Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Autism Res. 2021 Nov 20;15(2):366–378. doi: 10.1002/aur.2640

Caregiver Responsiveness as a Mechanism to Improve Social Communication in Toddlers: Secondary Analysis of a Randomized Controlled Trial

Paige Huguely Davis 1, Heba Elsayed 1,4, Elizabeth R Crais 1,2, Linda R Watson 1,2, Rebecca Grzadzinski 1,3
PMCID: PMC8821206  NIHMSID: NIHMS1754809  PMID: 34799999

Abstract

Early intensive behavioral interventions (EIBI) for children at elevated likelihood for a later diagnosis of autism spectrum disorder (EL-ASD), are often delivered through parent-mediated models. An area of current exploration is whether changes in caregiver behaviors are a mechanism through which to improve and track child behaviors in these interventions. Toddlers and their caregivers participated in an intervention trial (RCT; Watson et al., 2017) and were randomized to either a parent-mediated intervention (Adapted Responsive Teaching; ART) or a control condition (Referral to Early Intervention and Monitoring; REIM). Changes in toddler social communication (SC) behaviors and characteristics of caregiver responsiveness (CR) were quantified over 8 months. Analyses were conducted to assess whether changes in CR mediated the relation between group (ART vs REIM) and changes in child SC. Results of the current study indicated that caregivers who participated in a parent-mediated intervention improved in three domains of CR (Contingent Verbal Sensitivity, Responsivity, Affect). CR was also found to be a mechanism through which children’s SC skills improved. This work provides evidence that qualities of CR serve as mechanisms through which to improve and monitor child behaviors over the course of EIBIs. These results may lead to novel intervention targets, methods for tracking change, and tailored treatment planning for toddlers with EL-ASD. The data used in this study comes from a clinical trial that was prospectively registered with the Registry of Efficacy and Effectiveness Studies (REES; Registry ID: 316.1v1).

Keywords: Early Intervention, Social Communication, Treatment Response Measure, Caregiver Responsiveness, BOSCC, Autism Spectrum Disorder

Lay Summary:

Interventions for toddlers with high likelihood for a later diagnosis of autism often include the caregiver as an active participant in the intervention. In this study, we aimed to understand qualities of caregiver responsiveness (CR) that facilitate improvements in child behaviors during intervention. Results show that increasing verbal CR and affect are ways to improve child social skills over the course of intervention.

Introduction

Research on interventions for toddlers at elevated likelihood of receiving a diagnosis of autism spectrum disorder (EL-ASD) is limited. Interventions that have been proposed are typically implemented through a parent-mediated model (Baranek et al., 2015; Green et al., 2015, 2017; Jones et al., 2017; Koegel et al., 2014; Rogers et al., 2014; Watson et al., 2017; Whitehouse et al., 2019; Yoder et al., 2021a, 2021b). These studies have shown some promise for reducing emerging ASD characteristics (Baranek et al., 2015; Green et al., 2015, 2017; Jones et al., 2017; Koegel et al., 2014; Rogers et al., 2014; Trembath et al., 2019; Watson et al., 2017; Whitehouse et al., 2019; Yoder et al., 2021b, 2021a); however, the specific mechanisms that contribute to improvements is still unknown. Since many of these interventions work to incorporate techniques into a caregiver and toddler’s daily routines (Baranek et al., 2015; Green et al., 2015, 2017; Jones et al., 2017; Koegel et al., 2014; Rogers et al., 2014; Watson et al., 2017; Whitehouse et al., 2019; Yoder et al., 2021a, 2021b), they can become part of ongoing family life rather than an additional activity. If such interventions are effective, they may also reduce the need for costly and time-intensive post-diagnosis early intensive behavioral interventions (EIBI). Identifying the specific, active ingredients of positive change through intervention will allow researchers and clinicians to tailor treatment and future intervention programs to maximize outcomes.

Research has found that increases in caregiver responsiveness (CR), which is malleable over the course of intervention (Hampton & Rodriguez, 2021; Kasari et al., 2014; Siller et al., 2013; Venker et al., 2011; Watson et al., 2017), is linked to child improvements, such as expressive and receptive language growth (Watson et al., 2017; Yoder et al., 2015). This suggests that CR may support formative skill growth, such as joint attention, by engaging the child’s intrinsic interests, following the child’s lead, imitating, expanding on the child’s interest, commenting, and recognizing non-verbal communicative efforts such as eye gaze and facial expressions (Kasari et al., 2010; Shire et al., 2016). Some researchers have found that predictors of treatment outcome include caregiver involvement as well as treatment intensity and duration (Klinger et al., 2020), though the influence of mechanisms remains unclear (Crank et al., 2021). Caregiver behavior, such as responsiveness, is a key part of the child’s environment and may be a critical mechanism to evaluate changes that occur in the child though additional research is needed to clarify specific mechanisms of treatment outcomes.

In a recent systematic review, Hampton and Rodriguez (2021) identified 13 studies that found evidence in support of the malleability of caregiver behavior and the influence caregiver behavior has on the long-term outcomes of children at EL-ASD. Watson et al. (2017), one study included in the systematic review, examined the effectiveness of CR within a randomized controlled trial (RCT). The study tested the main effects of the caregiver-mediated intervention known as Adaptive Responsive Teaching (ART; Baranek et al., 2015; Watson et al., 2017), as well as the mediating effects of global CR on changes in a variety of toddler skills, as prespecified outcomes for the trial. To measure ASD symptom severity outcomes, Watson et al. (2017) used the Autism Diagnostic Observation Schedule, second edition (ADOS-2; Lord et al., 2012) Module 1 total algorithm score, a measure with good diagnostic validity but inconsistent sensitivity to changes in child behaviors over short periods of time (Green et al., 2010; Owley et al., 2001). Mediating effects of CR on the association between group assignment and child outcomes were found for more than half of the child outcomes used in this study, including expressive language, receptive language, fine motor skills, adaptive communication skills, adaptive social skills, and autism characteristics, with indirect effects that favored the treatment group in each case (Watson et al., 2017). However, CR did not mediate child outcomes on the Communication and Symbolic Behavior Scales Behavior Sample (Wetherby & Prizant, 2002), an assessment that includes many social-communication items. Although these results provide insight into the examination of CR as a mechanism of treatment change, further exploration is needed into the specific characteristics of CR, which may be enhanced by a using a measure developed to be more sensitive to changes in social communication (SC; Vivanti et al., 2014).

