Secondary Logo

Journal Logo

Original Articles

Developmental Outcomes of Children Served in a Part C Early Intervention Program

Elbaum, Batya PhD; Celimli-Aksoy, Seniz PhD

Author Information
doi: 10.1097/IYC.0000000000000205
  • Free
  • ISEI Article


THERE is broad consensus on the benefits of early intervention for infants and toddlers with developmental delays or conditions associated with developmental risk (Guralnick, 2011; Sandler, 2019). Numerous studies have demonstrated the positive effects of early intervention for children with a variety of diagnoses or delays, including infants born preterm (Spittle, Orton, Anderson, Boyd, & Doyle, 2015), late talkers (Kruythoff-Broekman, Wiefferink, Rieffe, & Uilenburg, 2019), children with autism spectrum disorder (ASD; Dawson et al., 2010; Fuller & Kaiser, 2020; Sacrey, Bennett, & Zwaigenbaum, 2015; Zwaigenbaum et al., 2015), and children with Down syndrome (DS; O'Toole, Lee, Gibbon, van Bysterveldt, & Hart, 2018).

Given the growing research base on early interventions to improve children's developmental outcomes, it is notable that few studies to date have evaluated outcomes of children's participation in services provided through Part C of the Individuals with Disabilities Education Act (Individuals with Disabilities Education Act [IDEA], 2004). In the United States, Part C Early Intervention (EI) programs are the frontline providers of development services to infants and toddlers aged birth to 3 years who have, or are at risk for, developmental delays or disabilities. In 2019, the point-in-time count of children enrolled in Part C EI programs was 427,234, representing 3.70% of children aged birth to 3 years in the U.S. population (U.S. Department of Education, n.d.).

The evaluation of outcomes for children receiving intervention through Part C EI (hereinafter referred to simply as EI) is of critical importance for multiple groups of stakeholders (Eigsti, Chandler, Robinson, & Bodkin, 2010). Policy makers, taxpayers, and families of children eligible for EI services need evidence that public resources are being used effectively. As well, EI service providers and program administrators need outcome data, in combination with other types of data, to guide program improvement efforts (Shah, 2019). Studies of EI are sparse, however, and rarely address the heterogeneity of developmental challenges that characterize the population of children served. For example, Eigsti et al. (2010) conducted a prospective longitudinal study of developmental progress in 34 children who had enrolled in EI services in the Denver area. The children had a variety of diagnosed conditions and delays but were not selected specifically to be representative of the population of children served in the program. In contrast, Woodman, Demers, Crossman, Warfield, and Hauser-Cram (2018), using data from the Early Intervention Collaborative Study, specifically selected for analysis a sample of 162 children identified as having a motor impairment, a diagnosis of DS, or a developmental delay of unknown etiology, as these diagnostic categories were posited to represent the most common types of disability in children enrolled in EI (in Massachusetts and New Hampshire) between 1985 and 1987 (Shonkoff, Hauser-Cram, Krauss, & Upshur, 1992). Data from the Vineland Adaptive Behavior Scales-Interview (Sparrow, Balla, & Cicchetti, 1984) indicated that when the children were 3 years of age (coinciding with transition out of EI), their mean age-equivalent scores were 1.94 in communication, 1.61 in socialization, and 1.65 in daily living skills. Thus, on average, these children were a year or more delayed, compared with typically developing toddlers, in these three areas of functioning. The data were not reported separately for the three subgroups of children, however, leaving unexamined the extent to which children's outcomes may have differed in relation to the nature of their developmental challenges.

Results from the National Early Childhood Longitudinal Study (Scarborough, Hebbeler, Spiker, & Simeonsson, 2011) provide a more nuanced picture of outcomes for the EI population. The study was based on a nationally representative sample of children who entered EI in 1997–1998. Scarborough et al. (2011) reported that 18% of children with a diagnosed condition, 31% of those with a developmental delay, 40% of children with an “at-risk” condition, and 50% of those with a delay in speech or language only reached all measured developmental milestones by 5 years of age. These results underscore the variability in outcomes by children's broad category of eligibility for EI.

The present study was designed to provide a finer-grained picture of EI outcomes for the diverse population of children served. In this regard, the study's focus is similar to that of Elbaum and Celimli-Aksoy (2017), who investigated outcomes of EI for four subgroups of children, empirically defined on the basis of mixture modeling. The present study, in contrast, takes as its starting point a classification of children into subgroups based on their primary diagnosis (e.g., prematurity vs. a neurological condition) or, for children without a specific diagnosis, their primary area of delay (e.g., motor, cognitive). Accordingly, we assigned children to a diagnostically or developmentally defined subgroup based on the child-level International Classification of Diseases, Ninth Edition (ICD-9; World Health Organization, 1978) codes that were entered into the EI program's database when children were determined eligible for services.


The data set used for analysis included each child's domain-level scores on the standardized developmental assessment administered when the child enrolled in the EI program and again when the child transitioned out of the program. Using these data, we addressed the following specific research questions:

Research question no. 1: To what extent did the defined subgroups of children present distinctive profiles of functioning across the five assessed developmental domains (cognitive, motor, adaptive, communication, and personal–social) on entry into the EI program?

Research question no. 2: To what extent did children in the defined subgroups demonstrate (a) significant change in functioning in each developmental domain, between the time they entered and exited the EI program, and (b) distinctive patterns of change across domains?

Research question no. 3: What percentage of children in each subgroup showed improved functioning in each developmental domain?

Research question no. 4: What percentage of children in each subgroup was functioning within age expectations in all five developmental domains at the time they exited the EI program?

Analytic approach

Given that any grouping scheme might be considered somewhat arbitrary, we first tested for the distinctiveness of the predefined subgroups. We did this by statistically comparing the identified subgroups using children's domain-level standard scores from the developmental assessment administered as part of their eligibility evaluation for EI. Once we confirmed the distinctiveness of the identified subgroups, we proceeded to examine subgroup variation in developmental progress and outcomes.

We took two approaches to examining developmental progress over time (cf. Eigsti et al., 2010). First, for each subgroup, we examined the statistical significance and effect sizes associated with mean change in standard scores from program entry to exit in each of the five assessed developmental domains. Second, we applied the Reliable Change Index, a metric widely used in clinical research, to derive robust estimates of the percentage of children in each subgroup who made statistically meaningful developmental gains in each domain. To examine variation in children's age-expected outcomes, we calculated the percentage of children in each subgroup who, at the time they exited EI, demonstrated a level of functioning that put them on par with typically developing children in all measured domains.

