts of Interest

The author declares no conflicts of interest.

Data Availability Statement

The data sets analyzed in this study (National Survey of Children's Health 2021–2023) are available online on the US Census Bureau Website: https://www.census.gov/programs‐surveys/nsch/data/datasets.html.