Studies have indicated that caregiver verbal responsiveness, broadly, influences their child’s language development, even up to three years later (Haebig et al., 2013; McDuffie & Yoder, 2010). A recent systematic review of the association between caregiver verbal responsiveness and the communication skills of children with ASD highlighted that studies using global ratings of parent verbal responsiveness were less likely to find significant associations with child communication skills than studies that coded specific aspects of parent verbal responsiveness (Edmunds et al., 2019). Additionally, in some studies, including Watson et al. (2017), CR measures have combined both verbal and nonverbal aspects of CR rather than examining whether there are separate or additive effects of these types of responsiveness. Researchers have also looked at the various nuances of CR such as affect (Brian et al., 2017; Yoder et al., 2021b) and praise (Crowell et al., 2019). Thus, by assessing changes in specific coded aspects of verbal and nonverbal CR in a mediation model, in addition to changes in more qualitative aspects of CR captured via ratings (e.g., affect, praise, sensitivity and gestures) we may be able to elucidate the specific mechanisms of an efficacious intervention and subsequently provide new treatment targets for toddlers with EL-ASD. Evaluating a broad and narrow scope of CR characteristics can provide interventionists and researchers a more in-depth view of effective caregiver strategies.

Beyond the world of ASD parent-mediated intervention research, scholars have worked to understand the way specific parent behaviors affect a variety of child and adolescent outcomes (Baumrind, 1991; Cimpian et al., 2007; Eckshtain et al., 2018; Gunderson et al., 2018). Some of these behaviors include caregivers’ use of praise (Gunderson et al., 2018), “subtle linguistic cues” (Cimpian et al., 2007), parenting style (Baumrind, 1991), and parental depressive symptoms (Eckshtain et al., 2018). There is a need for research regarding the connection between such caregiver behaviors and child outcomes in children with ASD. The current study takes steps to address this need.

Mediation analyses are capable of answering questions about why or how an intervention works, but are underutilized (Fairchild & Mcdaniel, 2017), including in studies of parent-mediated interventions for young children with ASD (Trembath et al., 2019). Conducting mediation analyses help researchers pinpoint mechanisms that can be acted upon more directly in prevention and intervention (Fairchild & Mcdaniel, 2017). Evaluating the mediating effect of different aspects of CR provides an opportunity to understand specific mechanisms of change in treatment (Lerner et al., 2012; Vivanti et al., 2014), which will eventually increase our understanding of the complex, dynamic interplay between infants and their caregivers, and assist in the development of tailored interventions for children with EL-ASD.

Furthering the work of Watson et al. (2017), the aims of this study were to 1) measure subtle changes in SC behaviors using the Brief Observation of Social Communication Change (BOSCC), a new treatment response measure (Grzadzinski et al., 2016) and 2) evaluate both global CR as well as specific components of CR as mechanisms of change in BOSCC scores.

Methods

Participants.

The current study used data gathered as part of a completed RCT (see Watson et al., 2017). Families were recruited from six counties in central North Carolina and were seen at two time points for this pretest-posttest design. Caregivers completed the First Year Inventory, Version 2.0 (FYI; Baranek et al., 2003), a screening tool used to identify infants at risk for ASD; families were asked to participate if their toddler scored in the “at-risk” range (pretest age of around 14 months). At posttest (around 23 months of age), 71% of the sample scored at or above the algorithm cut-off for “Autism Spectrum” on the ADOS-2 (Lord et al., 2012; Watson et al., 2017). Other eligibility criteria included birth weight >2500g, English as the primary language in the home, and at least one caregiver able to participate in the home-based intervention. All caregivers completed written informed consent as approved by the Institutional Review Board. This clinical trial project was prospectively registered with the Registry of Efficacy and Effectiveness Studies (REES; Registry ID: 316.1v1; see Consort Checklist in Supplemental Materials).

A sample of 87 toddlers with EL-ASD was enrolled, though four families did not complete Time 2 assessments. Thus, the final study sample consisted of the 83 caregiver-child dyads (56 male infants; 68%) that completed assessments at 14 months of age (Baseline; Time 1; ±0.73) and at 23 months of age (Post-test; Time 2; ±0.86). Forty-four (53%) of these dyads were randomly assigned to the treatment group receiving a caregiver-mediated intervention known as Adaptive Responsive Teaching (ART; Baranek et al., 2015; Watson et al., 2017), and the other 39 were randomly assigned to a Referral to EI and Monitoring (REIM) control group (see Figure 1 and Watson et al., 2017 for additional randomization, assignment, and statistical methods used). This variable is referred to as Group. The ART intervention, an adapted version of Responsive Teaching Curriculum (Mahoney & MacDonald, 2007), focuses on the improvement of pivotal behaviors in toddlers with EL-ASD through caregiver coaching (Baranek et al., 2015; Watson et al., 2017). During in-home visits, caregivers of infants were taught strategies for responding to their toddler’s (often subtle) social and sensory regulation cues and methods for maximizing toddler engagement, interest, and social reciprocity (see Watson et al., 2017).

Figure 1. Consort Diagram.

Figure 1.

Open in a new tab

Note. Diagram retrieved from original RCT:

Watson, L. R., Crais, E. R., Baranek, G. T., Turner-Brown, L., Sideris, J., Wakeford, L., Kinard, J., Reznick, J. S., Martin, K. L., & Nowell, S. W. (2017). Parent-mediated intervention for one-year-olds screened as at-risk for autism spectrum disorder: A randomized controlled trial. Journal of Autism and Developmental Disorders, 47, 3520–3540. DOI: 10.1007/s10803-017-3268-0

The sample, mirroring the catchment area in racial composition, was predominantly White (71%; n=59). Although the sample included a range of educational levels, caregivers overall were better educated than the population in the catchment area (70% with college degrees; n=58). The majority (93%) of participating primary caregivers in the sample were females. Results from the original RCT showed only one significant main effect on a secondary outcome measure of motor skills (caregiver report) in the ART group. See Watson et al. (2017) for additional sample and recruitment details, as well as detailed results. See Table 1 for a summary of background and demographic information.

Table 1.

Background Information (n=83)

REIM ART
n=39 n=44
Time 1 Time 2 Time 1 Time 2
Mean (SD) Mean (SD) Mean (SD) Mean (SD)
Ages in months 13.7 (0.8) 22.4 (0.8) 13.8 (0.7) 22.7 (1.0)
MSEL (T-Scores)
 Fine Motor 49.1 (8.2) 42.9 (11.1) 47.4 (10.6) 39.7 (13.7)
 Visual Reception 47.8 (11.3) 47.4 (12.3) 42.6 (10.2) 45.2 (13.8)
 Receptive Language 33.4 (12.5) 46.6 (15.8) 32.8 (9.6) 42.5 (17.0)
 Expressive Language 35.3 (12.3) 41.3 (13.5) 34.1 (10.8) 41.3 (11.9)
n (%) n (%)
Sex (males) 29 (74) 27 (61)
Child Race
 White 30 (77) 29 (66)
 African-American 5 (13) 11 (25)
 Mixed race/Other 4 (10) 4 (9)
Open in a new tab

ART = Adapted Responsive Teaching; MSEL = Mullen Scales of Early Learning; REIM = Referral to EI and Monitoring

Measures.