To provide additional context for these analyses, we also present data on the core EI services provided through this EI program: speech–language therapy (ST); occupational therapy (OT); physical therapy (PT); and infant toddler developmental specialist (ITDS) intervention.



The EI program providing the data for this study is part of a clinical and research center within a university-based medical school. The program serves a large portion of Miami-Dade County, Florida. In 2018, the population of this urban/suburban area was 69.1% Hispanic or Latino, 15.5% Black or African American, and 12.9% White alone (Data USA, n.d.). Table 1 displays demographic data for this sample, which included children exiting the EI program between July 2009 and June 2013. Only children who participated in the EI program for 6 months or more were included in the data set, as children who received fewer than 6 months of services were exempt from the state's exit assessment requirement. At the time of their entry assessment, children in this sample ranged in age from newborn to 30.5 months (M = 18.9 months).

Table 1. - Demographic Information (N = 1,337)
n %
Female 428 32.0
Male 909 68.0
Medicaid eligible
Yes 1,050 78.5
No 287 21.5
Hispanic/Latino 698 52.2
Black or African American only 368 27.5
White only 91 6.8
Asian only 5 0.4
Multiracial 4 0.3
Unknown federal race category 171 12.8
Diagnostic category
Neurological condition 107 8.0
Genetic or metabolic condition 93 7.0
Extreme prematurity 90 6.7
ASD concern 248 18.5
Global Developmental Delay 270 20.2
Delay in cognitive development 69 5.2
Delay in motor development 45 3.4
Delay in adaptive development 32 2.4
Delay in language development 383 28.6
Note. ASD = autism spectrum disorder.

Children received varying types, durations, and intensity of EI services. Almost all children (93.1%) received intervention from an ITDS. As defined by the EI program, an ITDS is:

a non-licensed provider of Early Intervention Services ... [who] focuses on infant/toddler development and ways to promote development and learning, including designing learning environments and activities to promote development across all domains. The ITDS, in consultation with other early intervention providers on the child and family's team, assists the family in understanding the special needs of the child and enhancing the child's development. (Children's Medical Services, n.d.)

Fewer children received services provided by a licensed disciplinary-based professional: 31.4% received ST, 28.6% received OT, and 16.8% received PT. Overall, 54.2% of children received only ITDS services; 22.3% received one type of disciplinary-based service (ST, OT, or PT); 16.1% received two types; and 7.5% received all three types.

Table 2 displays information on the intensity of services received by children in each of the nine subgroups. We report intensity of services (hours/month) rather than overall amount of services to take into account variation in time spent in EI. As seen in the table, children who exited this local EI program received an average of 5.43 (SD = 9.35) hours/month of services. Given the skewed distribution of this variable, we also report the median and interquartile range (IQR). For the whole sample, the median hours/month of received services was 3.73, with an IQR of 3.52 (first to third quartiles = 1.81–5.34).

Table 2. - Intensity of EI Services Receiveda
Subgroupb n Minimum Maximum Mean SD Median 1st to 3rd quartiles (Interquartile Range)
Genetic or metabolic condition 93 0.04 71.41 8.56 12.49 5.07 2.21–8.14 (5.93)
ASD concern 248 0.05 84.19 7.46 11.94 5.01 2.88–7.75 (4.87)
Global Developmental Delay 270 0.07 84.05 7.16 12.49 4.12 2.38–6.09 (3.71)
Delay in adaptive development 32 0.14 79.45 5.62 13.63 3.20 1.88–4.68 (2.80)
Delay in language development 383 0.09 53.37 3.82 4.07 3.57 2.00–4.60 (2.60)
Extreme prematurity 90 0.04 75.24 3.56 8.24 1.77 0.98–3.53 (2.54)
Neurological condition 107 0.09 20.13 3.56 3.77 2.48 1.15–4.44 (3.29)
Delay in cognitive development 69 0.03 9.40 2.94 1.95 3.00 1.13–4.40 (3.27)
Delay in motor development 45 0.09 10.96 2.91 2.19 2.20 1.37–4.28 (2.91)
Total 1,337 0.03 84.19 5.43 9.35 3.73 1.82–5.34 (3.52)
Note. ASD = autism spectrum disorder.
aHours/month of any combination of infant toddler developmental specialist intervention, speech therapy, occupational therapy, and/or physical therapy.
bSubgroups are listed in descending order of their mean number of hours/month of services.


Permission to conduct the study was obtained from the Institutional Review Boards of the Florida Department of Health and the researchers' university. The obtained data sets contained no personally identifiable information on any children or families.

Data source

Data for this study were drawn from two databases, one containing children's developmental assessment data and another containing children's diagnostic, demographic, and service use information. Child-level variables from the two databases were linked via a unique program ID assigned to children when they were referred to the EI program for eligibility evaluation.

Classification of children into subgroups

The child-level ICD-9 codes in the database were entered by the clinic's multidisciplinary evaluation team following each child's eligibility assessment. For children without a confirmed medical diagnosis, codes related to developmental delays were assigned on the basis of scores from the Battelle Developmental Inventory, Second Edition (BDI-2; Newborg, 2005), administered as part of each child's initial evaluation for services. If and when indicated, the multidisciplinary team also administered one or more screening tools for autism; for example, the Modified Checklist for Autism in Toddlers (Robins, Fein, Barton, & Green, 2001). If the results of the screening test indicated the need for further evaluation, the child's record was flagged for suspected ASD.

Children enrolling in the EI program fell into two broad categories: those with medical diagnoses and those with a developmental delay of unknown etiology. Children with medical diagnoses (“established conditions”) were classified into one of five subgroups: neurological condition (e.g., quadriplegic cerebral palsy), genetic or metabolic condition (e.g., DS), extreme prematurity (birth weight ≤1,200 gr.), deaf or hearing impaired, or blind or visually impaired. Children for whom an ASD concern was identified at the child's EI clinic visit were classified into a separate subgroup.

Children with developmental delays of unknown etiology were classified into one of six subgroups based on their noted area or areas of delay. Consistent with widely accepted guidelines for a diagnosis of Global Developmental Delay (GDD; Jimenez-Gomez & Standridge, 2014), children with standard scores 2 or more standard deviations below the mean in two or more domains of development were classified as having GDD. Children presenting with a severe delay in no more than one domain were classified on the basis of their primary area of delay: cognitive development, motor development, adaptive development, language development, or social–emotional development.

Of the 12 subgroups defined previously, three with very few children (n ≤ 15) were excluded from the analysis. These were the subgroups of children with hearing impairment (n = 15), vision impairment (n = 7), or a single-domain delay in the social–emotional domain (n = 4). The final data set included 1,337 children classified into nine subgroups (see Table 1).


Developmental assessment

The BDI-2 (Newborg, 2005) was validated for the assessment of developmental functioning in children aged birth through 7 years, 11 months, in five domains: cognitive, motor, adaptive, communication, and personal–social. The assessment consists of a total of 450 items reflecting milestones of development across the age range. Domain-level developmental quotients (DQs) range from 55 to 145, with M = 100 and SD = 15. For children 0–29 months, reported test–retest reliabilities for the domain DQs range from 0.87 to 0.95.

The BDI-2 assessment was administered by members of the EI clinic's multidisciplinary team at the time of each child's eligibility determination and again when they exited the program. All clinical team members were trained in the administration and scoring of the assessment, either by trainers employed by the publisher of the BDI-2 or by the EI lead clinical psychologist, who had participated in a Train-the-Trainer course provided by the publisher.

Statistical approach

To examine the distinctiveness of patterns of developmental functioning across subgroups on children's entry into EI (Research question no. 1), we conducted a one-way multivariate analysis of variance (MANOVA), using subgroup as the independent variable and the five entry DQs as the dependent variables (DVs). The main analysis was followed up with univariate analyses of variance (ANOVAs) using the DQs in each domain.

To investigate children's developmental change in each domain, as well as the distinctiveness of patterns of change across subgroups (research question no. 2), we conducted a doubly MANOVA and a series of follow-up paired t tests for each subgroup, comparing pre- and post-DQs in each domain. Effect sizes were calculated using partial η2 for MANOVA and Cohen's d for paired t tests (Cohen, 1988). A Bonferroni correction was applied for paired t tests to correct for inflated alpha.

To calculate the percentage of children showing reliable improvement in developmental functioning (research question no. 3), we calculated the Reliable Change Index (RCI) for each child in each developmental domain. The RCI was first introduced by Jacobson, Follette, and Revenstorf (1984) and revised by Christensen and Mendoza (1986) as follows: RCI=x2-x1S diff

where x1 represents a child's pretest score, and x2 is the child's posttest score. S diff is the standard error of difference between pre- and posttest scores and is computed as a function of the standard error of measurement, SE: S diff =2SE2SE=s11-r xx

where s1 is the standard deviation of the sample pretest DQ and r is the reliability of the measure. For this study, we used the domain-level reliability coefficients reported in the BDI-2 Examiner's Manual (Newborg, 2005) for children aged 0 through 29 months and applied these to the whole sample. (We did not calculate within-subgroup RCIs as we did not have published estimates of reliabilities for these subgroups.) An RCI value larger than 1.96, or smaller than −1.96 would be considered very unlikely (p < .05) if no real change had occurred between pre- and postassessments. Given this study's focus on the percentage of children who made meaningful gains in developmental functioning, we report only the percentage of children whose RCI was 1.96 or greater.

To determine the percentage of children whose developmental functioning was within age expectations at the time they exited EI (research question no. 4), we defined “within age expectations” as a DQ of 78 or greater (> −1.5 SD) in all five developmental domains.

All statistical analyses were performed using IBM SPSS version 26 (


In this section, we report the results of each analysis by research question, beginning with the question concerning the extent of subgroup variation on entry and then addressing differences across subgroups in developmental gains and normative attainment.

Research question no. 1: Subgroup entry profiles

Box's M value of 619.86 indicated that the assumption of homogeneity of the covariance matrices was violated, F(120, 189644.09) = 5.04, p < .001. We therefore used Pillai's criteria to interpret the result of the MANOVA. For these data, Pillai's trace = 1.10, F(40, 6,640) = 46.74, p < .001, ηp2 = .22, indicating that there was a statistically significant difference across subgroups in their mean domain-level DQs. Slightly more than one-fifth of the variation in children's developmental domain scores was explained by subgroup assignment. Subgroup mean entry DQs by domain are displayed in Table 3.

Table 3. - Entry DQ, DQ Change, and ES of the DQ Change by Subgroup and Developmental Domain
Subgroup Developmental Domain
Cognitive Motor Adaptive Communication Personal–Social
Entry DQ Entry DQ Entry DQ Entry DQ Entry DQ
M (S D) DQ Change ESa M (S D) DQ Change ES M (SD) DQ Change ES M (S D) DQ Change ES M (S D) DQ Change ES
Neurological condition 72.1 (11.7) −0.7* −0.1 71.1 (13.4) 10.81* 0.7 81.5 (14.6) −3.5 −0.2 72.9 (12.0) −1.7 −0.1 80.0 (10.0) −1.9 −0.2
Genetic or metabolic condition 74.6 (12.9) −6.29* −0.5 74.0 (15.3) 3.9 0.2 81.8 (16.8) −6.2 −0.3 74.3 (14.0) −9.74* −0.6 84.2 (10.9) −8.32* −0.7
Extreme prematurity 72.2 (8.9) 4.7 0.4 71.8 (12.4) 17.98* 1.3 80.4 (11.9) 3.4 0.2 68.9 (8.8) 5.0 0.4 83.3 (9.4) −0.7 −0.1
ASD concern 69.6 (8.3) −0.4 0.0 80.8 (12.7) 4.99* 0.4 75.1 (11.5) 2.2 0.2 58.7 (6.1) 3.76* 0.4 76.2 (9.6) −4.06* −0.3
Global Developmental Delay 69.9 (8.8) 3.40* 0.3 77.9 (14.1) 12.60* 0.8 72.5 (12.4) 7.32* 0.5 61.9 (6.4) 9.13* 0.7 79.8 (8.7) −1.4 −0.1
Delay in cognitive development 72.1 (5.2) 7.71* 0.7 93.8 (11.3) 4.0 0.3 90.7 (11.1) −0.3 0.0 80.2 (8.0) 0.8 0.1 91.2 (8.5) −5.44* −0.6
Delay in motor development 83.6 (7.5) −0.69 −0.1 73.2 (9.7) 22.31* 1.4 92.6 (12.3) −4.9 −0.3 84.3 (7.5) 4.4 0.0 88.1 (8.2) −0.9 −0.1
Delay in adaptive development 81.2 (6.2) 1.22 0.1 90.1 (7.9) 6.3 0.5 68.8 (9.4) 14.63* 0.9 76.9 (6.8) 8.22* 0.3 85.9 (6.8) −2.6 −0.2
Delay in language development 80.6 (7.3) −1.27 −0.1 94.1 (9.9) 12.60* 0.2 88.6 (9.7) −0.2 0.0 67.4 (7.8) 9.13* 0.6 89.7 (8.1) −5.91* −0.6
Full sample 74.4 (10.0) 0.5 0.0 82.8 (15.1) 7.63* 0.5 81.1 (13.9) 1.50* 0.1 67.2 (10.7) 4.54* 0.3 83.5 (10.4) −3.89* −0.3
Note. ASD = autism spectrum disorder; DO = developmental quotient; ES = effect size.
aEffect sizes calculated as Cohen's d (Cohen, 1988).
*Statistical significance at p < .001.