Cognitive Abilities.

Children completed the Mullen Scales of Early Learning (MSEL; Mullen, 1995) at Time 1 and Time 2. The MSEL yields T-scores in the domains of fine motor, visual reception, receptive language, and expressive language. Visual reception scores at Time 1 (T1VR) were used as a covariate in the analyses to control for baseline non-verbal cognitive abilities, as the randomly assigned groups differed significantly on T1VR (Watson et al, 2017). See Table 1 for MSEL scores at both time points.

Child Social Communication (SC).

For the purpose of this secondary data analysis, the BOSCC (Grzadzinski et al., 2016) SC domain scores (BOSCC_SC) were used to measure changes in SC characteristics from Time 1 to Time 2. At each time point, caregiver-toddler dyads participated in video recorded free-play interactions for 10 minutes. A standard set of interactive toys such as xylophones, toy phones, planes, shape sorters, and pop-up toys were provided. All videos were coded using standard coding procedures by trained undergraduate research assistants who attained and maintained reliability on the BOSCC. In accordance with standard coding procedures, each video was coded in two, five-minute segments (see Grzadzinski et al., 2016); videos less than 10 minutes long (2% of all videos) were segmented in half. No video less than 9 minutes long was coded. Twenty-seven videos (16%) were randomly chosen for independent coding by two coders to assess Inter-Rater Reliability (IRR). The coders were unaware of which videos were being double-coded. The inter-rater reliability for the SC domain was high [Intraclass Correlation Coefficient (ICC)= 0.85, 95% CI (0.69–0.93)]. On the BOSCC, higher scores indicate greater impairments in SC behaviors. A BOSCC-SC Change (ΔBOSCC-SC) score was calculated by subtracting each child’s BOSCC-SC score at Time 1 (T1BOSCC-SC) from their BOSCC-SC score at Time 2. Thus, decreases in a child’s BOSCC score over time correspond to an improvement in SC skills (Grzadzinski et al., 2016). For ease of interpretation, BOSCC-SC scores have been inverted, such that increases in a child’s BOSCC-SC score corresponds to improvements in SC skills. ΔBOSCC-SC scores were used in the current study as the outcome measure rather than using T2BOSCC-SC scores to remain in line with the BOSCC’s intended use: scores at a single time point must be interpreted relative to another time point (Grzadzinski et al., 2016).

Caregiver Responsiveness (CR).

Two different coding schemes were used to quantify global and specific CR, respectively: the Parent Responsiveness Coding System (PRCS; based on Yoder et al., 2015) and the Maternal Behavior Rating Scale (MBRS; Mahoney et al., 1986). The coding scheme for the PRCS (Yoder et al., 2015) calculates a percentage of intervals during which a parent responds contingently to the child’s focus of attention. In contrast, the MBRS (Mahoney et al., 1986) coding scheme uses Likert 1–5 scales with qualitative descriptions assigned to each value, with higher values representing more positive ratings of caregiver behaviors. The MBRS more specifically judges the quality of an interaction between the caregiver-child dyad. Including both quantitative and qualitative-focused measures contributes to our understanding of the nuances of CR that may impact outcome. The PRCS and MBRS were coded by research assistants (different from the coders of the BOSCC) blind to treatment group and time point in accordance with standard procedures for each measure, based on the same caregiver-toddler free-play videos coded for the BOSCC (described above). For the PRCS, 21 pretest and 12 posttest videos (18%) were used to assess inter-rater reliability, which was very high for both pretest and posttest videos [Pretest ICC = 0.93; Posttest ICC = 0.95]. For the MBRS, 13.5% of pretest and 23% of posttest (18% of total videos) were independently coded to assess inter-rater reliability. For pretest, ICCs for the MBRS scales were Responsiveness, 0.63; Affect, 0.80; Achievement Orientation, 0.83; and Directiveness, 0.79. ICCs for posttest were Responsiveness, 0.93; Affect, 0.72; Achievement Orientation, 0.84; and Directiveness, 0.87.

The PRCS uses partial interval coding (5-s intervals). For all intervals with a child lead (object or activity on which the child’s attention was focused), coders indicated whether the caregiver provided a verbal follow-in response to the child’s lead, a nonverbal follow-in response, both a verbal and a nonverbal response, or neither. The PRCS yields metrics of both verbal and nonverbal responsiveness. Caregiver Contingent Verbal Sensitivity (will be referred to as Verbal Sensitivity) was defined as the percent of intervals in which the caregiver gave a follow-in verbal response to the child’s lead (Yoder et al., 2015). Contingent Nonverbal Sensitivity (will be referred to as Nonverbal Sensitivity) was calculated by subtracting the verbal responses from the total CR variable within the dataset. Nonverbal Sensitivity thus captures responses by caregivers that were only nonverbal and not combined verbal and nonverbal behaviors. Scores for Verbal Sensitivity and Nonverbal Sensitivity at T1 and T2, respectively (T1Verbal-Sensitivity, T2Verbal-Sensitivity, T1Nonverbal-Sensitivity, T2Nonverbal-Sensitivity) were calculated. Change in Verbal Sensitivity (ΔVerbal-Sensitivity) from T1 to T2 was calculated by subtracting T1Verbal-Sensitivity from T2Verbal Sensitivity. The same was done for calculating a Change in Nonverbal Sensitivity (ΔNonverbal-Sensitivity) variable. For both ΔVerbal-Sensitivity and ΔNonverbal-Sensitivity, positive values correspond with an increase in the frequency of responses over time.

To look beyond the more quantitative measures of verbal versus nonverbal CR, two CR variables were extracted from the MBRS (Mahoney et al., 1986) coding scheme: (1) Responsivity and (2) Affect. In addition to the consistency with which a caregiver responds appropriately to a child’s behaviors (facial expressions, vocalizations, body language, demands, intentions, signs of discomfort), the MBRS Responsivity code also accounts for caregiver sensitivity and effectiveness. Sensitivity includes a caregiver’s awareness of, and responsiveness to, cues regarding the child’s interest during the child-free play video and includes consideration of whether the caregiver recognizes more subtle cues, such as a changed direction of the child’s gaze. Effectiveness involves a caregiver’s ability to engage in reciprocal interactions with their child, leading to cooperative interactions in which the caregiver and child engage in more balanced turn taking in play and/or conversation. Thus, the MBRS Responsivity variable contributes to the study’s overall CR paradigm by accounting for qualitative aspects of CR that extend well beyond the caregiver’s contingent verbal and nonverbal responses to the child’s focus of attention, as measured by the PRCS. The MBRS variable of Affect captures the caregiver’s positive animation, enjoyment, expressiveness, inventiveness and warmth directed toward the toddler during their interactions. These variables were not exclusively verbal or nonverbal but included both types of communication. Variables for Responsivity and Affect at Time 1 and Time 2 were created via this coding process (T1Responsivity and T1Affect; T2Responsivity and T2Affect). Change scores were calculated for all variables via the same procedures as the PRCS variables described above (T2-T1; ΔResponsivity; ΔAffect). All video-coded variables of change were normally distributed without outliers with the exception of ΔNonverbal-Sensitivity (see Figure S2). Two outliers were removed from the sample and linear regression and correlation analyses were rerun for the ΔNonverbal-Sensitivity variable.

Data Analysis

Preliminary analyses (bivariate correlations and linear regressions) were conducted in order to identify which CR variables were associated with changes in child SC behaviors (as measured by the BOSCC). In the linear regression model, the different CR variables served as the independent variables, and ΔBOSCC-SC served as the dependent variable. Separate models were run for each independent variable. Covariates for each model included the corresponding CR variable at T1, T1BOSCC-SC, and T1VR. Based on this analysis, the CR variables that correlated significantly with changes in SC behaviors were used as individual mediators in separate mediation models. This study follows a mediation for explanation model (Fairchild & MacKinnon, 2014; Hayes, 2017; MacKinnon, 2008) in that we are evaluating new data from a completed RCT. The covariates included in all mediation models were T1BOSCC-SC, T1VR, and the associated CR variable at T1 (See Figure 2). Mediation analyses were conducted using model 4 within Process v3.1 (http://www.afhayes.com), and a bootstrap estimation approach of 10,000 samples was employed.

Figure 2. The Components of Caregiver Responsiveness.

Figure 2.

Open in a new tab

Note. PRCS = Parent Responsiveness Coding System MBRS = Maternal Behavior Rating Scale

Results

PRCS.

Verbal Sensitivity.

In the linear regression models, ΔVerbal-Sensitivity significantly predicted ΔBOSCC-SC (β=0.16, t=4.004, p<0.01; [r2=0.13; F(2,80)=6.20, p<0.01]). T1 BOSCC-SC (t=5.04, p<0.01) and T1Verbal-Sensitivity (t=3.48, p<0.01) also were significant predictors of ΔBOSCC-SC in the linear regression model. In the mediation analysis, the direct effect of Group on ΔBOSCC-SC was non-significant. About one-third of the variance (R2=0.36) in ΔBOSCC-SC was accounted for by ΔVerbal-Sensitivity, as well as the covariates (T1BOSCC-SC, T1VR, and T1Verbal-Sensitivity). The indirect coefficient was significant, b=1.55, SE=0.66, 95 CI [0.44, 2.99], indicating that change in Verbal Sensitivity had a significant mediation effect on the relation between Group assignment and ΔBOSCC-SC.

Nonverbal Sensitivity.

ΔNonverbal-Sensitivity did not correlate with ΔBOSCC-SC (r −0.05, p=0.67), nor was it a significant predictor in linear regression models (β= −0.09, t= −0.86, p=0.39. T1BOSCC-SC (t=4.07, p<0.001) was a significant covariate in the linear regression model, highlighting the importance of baseline social skills in treatment response.

MBRS.

Responsivity.

In the linear regression models, ΔResponsivity significantly predicted ΔBOSCC-SC (β=0.96, t=2.93, p=0.004, [r2=0.30; F(4,78)=8.48, p<0.01]). T1BOSCC-SC (t=4.84, p<0.01) and T1Responsivity (t=2.91, p<0.01) also were significant predictors in the linear regression model. 31% of the variance (R2=0.31) in ΔBOSCC-SC was accounted for by ΔResponsivity, as well as the covariates included in the model (T1 BOSCC-SC, T1VR, and T1Responsivity). When examining the Responsivity variable in the mediation analysis, the direct effect of Group on ΔBOSCC-SC was non-significant. The indirect coefficient, however, was significant, b=1.45, SE=0.56, 95% CI [0.41, 2.58], indicating a significant mediation effect of ΔResponsivity on the relation between Group assignment and ΔBOSCC-SC.

Affect.

In the linear regression model, ΔAffect significantly predicted BOSCC-SC Change (β=0.54, t=2.05, p<0.05, [r2=0.25; F(4,78)=6.38, p<0.01]). T1BOSCC-SC (t=4.27, p<0.01) and T1VR (t=2.13, p<0.05) also were significant predictors in the linear regression model. A quarter of the variance (R2=0.25) in ΔBOSCC-SC was accounted for by ΔAffect, as well as the covariates included in the model (T1BOSCC-SC, T1VR, and T1Affect). In the mediation analysis using the Affect variable as an individual mediator, the direct effect of Group on ΔBOSCC-SC was non-significant. The indirect coefficient was significant, b=1.01, SE=0.56, 95% CI [0.11, 2.31], indicating that change in caregiver affect had a significant mediation effect on the relation between Group assignment and ΔBOSCC-SC. See Table 2 and Figure 3.

Table 2.

Linear Regression Statistics for CR Change Variables and ΔBOSCC-SC

β t p r2 F(df1, df2) p
ΔVerbal-Sensitivity 0.16 4.00 0.00 0.35 F(4, 78)=10.71 0.00
ΔNonverbal-Sensitivity −0.07 −0.74 0.46 0.22 F(4, 78)=5.43 0.00
ΔResponsivity 0.96 2.93 0.004 0.30 F(4, 78)=8.48 0.00
ΔAffect 0.54 2.05 0.04 0.25 F(4, 78)=6.38 0.00
Open in a new tab

Note. Each CR Change variable was used in a separate linear regression model that includes the covariates: respective T1 CR variable, T1BOSCC-SC, T1VR (not shown; see text)

Figure 3. Mediation Analysis for CR Variables.

Figure 3.

Open in a new tab

Note. Mediation analysis testing mediating effect of CR Change variable on the relationship between Group and BOSCC-SC Change. *p<0.05 **p<0.01 ***p<0.001; Covarying for BOSCC-SC Score at Time 1 and CR at Time 1; 10,000 bootstraps; Group= REIM (Referral to Early Intervention and Monitoring) vs. ART (Adaptive Responsive Teaching); BOSCC=Brief Observation of Social Communication Change; SC=Social Communication

Post hoc analyses.

To understand the relations among the three significant mediation variables, we ran posthoc Pearson correlations for ΔVerbal-Sensitivity, ΔResponsivity, and ΔAffect (see Table S1 in supplementary materials). All three CR variables correlated significantly, with p-values ≤0.01 (2-tailed). However, all of the correlation coefficients were less than 0.5, indicating that each of these CR variables captures some unique aspects of CR.