Follow-up ANOVAs indicated that there were statistically significant subgroup differences in mean DQs for each domain: cognitive, F(8, 1,328) = 53.68, p < .001, ηp2 = .24; motor, F(8, 1,328) = 85.50, p < .001, ηp2 = .34; adaptive, F(8, 1,328) = 59.62, p < .001, ηp2 = .26; communication, F(8, 1,328) = 111.76, p < .001, ηp2 = .40; and personal–social, F(8, 1,328) = 59.35, p < .001, ηp2 = .26. For example, in the cognitive domain, means for the neurological, genetic–metabolic, extreme prematurity, cognitive delay, GDD, and ASD subgroups were not statistically different from one another (range: M = 69.6 to M = 72.2); however, means for all of these subgroups were statistically significantly different from those of the subgroups defined by a delay in the sole area of language, motor, or adaptive development (range: M = 80.6 to M = 83.6), which did not differ significantly from one another. No two subgroups had statistically equivalent mean DQs in all five domains, indicating that each of the nine subgroups demonstrated a distinct profile, or pattern of developmental strengths and challenges, on entry into the EI program. Figure 1 provides a visual display of subgroup developmental profiles on entry.

Figure 1.:
Subgroup developmental profiles on entry into early intervention. ASD = autism spectrum disorder; DQ = developmental quotient.

Research question no. 2: Subgroup developmental gains

Results of the doubly MANOVA using entry and exit domain-level DQs as the within-subjects DVs indicated that the developmental change by subgroup interaction was statistically significant, Pillai's trace = 0.43, F(40, 6,640) = 15.73, p < .001, ηp2 = .09, indicating that changes between entry and exit DQs differed significantly across subgroups. Follow-up paired t tests were conducted to assess the statistical significance of each difference, using the Bonferroni-corrected Type 1 error rate of p < .001. Mean DQ changes and their associated effect sizes are displayed in Table 3.

Research question no. 3: Subgroup reliable progress

Table 4 displays the percentage of children in each subgroup who made reliable progress in each domain. Overall, approximately 40% of children made reliable gains in the motor and communication domains; slightly more than 20% improved, compared with norms, in the cognitive and adaptive domains; and fewer than 20% made gains in the personal–social domain. As seen in Table 4, within each domain, the percentage of children making reliable gains varied considerably by subgroup.

Table 4. - Percentage of Children Who Made a Reliable Gain by Subgroup and Developmental Domaina
Subgroup Developmental Domain
Cognitive Motor Adaptive Communication Personal–Social
Neurological condition 23.4 53.3 19.6 28.0 27.1
Genetic or metabolic condition 8.6 38.7 19.4 14.0 10.8
Extreme prematurity 38.9 73.3 25.6 42.2 25.6
ASD concern 14.9 34.7 19.8 25.0 19.8
Global Developmental Delay 28.1 56.7 40.7 51.1 21.9
Delay in cognitive development 40.6 26.1 11.6 26.1 11.6
Delay in motor development 24.4 77.8 13.3 31.1 17.8
Delay in adaptive development 21.9 37.5 59.4 43.8 15.6
Delay in language development 16.4 25.1 13.3 48.3 8.9
Full sample 21.7 41.8 22.8 38.3 16.8
Note. ASD = autism spectrum disorder.
aThe percentage in each cell is the percentage of children with a Reliable Change Index of 1.96 or greater.

Research question no. 4: Subgroup attainment of age norms

Overall, 25% of children who received EI were within age expectations (DQ ≥78) in all domains when they exited the EI program. Similar to the results related to reliable gains, the percentage of children exiting within age expectations varied considerably across subgroups, ranging from 10% of children with suspected ASD to 49% of children with a delay in the area of motor development only. Corresponding percentages for the other subgroups were 12% for children with a genetic or metabolic condition, 19% for children with GDD, 22% for children with a neurological condition, 31% for children with extreme prematurity, 35% for children with a delay in language development, and 41% for children with either a cognitive delay or an adaptive delay.

To summarize, the results of our first analyses provided support for the validity of the subgroups we created for purposes of this study. Across all five domains assessed on entry to EI, subgroups accounted for 22% of the variation in children's scores, representing a large effect size (Cohen, 1988). Subgroups accounted for 24%–40% of the variation in individual domain scores. In addition, children in different subgroups displayed unique patterns of development; variation in developmental gains, as indexed by the change by subgroup interaction, accounted for 9% of the variation, considered a medium effect size (Cohen, 1988). Finally, the percentage of children exiting EI who displayed developmental functioning on par with typically developing peers, across all domains, varied markedly by subgroup. From one-third to approximately one-half of children with a delay in a single domain of development showed remediation of their delay by the time they exited EI. Children with more pervasive developmental challenges were far less likely to be developmentally on track by 3 years of age.


This study investigated developmental outcomes of a Part C EI program for clinically defined subgroups of infants and toddlers, using scores from a standardized developmental assessment administered at program entry and exit. The study revealed significant variation in the progress and outcomes of children with different diagnoses or developmental delays.

Subgroup differences in initial developmental profiles and in developmental progress over time

The nine subgroups we studied showed distinctive developmental profiles on entry into the EI program as well as distinctive patterns of developmental progress over time. These results suggest that the focus on subgroups can provide an important perspective on EI that is not captured by results reported in the aggregate for all children exiting the program. This is not to say that the remaining variability within subgroups is not important; indeed, variability in outcomes among children with the same diagnosis is to be expected and is itself the object of considerable developmental research (e.g., Chen et al., 2019; Clark, Barbaro, & Dissanayake, 2017; Spittle et al., 2016; Tsao & Kindelberger, 2009). However, from the standpoint of evaluating outcomes of EI, there are practical as well as statistical advantages to creating a manageable number of subgroups that account for significant variability in the population served.