Additionally, to further understand the unique impact of the CR variables on ΔBOSCC-SC, a multiple mediation model was run (Model 6 within Process v3.1 http://www.afhayes.com; bootstrap estimation approach of 10,000 samples was employed). Using both ΔVerbal-Sensitivity and ΔResponsivity as mediators, including covariates, the model highlights that ΔVerbal-Sensitivity remains significant above and beyond ΔResponsivity (See Figure 4). Additionally, when including both ΔVerbal-Sensitivity and ΔAffect as mediators, neither change variable remained statistically significant; however, the confidence intervals indicate that trends for both mediators remain (See Figure 4). In this model, both ΔVerbal-Sensitivity and ΔAffect may be equally important contributors to SC improvements.

Figure 4. Mediation Analyses with Multiple CR Mediators.

Figure 4.

Open in a new tab

Note. Mediation analysis testing mediating effect of 2 CR Change variable on the relationship between Group and ΔBOSCC-SC. *p<0.05 **p<0.01 ***p<0.001; Covarying for BOSCC-SC Score at T1, VR at T1 and CR variables at T1; 10,000 bootstraps; BOSCC=Brief Observation of Social Communication Change; SC=Social Communication; VR = Visual Reception

Discussion

Previous studies have found that CR is malleable over the course of intervention, although changes in CR are not consistently associated with improved child outcomes in intervention trials (Edmunds et al., 2019). In this study, group assignment had a differential impact on the three types of CR (Verbal Sensitivity, Responsivity, and Affect) such that caregivers in the ART (treatment) group exhibited greater changes in the three CR variables​; in turn, these improvements in CR were significantly associated with BOSCC-SC improvements, indicating that increasing CR is a mechanism through which children in the ART group showed improvements in SC skills. Importantly, the evidence for CR as a mechanism of change for child outcomes was only apparent through the mediation analyses, as Group did not have a significant main effect on BOSCC-SC outcomes. This is an empowering message to caregivers and treatment providers that CR can be increased over the course of intervention and is a mechanism through which children respond to early intervention (Baranek et al., 2015; Green et al., 2015; Hampton & Rodriguez, 2021; Siller et al., 2013; Venker et al., 2011; Watson et al., 2017). This work highlights the significance of CR as a worthy target of intervention and extends the work of Hampton and Rodriguez (2021) by suggesting that there may be specific aspects of CR that yield improvements in child SC characteristics that are diagnostic markers for ASD. Further, our results emphasize the importance of identifying the most potent aspects of CR for child SC improvements along with caregiver-mediated intervention strategies to effectively increase these aspects of CR across diverse caregivers.

The significant mediating effects of ΔVerbal-Sensitivity (coded quantitatively based on operational definitions) and ΔResponsivity and ΔAffect (evaluated via qualitative ratings) between Group and ΔBOSCC-SC provide insight into types of CR that may be most effective in treatment. For example, helping caregivers to respond verbally to subtle child cues and increase their positive affect (e.g., facial expressions, intonation) in responding may be a particularly effective treatment mechanism. The current project found that verbal responses had a significant mediating effect between group and child SC outcomes, but that nonverbal responses alone were not significantly related to child SC outcomes. This indicates that helping caregivers to respond verbally is particularly important to promote child improvements (Brian et al., 2017; Haebig et al., 2013; McDuffie & Yoder, 2010). The caregivers assigned to the ART group made more gains in all these aspects of CR than caregivers assigned to the REIM group; high levels of Verbal Sensitivity as well as nuanced affect and responsivity may be important in mediating improvements in child SC.

The above interpretations are further supported by the post hoc analyses, which provided evidence that, when both ΔResponsivity and ΔVerbal-Sensitivity are included in a multiple mediation model, ΔVerbal-Sensitivity is the mediator of maximal impact. Additionally, when both ΔAffect and ΔVerbal-Sensitivity were included in another multiple mediation model, neither mediator remained significant, providing evidence for the important contributions of both ΔAffect and ΔVerbal-Sensitivity. These results may be helpful to interventionists and researchers striving to prioritize the effectiveness of caregiver-mediated interventions. However, additional research is needed to better understand the separate effects of these intercorrelated CR dimensions on ΔBOSCC-SC or other child outcome measures.

The significant effect of T1VR (covariate) on ΔBOSCC-SC suggests that the greater nonverbal skills of a child at Time 1 corresponds to greater gains in SC skills during intervention. The scores for T1BOSCC-SC also positively correlated with the amount of ΔBOSCC-SC a toddler experienced (see Figure S1), suggesting that child characteristics at baseline may provide information about how much improvement to expect over the course of an intervention (Ben-Itzchak et al., 2014; Contaldo et al., 2020; Zachor & Ben-Itzchak, 2017). Understanding the specific impacts of initial SC skills of toddlers and the way these baseline measures may impact the degree to which they are able to improve should be a focus of future projects. Further research identifying baseline child characteristics that inform changes or outcomes is necessary to determine which treatments are best to implement with which children.

Caregivers serve an influential role in early intervention for infants and young toddlers with EL-ASD. Our work adds to evidence of caregivers as key contributors to treatment for children with EL-ASD despite questions raised regarding the role of CR in impacting child outcomes (Kasari, 2019). Since caregivers are involved in dyadic interactions with their child more often than therapy providers, they can implement treatment strategies as part of daily routines (Schreibman et al., 2015), capitalizing on the capable and teachable “caregiver interventionist.”

CR as a treatment mechanism may be especially relevant in pre-emptive interventions for infants and young toddlers with EL-ASD. Many infants and young toddlers showing early characteristics associated with ASD will not be identified as candidates for early intervention in the absence of screening; for example, among the participants in this study, only 12 of the 87 infants randomized had received any early intervention services in the community prior to their screening at 12 months of age (Watson et al., 2017). And even when identified as with EL-ASD, these infants may not yet show sufficient developmental delays or differences to qualify them for direct intervention services, and/or caregivers may decide against pursuing direct intervention services while their child is very young. So, although options for direct intervention services may be limited for young children at EL-ASD, working with caregivers may provide a unique opportunity to influence child outcomes without the need for a child diagnosis or specific level of delay, and may be more acceptable to some families than enrolling their toddlers in direct services.

For the sample used in this study, researchers were able to re-recruit 65% of the sample at preschool age (M=54 months, SD=11 months; Time 3;Grzadzinski et al., 2020) to obtain diagnostic follow-up information, and one-third of those children were diagnosed with ASD and noting characteristics associated with autism in an additional 30% (totally >60% of follow up sample with diagnosis or symptoms of ASD). An additional 10% also displayed significant developmental delays, though not autism. Nevertheless, fewer than half of the infants had accessed community early intervention services by the end of the RCT, despite referrals and supports offered by project staff to facilitate such access. Given the high preponderance of autism symptoms and developmental delays continuing into preschool, the results of this work highlight the need to better understand the nuances of child and caregiver characteristics in early development that may predict outcomes and ultimately be targeted to promote improved outcomes.

The variables in our models accounted for up to 36% of the variance in child SC outcomes. These results in combination with improvements observed in interventions that do not use caregivers as active participants (e.g., child-therapist models of speech, occupational, or behavioral therapy; Magiati et al., 2012; Smith et al., 2000), highlight that CR is not the only possible active ingredient in interventions. Future work should continue to examine which variables are active treatment mechanisms beyond CR. In addition, it is possible that integrating caregivers into interventions that are typically clinician-implemented may bolster those treatments’ effectiveness (Hampton & Kaiser, 2016).

Future research should focus on measuring caregiver behaviors that may lead to cascading improvements in a child’s skills throughout an intervention. While pre-emptive early intervention studies such as this one may find limited direct impacts of their intervention on toddler improvements (Green et al., 2015; Watson et al., 2017; Whitehouse et al., 2019), measures of caregiver interactional behaviors, such as responsiveness, sensitivity, and affect, may aid in disentangling the heterogeneous treatment outcomes observed across children (Howlin et al., 2011; Vivanti et al., 2014). Though it is not expected that all children will show improvements over the course of an intervention, understanding for whom and why an intervention is effective is critical for individualizing treatments and optimizing impacts (Vivanti et al., 2014). Therefore, if children of caregivers whose responsiveness increases over treatment also show more gains in SC skills, as shown in this work, caregiver-mediated models may be particularly effective for at least a subgroup of children. One possible avenue is to assess CR at midpoint and implement therapeutic changes for caregivers whose CR is not changing in an effort to promote maximal impacts (e.g., through a SMART RCT model; (Almirall et al., 2014)).

Current research provides ample evidence that the way a caregiver responds, and the frequency of these responses is associated with later language outcomes of their children (Haebig et al., 2013; McDuffie & Yoder, 2010; Siller & Sigman, 2008; Warren et al., 2010; Yoder et al., 2015). This study supports evidence that CR is malleable within intervention (Kasari et al., 2014; Siller et al., 2013; Venker et al., 2011; Watson et al., 2017) and CR change influences improvements in child behaviors during intervention (Watson et al., 2017). Since SC measures on the BOSCC include verbal as well as nonverbal behaviors—such as gestures, eye contact, and facial expressions—this work extends recent studies beyond language outcomes in toddlers while also testing the changes in CR and child SC behaviors over time. Furthermore, this study adds to current research by beginning to understand the specific types of CR that mediate later child outcomes in SC.

While this work has focused on changes in CR over the course of an early intervention, it is important to clarify that CR is not impaired or different in caregivers of children with EL-ASD compared to caregivers of typically developing children (Crowell et al., 2019). Rather, it is likely that toddlers with EL-ASD need enhanced, tailored, or optimized CR in order to maximize growth in skills. This work does not ask the question of “how much” or whether there is an optimal combination of CR qualities that is necessary to promote positive SC outcomes in toddlers. Previous research has shown that specific levels of visual and auditory input promote optimal outcomes, such that too little as well as too much may not be optimally effective (Kidd et al., 2012, 2014). Similarly, there may be optimal levels of CR that lead to maximum improvements. In addition, optimal levels of CR may differ depending on different child characteristics or dyadic qualities. More nuanced examinations of these variables are necessary to truly advance these research questions.

The results of this work should be considered in light of several limitations. This sample consists of predominantly White, highly-educated families, and the results may not extend to more diverse samples. Certain models with additional variables or testing moderation effects could not be run due to the insufficient sample size of this study. Additionally, change scores were used due to only having two time points, yet the BOSCC was designed to detect change over time in ASD characteristics (Grzadzinski et al., 2016). Results indicate that the higher the T1BOSCC-SC score, the more improvements a child displayed in BOSCC-SC scores. Although this raises a concern regarding whether or not the sample is regressing to the mean, we cannot address this possibility with only 2 timepoints. This limitation should be addressed in future research projects. Because the majority of primary caregivers (93%) in this study were females, future work should look at caregiver and toddler changes with fathers as the main caregiver-participant, or in multiple family caregivers when more than one caregiver participates in coaching sessions (as was the case for more than 40% of the families in the ART group in this trial, although CR was only measured for the primary caregiver).

Further research will elucidate whether interventions should focus predominately on CR forms similar to those that were significant in the current study and dive deeper into the combined qualities of these CR variables in child outcomes. For example, are sensitive verbal responses of caregivers delivered with high positive affect more effective than sensitive nonverbal responses delivered with high positive affect? Do different aspects of CR play a mediating role in different targeted child outcomes? CR measures that include both qualitative and quantitative measures can allow for more nuanced comparisons between variables. Measures such as the Measure of NDBI Strategy Implementation-Caregiver Change (MONSI-CC; Vibert et al., 2020), a coding scheme that captures the amount and effectiveness of caregiver strategies, may be helpful in improving the way researchers capture caregiver behavior during intervention, thereby increasing researchers’ ability to make stronger claims about the association between CR and the development of child communication skills. The impact of child factors (e.g., social attention or language comprehension level) on different forms of CR also should be examined in future studies.

The results of this study are a stepping stone to future research examining child variables, caregiver variables, and dyadic qualities that relate to child outcomes in parent-mediated interventions. As a nuanced understanding of these relations emerges, researchers will be well-positioned to develop informed, tailored intervention programs for young children with or at risk for ASD.

Supplementary Material

supinfo
fS1
tS1
fS2

Acknowledgements:

This research reported here was supported by the National Institute of Child Health and Human Development (T32 HD040127) and the Institute of Education Sciences, U.S. Department of Education (R324A100305) through grants awarded to the University of North Carolina at Chapel Hill. This research was also supported by a K award from the National Institute of Child Health and Human Development awarded to Dr. Grzadzinski (KL2TR002490). The clinical trial was prospectively registered with the Registry of Efficacy and Effective Studies (REES; Registry ID: 316.1v1). The opinions expressed are those of the authors and do not represent views of the funding agencies. Contributing coders included Alapika Jatkar, Alex Aspuru, Paige Davis, Joe Fortkort and Taylor Kennedy.

Footnotes

Conflict of Interest: Authors report the following potential conflicts of interest: Dr. Grzadzinski has received compensation for trainings she has led on the ADOS and BOSCC; she did not receive compensation for trainings associated with this manuscript. Additionally, Dr. Watson and Dr. Crais have received compensation for invited talks/workshops on the ART intervention and its strategies.