The specificity of the demonstrated patterns of gains across our identified subgroups sheds new light on the outcomes of EI for children with different developmental challenges. Of children with a single-domain delay, those with a cognitive delay made large gains in cognitive skills; those with a motor delay made large gains in motor functioning; those with a delay in adaptive skills made large gains in adaptive functioning; and those with a language delay made large gains in communication. Even children with GDD, who as a group were among the most severely delayed on entry, made notable gains in the communication and motor domains.

Gains for children with an ASD concern or a medical diagnosis were also worthy of note. Children with an ASD concern showed moderate gains in the motor and communication domains, as well as smaller but still significant gains in adaptive functioning. Children born extremely premature made standard score gains in all but the social–emotional realm, with effect sizes interpreted as moderate to large in the cognitive, communication, and motor domains. Children with neurological diagnoses made significant gains only in the motor domain. Children with genetic or metabolic conditions fell further behind same-age peers in four of five domains and showed no change, relative to typically developing peers, in the fifth.

It bears reiterating that in this study, improvement was defined as an increase in standard scores, indicating a positive shift in functioning compared with typically developing peers. At the same time, we note that in children with developmental disabilities, an age-related decline in standard scores might actually be expected, based on these children's slower rate of development (Tobia, Brigstocke, Hulme, & Snowling, 2018). Emphasizing this point, Will, Caravella, Hahn, Fidler, and Roberts (2018), in their study of changes in the adaptive behavior of infants and toddlers with DS and fragile X syndrome (FXS), note that “Despite nearly all participants with DS or FXS in this study having a history of early intervention, we reported consistent declines in standard adaptive behavior scores across age, with the exception of motor skill acquisition for those with DS” (p. 366). Against this backdrop, gains in standard scores are particularly notable, as they suggest a lessening of the extent of a child's delay in a particular area of functioning.

Subgroup differences in reliable progress

As seen in Table 4, the percentage of children making reliable gains in any given domain tracks with the subgroup effect sizes displayed in Table 3 (Estrada, Ferrer, & Pardo, 2019). Evaluating children's developmental progress in the context of EI is important for multiple constituencies, including families, practitioners, program administrators, policy makers, and researchers. As such, the metrics used to report results—effect sizes, percentages, odds ratios, and so forth—may hold sway differently with different audiences. The RCI metric may be a particularly useful metric for both public reporting and accountability, as it is likely more readily understood by a public audience than are subgroup means and effect sizes. Moreover, from the perspective of estimating a mean population value, dispersion around the mean is “error”; in contrast, the RCI perspective acknowledges and validates individual differences in response to an intervention. This means that even in subgroups that did not demonstrate statistically significant aggregate gains, some children did make significant individual progress, and conversely, even in subgroups with large effect sizes, not all children showed progress. Knowing the estimated percentage of children with a particular diagnosis or delay who make progress in a particular developmental domain can provide information that is particularly important to practitioners and families. In addition, given the documented persistence of developmental delays over time (Wood, Christian, & Sampson, 2018), the percentages reflecting children's reliable gains can be seen as reasonable estimates of the extent to which children's delays can be mitigated through the provision of community-based developmental services.

Subgroup differences in attaining age-expected functioning

At the time they exited the EI program, 25% of children in this sample had standard scores within the typical range (> −1.5 SD) in all five developmental domains. Again, however, there was important variation across subgroups. Nearly one-half of children with a delay limited to the motor domain were on par with their typically developing peers at the time they transitioned out of EI; this was also true of more than 40% of the children with an identified delay in either language or cognitive development (but not both). For children with more pervasive delays, including children with ASD concern, GDD, or genetic/metabolic syndromes, the percentage of children exiting EI on par with typically developing same-age peers was less than 20%.

On the one hand, it is encouraging to note that approximately one out of four children served in this local EI program achieved this outcome. This information is extremely helpful both to policy makers and providers at the next age level (i.e., preschool), as it provides a basis for estimating the percentage of children who will likely continue to need targeted supports in order to maintain or improve their functioning. Outcome indicators framed in relation to norms for typically developing children exist not only for Part C EI and IDEA preschool programs (Early Childhood Technical Assistance Center, 2019) but also for all U.S. children aged 3–5 years (Ghandour et al., 2019). For example, data reported by Ghandour et al. (2019) indicate that in the United States, approximately 42% of children aged 3–5 years are “on track” in four key school readiness domains (early learning skills, self-regulation, social–emotional development, and physical health and motor skills).

On the other hand, comparisons of children's EI outcomes with normative expectations may be less meaningful for families of children with more pervasive developmental challenges. As our results show, only a small percentage of these children (e.g., children with GDD or genetic–metabolic syndromes) attain age-expected functioning across all domains of development by 3 years of age. An alternative approach to evaluating outcomes for these children might be to compare the progress of individual children against trajectories that typify the course of development for children with similar conditions or delays. For example, de Graaf, Levine, Goldstein, and Skotko (2018) used data on thousands of children with DS to create growth curves reflecting the range of normative development for these children. Data such as those reported in this study, including data from other states' administrative databases, could inform the bounds of these trajectories.


EI programs are tasked with serving a heterogeneous population of infants and toddlers with or at risk for developmental delays or disabilities. An important challenge for the field is that of evaluating program outcomes in ways that can inform policy and future research.

With regard to policy, state legislatures are loathe to fund programs that produce weak or uncertain outcomes. The results of our study, while indeterminate in terms of causality, indicate that many children participating in EI show significant developmental progress. The positive results noted for all but one of the subgroups we investigated may be especially compelling. As such, EI programs may benefit greatly, in terms of both public support and legislative leverage, if outcomes are reported for meaningfully defined subgroups of children.

Importantly, it should be emphasized that all EI-eligible children, regardless of their specific diagnosis or likelihood of attaining age norms, benefit from well-designed EI. As underscored by Kuhn and Marvin (2016):

The fact that the payoff in terms of measurable outcomes for children with severe disabilities is smaller than for some children with milder disabilities is not a reason to reduce dosage of early intervention supports and services for children with severe disabilities. (p. 3)

The reporting of EI outcomes by subgroup might also address the challenge of drawing comparisons across states, or among local programs within a state, with regard to program performance. Health systems typically acknowledge that client makeup differs across providers and that this can affect outcomes, irrespective of the quality of care provided (e.g., Kuhlthau, Ferris, & Iezzoni, 2004). Reporting of outcomes for well-defined subgroups could be an effective and relatively straightforward approach to taking into account differences in the case mix (cf. Mesterton et al., 2016) of different EI programs.