Ethical Approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

References

  1. Almirall D, Nahum-Shani I, Sherwood NE, & Murphy SA (2014). Introduction to SMART designs for the development of adaptive interventions: with application to weight loss research. Translational behavioral medicine, 4, 260–274. 10.1007/s13142-014-0265-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Baranek GT, Watson LR, Crais E, & Reznick S (2003). First Year Inventory (FYI) 2.0. Chapel Hill, NC: University of North Carolina at Chapel Hill. [Google Scholar]
  3. Baranek GT, Watson LR, Turner-Brown L, Field SH, Crais ER, Wakeford L, Little LM, & Reznick JS (2015). Preliminary efficacy of Adapted Responsive Teaching for infants at-risk for autism spectrum disorder in a community sample. Autism Research and Treatment, 2015, 386951. doi: 10.1155/2015/386951. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Baumrind D (1991). The Influence of Parenting Style on Adolescent Competence and Substance Use. The Journal of Early Adolescence, 11(1), 56–95. 10.1177/0272431691111004 [DOI] [Google Scholar]
  5. Ben-Itzchak E, Watson LR, & Zachor DA (2014). Cognitive ability is associated with different outcome trajectories in autism spectrum disorders. Journal of Autism and Developmental Disorders, 44, 2221–2229. doi: 10.1007/s10803-014-2091-0. [DOI] [PubMed] [Google Scholar]
  6. Brian JA, Smith IM, Zwaigenbaum L, & Bryson SE (2017). Cross-Site Randomized Control Trial of the Social ABCs Caregiver-Mediated Intervention for Toddlers with Autism Spectrum Disorder Clinical Trial Title: Social ABCs for Toddlers with Signs of Autism Spectrum Disorder: RCT of a Parent-Mediated Inter-vention. Autism Research, 10, 1700–1711. 10.1002/aur.1818 [DOI] [PubMed] [Google Scholar]
  7. Cimpian A, Arce H-MC, Markman EM, & Dweck CS (2007). Subtle Linguistic Cues Affect Children’s Motivation. Psychological Science, 18, 314–316. 10.1111/j.1467-9280.2007.01896.x [DOI] [PubMed] [Google Scholar]
  8. Contaldo A, Colombi C, Pierotti C, Masoni P, & Muratori F (2020). Outcomes and moderators of Early Start Denver Model intervention in young children with autism spectrum disorder delivered in a mixed individual and group setting. Autism, 24, 718–729. 10.1177/1362361319888344 [DOI] [PubMed] [Google Scholar]
  9. Crank JE, Sandbank M, Dunham K, Crowley S, Bottema-Beutel K, Feldman J, & Woynaroski TG (2021). Understanding the Effects of Naturalistic Developmental Behavioral Interventions: A Project AIM Meta-analysis. Autism Research, 14, 817–834. 10.1002/aur.2471 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Crowell JA, Keluskar J, & Gorecki A (2019). Parenting behavior and the development of children with autism spectrum disorder. Comprehensive Psychiatry, 90, 21–29. doi: 10.1016/j.comppsych.2018.11.007. [DOI] [PubMed] [Google Scholar]
  11. Eckshtain D, Marcette LK, Schleider J & Weisz JR (2018). Parental Depressive Symptoms as a Predictor of Outcome in the Treatment of Child Depression. Journal of Abnormal Child Psychology, 46, 825–837. 10.1007/s10802-017-0323-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Edmunds SR, Kover ST & Stone WL (2019). The relation between parent verbal responsiveness and child communication in young children with or at risk for autism spectrum disorder: A systematic review and meta-analysis. Autism Research, 12, 715–731. 10.1002/aur.2100 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fairchild AJ, & McDaniel HL (2017). Best (but oft-forgotten) practices: Mediation analysis. The American Journal of Clinical Nutrition, 105, 1259–1271. doi: 10.3945/ajcn.117.152546 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fairchild AJ, MacKinnon DP (2014). Using mediation and moderation analyses to enhance prevention research. In: Sloboda Z, Petras H, editors. Advances in prevention science. Vol. 1. Defining prevention science. New York: Springer. p. 537–55. [Google Scholar]
  15. Green J, Charman T, Pickles A, Wan MW, Elsabbagh M, Slonims V, … & Jones EJ (2015). Parent-mediated intervention versus no intervention for infants at high risk of autism: a parallel, single-blind, randomised trial. The Lancet Psychiatry, 2, 133–140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Green J, Charman T, McConachie H, Aldred C, Slonims V, Howlin P, … & Pickles A (2010). Parent-mediated communication- focused treatment in children with autism (PACT): A randomised controlled trial. Lancet (British Edition), 375, 2152–2160. doi: 10.1016/S0140-6736(10)60587-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Green J, Pickles A, Pasco G, Bedford R, Wan MW, Elsabbagh M…The British Autism Study of Infant Siblings (BASIS) Team. (2017). Randomised trial of a parent-mediated intervention for infants at high risk for autism: longitudinal outcomes to age 3 years. Journal of Child Psychology and Psychiatry, 58, 1330–1340. 10.1111/jcpp.12728 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Grzadzinski R, Carr T, Colombi C, McGuire K, Dufek S, Pickles A, & Lord C (2016). Measuring changes in social communication behaviors: preliminary development of the brief observation of social communication change (BOSCC). Journal of Autism and Developmental Disorders, 46, 2464–2479. doi: 10.1007/s10803-016-2782-9. [DOI] [PubMed] [Google Scholar]
  19. Grzadzinski R, Donovan K, Truong K, Nowell S, Lee H, Sideris J, Turner-Brown L, Baranek GT, & Watson LR (2020). Sensory Reactivity at 1 and 2 Years Old is Associated with ASD Severity During the Preschool Years. Journal of Autism and Developmental Disorders. 10.1007/s10803-020-04432-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gunderson EA, Sorhagen NS, Gripshover SJ, Dweck CS, Goldin-Meadow S, & Levine SC (2018). Parent praise to toddlers predicts fourth grade academic achievement via children’s incremental mindsets. Developmental Psychology, 54(3), 397–409. https://doi-org.pallas2.tcl.sc.edu/10.1037/dev0000444.supp (Supplemental) [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Haebig E, McDuffie A, & Weismer SE (2013). Brief report: Parent verbal responsiveness and language development in toddlers on the autism spectrum. Journal of Autism and Developmental Disorders, 43, 2218–2227. https://doi-org.pallas2.tcl.sc.edu/10.1007/s10803-013-1763-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hampton LH, & Rodriguez EM (2021). Preemptive interventions for infants and toddlers with a high likelihood for autism: A systematic review and meta-analysis. Autism, 13623613211050432. 10.1177/13623613211050433 [DOI] [PubMed] [Google Scholar]
  23. Hampton LH & Kaiser AP (2016). Intervention effects on spoken-language outcomes for children with autism: a systematic review and meta-analysis. Journal of Intellectual Disability Research, 60, 444–463. doi: 10.1111/jir.12283 [DOI] [PubMed] [Google Scholar]