Further research is needed to evaluate the generalizability of the pattern of outcomes we found in our study. From the standpoint of service intensity, the hours per month of services received by children in this sample (M = 5.43, median = 3.73) are close to state-level averages reported by state Part C EI coordinators in 2020 (M of state averages = 4.6, median = 3.6; IDEA Infant & Toddler Coordinators Association, 2020). Investigation of the association between intensity of services and children's outcomes is complicated, however. By design, service provision in EI is individualized to children's developmental needs; in part as a consequence of this design, children with greater developmental needs—and often slower rates of development—are provided more intensive services than children with less significant needs (Javalkar & Litt, 2017; Richardson et al., 2019). This pattern appears to hold true, at least to some extent, in the subgroup data we report in this study. As seen in Table 2, the subgroups of children who were least likely to demonstrate age-normative functioning on exit were those that received the greatest intensity of services.

Moreover, the last decade has seen a major shift in the thinking about service intensity, or “dosage” of intervention (Bagnato, Suen, & Fevola, 2011; Kuhn & Marvin, 2016). Billed service hours are still a readily accessible metric that may be associated with outcomes (e.g., Khetani, Richardson, & McManus, 2017; Richardson et al., 2019). But given changes in EI service delivery models (e.g., Dunst & Trivette, 2009; Romano & Schnurr, 2020; Shelden & Rush, 2014), particularly the adoption of models that prioritize family capacity building and center the coaching of families to implement developmentally supportive practices in the context of everyday family routines, measures of parent uptake and implementation of strategies may be a more powerful predictor of improvement in children's developmental and functional outcomes than contact hours with an EI service provider. In line with this view, there has been an increasing focus on parent/family outcomes, such as self-efficacy, as either mediators or moderators of children's outcomes and as meaningful outcomes in their own right (cf. Wainer, Hepburn, & Griffith, 2017).

Finally, further research is needed on the use of different types and timing of outcome measures to evaluate the impact of EI. The outcomes reported in this study were measured at or before 36 months of age. Whether the impact of EI on later outcomes fades, or becomes more accentuated with time, can be determined only through appropriately designed follow-up studies (Litt, Glymour, Hauser-Cram, Hehir, & McCormick, 2018; Spittle et al., 2016). In addition, the EI program that provided data for this study utilized a standardized developmental assessment as the basis for reporting outcomes of services for participating children. However, it has been argued that a sole focus on changes internal to the child represents a narrow view of the role and potential of interventions to improve children's participation in everyday contexts and their overall quality of life (Calder, Ward, Jones, Johnston, & Claessen, 2018; Cunningham et al., 2017). As these researchers note, the framework of the International Classification of Functioning, Disability and Health—Child and Youth (World Health Organization, 2007) may be especially useful in selecting diverse measures to evaluate intervention outcomes.


The results of this study should be interpreted in the light of several limitations. First, this study investigated outcomes of a single local EI program in a single state. Whether outcomes for similarly defined subgroups of children exiting other EI programs would show similar patterns of results is unknown. Second, the database we accessed did not include item-level data from the developmental assessment, preventing us from calculating the reliability of the domain-level scores for this sample of children. Relatedly, no information was available on the reliability of individual assessors. Given that the assessors were not blind to children's participation in EI, it is possible that this may have introduced some bias into children's assessment scores.

A further limitation is related to the use of pre-/postdata. A threat to the interpretation of all pre-/postdesigns is regression to the mean. Although regression artifacts may have affected our results to some extent, it is unlikely that regression alone could be responsible for the overall pattern of results documented in this study. It is nonetheless the case that because our study did not include a no-treatment group, we cannot disentangle changes that are the result of maturation from changes that can be credited to intervention (Christakis, Johnston, & Connell, 2001).

An additional limitation concerns the children flagged for ASD. The database included no follow-up diagnostic results for these children; hence, there is some uncertainty concerning these children's confirmed diagnosis.

Finally, the database available for our analysis did not include complete data on family income, parents' educational level, or other sociodemographic variables that have been associated with disparities in diagnoses and outcomes (Burkett, Pickler, Bowers, & Folker, 2020; Zablotsky et al., 2019). This information may have nuanced our conclusions at the subgroup level. More broadly, we acknowledge that environmental factors not measured in this study can interact in complex ways with child-level factors to affect the course of development (Baylor & Darling-White, 2020; Zachor & Ben-Itzchak, 2017).


Our results document meaningful variability in the developmental trajectories and outcomes of infants and toddlers who received services through a public Part C EI program. The differential progress of children defined in terms of their primary diagnosis or area of delay is obscured when results are reported only in the aggregate. Although a minority of children served demonstrated age-appropriate skills across all domains of development by the time they exited the program, a large percentage of children demonstrated significant improvement, particularly in their area or areas of specific challenge. These results can provide policy makers, professionals, EI providers, and families with actionable information for program evaluation and improvement planning.


Bagnato S. J., Suen H. K., Fevola A. V. (2011). “Dosage” effects on developmental progress during early childhood intervention: Accessible metrics for real-life research and advocacy. Infants & Young Children, 24(2), 117–132. doi:10.1097/IYC.0b013e3182104896
Baylor C., Darling-White M. (2020). Achieving participation-focused intervention through shared decision making: Proposal of an age- and disorder-generic framework. American Journal of Speech–Language Pathology, 29, 1335–1360. doi:10.1044/2020_AJSLP-19-00043
Burkett K., Pickler R., Bowers K., Folker A. (2020). Disparities affect developmental risk for Head Start preschoolers. Journal of Pediatric Nursing, 54, 86–92. doi:10.1016/j.pedn.2020.06.010
Calder S., Ward R., Jones M., Johnston J., Claessen M. (2018). The uses of outcome measures within multidisciplinary early childhood intervention services: A systematic review. Disability and Rehabilitation, 40(22), 2599–2622. doi:10.1080/09638288.2017.1353144
Chen L., Abrams D. A., Rosenberg-Lee M., Iuculano T., Wakeman H. N., Prathap S., Menon V. (2019). Quantitative analysis of heterogeneity in academic achievement of children with autism. Clinical Psychological Science, 7(2), 362–380. doi:10.1177/2167702618809353
Children's Medical Services. (n.d.). Infant toddler developmental specialist training. Retrieved March 11, 2021, from
Christakis D. A., Johnston B. D., Connell F. A. (2001). Methodologic issues in pediatric outcomes research. Ambulatory Pediatrics, 1(1), 59–62. doi:10.1367/1539-4409(2001)001<0059:MIIPOR>2.0.CO;2
Christensen L., Mendoza J. L. (1986). A method of assessing change in a single subject: An alteration of the RC index. Behavior Therapy, 17(3), 305–308. doi:10.1016/S0005-7894(86)80060-0
Clark M. L. E., Barbaro J., Dissanayake C. (2017). Continuity and change in cognition and autism severity from toddlerhood to school age. Journal of Autism and Developmental Disorders, 47(2), 328–339. doi:10.1007/s10803-016-2954-7
Cohen J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates.
Cunningham B. J., Washington K. N., Binns A., Rofle K., Robertson B., Rosenbaum P. (2017). Current methods of evaluating speech–language outcomes for preschoolers with communication disorders: A scoping review using the ICF-CY. Journal of Speech, Language, and Hearing Research, 60, 447–464. doi:10.1044/2016_JSLHR-L-15-0329
Data USA. (n.d.). Florida: Civics—diversity. Retrieved December 9, 2020, from
Dawson G., Rogers S., Munson J., Smith M., Winter J., Greenson J., Varley J. (2010). Randomized, controlled trial of an intervention for toddlers with autism: The Early Start Denver Model. Pediatrics, 125(1), e17–e23. doi: 10.1542/peds.2009-0958
de Graaf G., Levine S. P., Goldstein R., Skotko B. G. (2018). Parents' perceptions of functional abilities in people with Down syndrome. American Journal of Medical Genetics Part A, 179(2), 161–176. doi:10.1002/ajmg.a.61004
Dunst C. J., Trivette C. M. (2009). Using research evidence to inform and evaluate early childhood intervention practices. Topics in Early Childhood Special Education, 29(1), 40–52. doi:10.1177/0271121408329227
Early Childhood Technical Assistance Center. (2019). Early childhood technical assistance center. IDEA Child Outcomes Highlights. Retrieved October 25, 2021, from
Eigsti H. J., Chandler L., Robinson C., Bodkin A. W. (2010). A longitudinal study of outcome measures for children receiving early intervention services. Pediatric Physical Therapy, 22(3), 304–313. doi:10.1097/PEP.0b013e3181e94464
Elbaum B., Celimli-Aksoy S. (2017). Empirically identified subgroups of children served in Part C early intervention programs. Journal of Developmental and Behavioral Pediatrics, 38(7), 510–520. doi:10.1097/DBP.0000000000000475
Estrada E., Ferrer E., Pardo A. (2019). Statistics for evaluating pre-post change: Relation between change in the distribution center and change in the individual scores. Frontiers in Psychology, 9, 2696. doi:10.3389/fpsyg.2018.02696
Fuller E. A., Kaiser A. P. (2020). The effects of early intervention on social communication outcomes for children with autism spectrum disorder: A meta-analysis. Journal of Autism and Developmental Disorders, 50(5), 1683–1700. doi:10.1007/s10803-019-03927-z
Ghandour R. M., Moore K. A., Murphy K., Bethell C., Jones J. R., Harwood R., Lu M. (2019). School readiness among U.S. children: Development of a pilot measure. Child Indicators Research, 12(4), 1389–1411. doi:10.1007/s12187-018-9586-8
Guralnick M. J. (2011). Why early intervention works. Infants & Young Children, 24(1), 6–28. doi:10.1097/iyc.0b013e3182002cfe
IDEA Infant & Toddler Coordinators Association. (2020). 2020 tipping points survey. Demographics and challenges.
Individuals with Disabilities Education Act, Pub. L. No. 108-446, 118 Stat. 2647 (2004).
Jacobson N. S., Follette W. C., Revenstorf D. (1984). Psychotherapy outcome research: Methods for reporting variability and evaluating clinical significance. Behavior Therapy, 15(4), 336–352. doi:10.1016/S0005-7894(84)80002-7
Javalkar K., Litt J. S. (2017). Reason for referral predicts utilization and perceived impact of early intervention services. Journal of Developmental and Behavioral Pediatrics, 38(9), 706–713. doi:10.1097/DBP.0000000000000494
Jimenez-Gomez A., Standridge S. M. (2014). A refined approach to evaluating global developmental delay for the international medical community. Pediatric Neurology, 51(2), 198–206. doi:10.1016/j.pediatrneurol.2013.12.018
Khetani M. A., Richardson Z., McManus B. M. (2017). Social disparities in early intervention service use and provider-reported outcomes. Journal of Developmental and Behavioral Pediatrics, 38(7), 501–509. doi:10.1097/DBP.0000000000000474
Kruythoff-Broekman A., Wiefferink C., Rieffe C., Uilenburg N. (2019). Parent-implemented early language intervention programme for late talkers: Parental communicative behaviour change and child language outcomes at 3 and 4 years of age. International Journal of Language and Communication Disorders, 54(3), 451–464. doi:10.1111/1460-6984.12451
Kuhlthau K., Ferris T. G. G., Iezzoni L. I. (2004). Risk adjustment for pediatric quality indicators. Pediatrics, 113, 210–216.
Kuhn M., Marvin C. A. (2016). “Dosage” decisions for early intervention services. Young Exceptional Children, 19(4), 20–34. doi:10.1177/1096250615576807
Litt J. S., Glymour M. M., Hauser-Cram P., Hehir T., McCormick M. C. (2018). Early intervention services improve school-age functional outcome among neonatal intensive care unit graduates. Academic Pediatrics, 18(4), 468–474. doi:10.1016/j.acap.2017.07.011
Mesterton J., Lindgren P., Ekenberg Abreu A., Ladfors L., Lilja M., Saltvedt S., Amer-Wåhlin I. (2016). Case mix adjustment of health outcomes, resource use and process indicators in childbirth care: A register-based study. BMC Pregnancy and Childbirth, 16(1), 125. doi:10.1186/s12884-016-0921-0
Newborg J. (2005). Battelle Developmental Inventory (2nd ed.). Itasca, IL: Riverside. doi:10.1177/0734282907300382
O'Toole C., Lee A., Gibbon F. E., van Bysterveldt A. K., Hart N. J. (2018). Parent-mediated interventions for promoting communication and language development in young children with Down syndrome. Cochrane Database of Systematic Reviews, 10(10), CD012089. doi:10.1002/14651858.CD012089.pub2
Richardson Z. S., Khetani M. A., Scully E., Dooling-Litfin J., Murphy N. J., McManus B. M. (2019). Social and functional characteristics of receipt and service use intensity of core early intervention services. Academic Pediatrics, 19(7), 722–732. doi:10.1016/j.acap.2019.02.004
Robins D. L., Fein D., Barton M. L., Green J. A. (2001). The Modified Checklist for Autism in Toddlers: An initial study investigating the early detection of autism and pervasive developmental disorders. Journal of Autism and Developmental Disorders, 31(2), 131–144. doi:10.1023/A:1010738829569
Romano M., Schnurr M. (2020). Mind the gap: Strategies to bridge the research-to-practice divide in early intervention caregiver coaching practices. Topics in Early Childhood Special Education. Advance online publication. doi:10.1177/0271121419899163
Sacrey L. A. R., Bennett J. A., Zwaigenbaum L. (2015). Early infant development and intervention for autism spectrum disorder. Journal of Child Neurology, 30(14), 1921–1929. doi:10.1177/0883073815601500
Sandler A. D. (2019). Value-driven care in developmental-behavioral pediatrics, Part 1: The value proposition of developmental-behavioral pediatrics. Journal of Developmental and Behavioral Pediatrics, 40(6), 472–478. doi:10.1097/DBP.0000000000000685
Scarborough A. A., Hebbeler K. M., Spiker D., Simeonsson R. J. (2011). Using survival analysis to describe developmental achievements of early intervention recipients at kindergarten. Infants & Young Children, 24(2), 133–152. doi:10.1097/IYC.0b013e3182104a7e
Shah A. (2019). Using data for improvement. BMJ, 364, l189. doi:10.1136/bmj.l189
Shelden M. L., Rush D. D. (2014). IFSP outcome statements made simple. Young Exceptional Children, 17(4), 15–27. doi:10.1177/1096250613499246
Shonkoff J. P., Hauser-Cram P., Krauss M. W., Upshur C. C. (1992). Development of infants with disabilities and their families: Implications for theory and service delivery. Monographs of the Society for Research in Child Development, 57(6), v–vi, 1–153. doi:10.2307/1166087
Sparrow S. S., Balla D. A., Cicchetti D. V. (1984). Vineland adaptive behavior scales. Circle Pines, MN: American Guidance Service. doi:10.1007/978-0-387-79948-3_1602
Spittle A., Orton J., Anderson P., Boyd R., Doyle L. (2015). Early developmental intervention programs post hospital discharge to prevent motor and cognitive impairments in preterm infants. Cochrane Database of Systematic Reviews, 24(11), CD005495. doi:10.1002/14651858.CD005495.pub4
Spittle A. J., Barton S., Treyvaud K., Molloy C. S., Doyle L. W., Anderson P. J. (2016). School-age outcomes of early intervention for preterm infants and their parents: A randomized trial. Pediatrics, 138(6), e20161363. doi:10.1542/peds.2016-1363
Tobia V., Brigstocke S., Hulme C., Snowling M. J. (2018). Developmental changes in the cognitive and educational profiles of children and adolescents with 22q11.2 deletion syndrome. Journal of Applied Research in Intellectual Disabilities, 31(1), e177–e181. doi:10.1111/jar.12344
Tsao R., Kindelberger C. (2009). Variability of cognitive development in children with Down syndrome: Relevance of good reasons for using the cluster procedure. Research in Developmental Disabilities, 30(3), 426–432. doi:10.1016/j.ridd.2008.10.009
U.S. Department of Education. (n.d.). Number and percent of infants and toddlers receiving early intervention services under IDEA, Part C, by age group, gender and state. Retrieved December 2, 2020, from
    Wainer A. L., Hepburn S., Griffith E. M. (2017). Remembering parents in parent-mediated early intervention: An approach to examining impact on parents and families. Autism, 21(1), 5–17. doi:10.1177/1362361315622411
    Will E. A., Caravella K. E., Hahn L. J., Fidler D. J., Roberts J. E. (2018). Adaptive behavior in infants and toddlers with Down syndrome and fragile X syndrome. American Journal of Medical Genetics, 177(3), 358–368. doi:10.1002/ajmg.b.32619
    Wood S., Christian M. P., Sampson A. (2018). Audit of outcomes following a community-based early intensive behaviour intervention program for children with autism in Australia. Australian Journal of Psychology, 70(3), 217–224. doi:10.1111/ajpy.12193
    Woodman A. C., Demers L., Crossman M. K., Warfield M. E., Hauser-Cram P. (2018). Part C early intervention dosage and growth in adaptive skills from early childhood through adolescence. Early Childhood Research Quarterly, 43, 73–82. doi:10.1016/j.ecresq.2018.01.007
    World Health Organization. (1978). International classification of diseases: Basic tabulation list with alphabetic index (9th ed.). Geneva, Switzerland: World Health Organization.
    World Health Organization. (2007). International classification of functioning, disability, and health: Children & youth version: ICF-CY. Geneva, Switzerland: World Health Organization.
    Zablotsky B., Black L. I., Maenner M. J., Schieve L. A., Danielson M. L., Bitsko R. H., Boyle C. A. (2019). Prevalence and trends of developmental disabilities among children in the United States: 2009–2017. Pediatrics, 144(4), 2009–2017. doi:10.1542/peds.2019-0811
    Zachor D. A., Ben-Itzchak E. (2017). Variables affecting outcome of early intervention in autism spectrum disorder. Journal of Pediatric Neurology, 15(3), 129–133. doi:10.1055/s-0037-1601444
    Zwaigenbaum L., Bauman M. L., Choueiri R., Kasari C., Carter A., Granpeesheh D., Natowicz M. R. (2015). Early Intervention for children with autism spectrum disorder under 3 years of age: Recommendations for practice and research. Pediatrics, 136(Suppl. 1), S60–S81. doi:10.1542/peds.2014-3667E

    developmental delay; developmental progress; early intervention; outcomes; Part C; Reliable Change Index

    © 2022 Wolters Kluwer Health, Inc. All rights reserved.