  24. Hayes AF (2017). Introduction to Mediation, Moderation, and Conditional Process Analysis: Second Edition: A Regression-Based Approach (Kenny DA & Little TD (eds.); 2nd ed.). Guilford Publications. https://www.guilford.com/books/Introduction-to-Mediation-Moderation-and-Conditional-Process-Analysis/Andrew-Hayes/9781462534654 [Google Scholar]
  25. Howlin P, Charman T, & Ghaziuddin M (Eds.). (2011). The SAGE handbook of developmental disorders. Sage. [Google Scholar]
  26. Jones JHE, Dawson G, Kelly J, Estes A, Webb SJ (2017). Parent-delivered early intervention in infants at risk for ASD: Effects on electrophysiological and habituation measures of social attention: Intervention in infants at risk for ASD. Autism Research, 10, 961–972. 10.1002/aur.1754 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kasari C (2019). Time to rethink pre-emptive interventions for infants with early signs of autism spectrum disorder. Lancet Child & Adolescent Health, 3, 586–587. doi: 10.1016/S2352-4642(19)30234-2. [DOI] [PubMed] [Google Scholar]
  28. Kasari C, Gulsrud A, Wong C, Kwon S, Locke J, (2010). Randomized Controlled Caregiver Mediated Joint Engagement Intervention for Toddlers with Autism. Journal of Autism and Developmental Disorders, 40, 1045–1056. DOI 10.1007/s10803-010-0955-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kasari C, Gulsrud A, Paparella T, Hellemann G, & Berry K (2015). Randomized comparative efficacy study of parent-mediated interventions for toddlers with autism. Journal of Consulting and Clinical Psychology, 83(3), 554–563. 10.1037/a0039080 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kasari C, Siller M, Huynh LN, Shih W, Swanson M, Hellemann GS, & Sugar CA (2014). Randomized controlled trial of parental responsiveness intervention for toddlers at high risk for autism. Infant Behavior and Development, 37, 711–721. doi: 10.1016/j.infbeh.2014.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kidd C, Piantadosi ST, & Aslin RN (2012). The goldilocks effect: Human infants allocate attention to visual sequences that are neither too simple nor too complex. PLoS ONE, 7, p.e36399–e36399. DOI: 10.1371/journal.pone.0036399 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kidd C, Piantadosi ST, & Aslin RN (2014). The Goldilocks effect in infant auditory attention. Child Development, 85, 1795–1804. https://doi-org.pallas2.tcl.sc.edu/10.1111/cdev.12263 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Klinger LG, Cook ML & Dudley KM (2020). Predictors and moderators of treatment efficacy in children and adolescents with autism spectrum disorder. Journal of Clinical Child and Adolescent Psychology, 1–8. 10.1080/15374416.2020.1833735 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Koegel LK, Singh AK, Koegel RL, Hollingsworth JR, Bradshaw J (2014). Assessing and improving early social engagement in infants. Journal of Positive Behavior Interventions, 16, 69–80. 10.1177/1098300713482977 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lerner MD, White SW, & McPartland JC (2012). Mechanisms of change in psychosocial interventions for autism spectrum disorders. Dialogues in Clinical Neuroscience, 14, 307. doi: 10.31887/DCNS.2012.14.3/mlerner [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lord C, Rutter M, DiLavore PC, Risi S, & Gotham K, & Bishop SL (2012). Autism diagnostic observation schedule, second edition (ADOS-2). Torrence: Western Psychological Services. [Google Scholar]
  37. MacKinnon DP (2008). Introduction to statistical mediation analysis. Mahwah (NJ): Erlbaum. [Google Scholar]
  38. Magiati I, Tay XW, & Howlin P (2012). Early comprehensive behaviorally based interventions for children with autism spectrum disorders: A summary of findings from recent reviews and meta-analyses. Neuropsychiatry, 2, 543–570. doi: 10.2217/npy.12.59. [DOI] [Google Scholar]
  39. Mahoney G, & Macdonald J (2007). Autism and developmental delays in young children: The responsive teaching curriculum for parents and professionals. Austin: PRO-ED. [Google Scholar]
  40. Mahoney G, Powell A, & Finger I (1986). The maternal behavior rating scale. Topics in Early Childhood Special Education, 6, 44–56. doi: 10.1177/027112148600600205 [DOI] [Google Scholar]
  41. McDuffie A, & Yoder P (2010). Types of parent verbal responsiveness that predict language in young children with autism spectrum disorder. Journal of Speech, Language, and Hearing Research, 53, 1026–1039. doi: 10.1044/1092-4388(2009/09-0023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Mullen EM (1995). Mullen Scales of Early Learning. Circle Pines, MN: AGS. [Google Scholar]
  43. Owley T, McMahon W, Cook EH, Laulhere T, South M, Mays LZ, et al. (2001). Multisite, double-blind, placebo- controlled trial of porcine secretin in autism. Journal of the American Academy of Child and Adolescent Psychiatry, 40, 1293–1299. doi: 10.1097/00004583-200111000-00009. [DOI] [PubMed] [Google Scholar]
  44. Rogers SJ, Vismara L, Wagner AL, McCormick C, Young G, Ozonoff S (2014). Autism treatment in the first year of life: A pilot study of infant start, a parent-implemented intervention for symptomatic infants. Journal of Autism and Developmental Disorders, 44, 2981–2995. 10.1007/s10803-014-2202-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Schertz HH, Odom SL, Baggett KM, & Sideris JH (2018). Mediating Parent Learning to Promote Social Communication for Toddlers with Autism: Effects from a Randomized Controlled Trial. Journal of Autism and Developmental Disorders, 48(3), 853–867. 10.1007/s10803-017-3386-8 [DOI] [PubMed] [Google Scholar]
  46. Schreibman L, Dawson G, Stahmer AC, Landa R, Rogers SJ, McGee GG, … & Halladay A (2015). Naturalistic developmental behavioral interventions: Empirically validated treatments for autism spectrum disorder. Journal of Autism and Developmental Disorders, 45, 2411–2428. https://doi-org.pallas2.tcl.sc.edu/10.1007/s10803-015-2407-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Shire S, Gulsrud A, & Kasari C, (2016). Increasing responsive parent-child interactions and joint engagement: Comparing the influence of parent-mediated intervention and parent psychoeducation. Journal of Autism and Developmental Disorders, 46, 1737–1747. doi: 10.1007/s10803-016-2702z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Siller M, Hutman T, & Sigman M (2013). A parent-mediated intervention to increase responsive parental behaviors and child communication in children with ASD: A randomized clinical trial. Journal of Autism and Developmental Disorders, 43, 540–555. https://doi-org.pallas2.tcl.sc.edu/10.1007/s10803-012-1584-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Siller M, & Sigman M (2008). Modeling longitudinal change in the language abilities of children with autism: parent behaviors and child characteristics as predictors of change. Developmental Psychology, 44, 1691–1704. [DOI] [PubMed] [Google Scholar]
  50. Smith T, Groen AD, Wynn JW, & Smith T (2000). Randomized trial of intensive early intervention for children with pervasive developmental disorder. American Journal on Mental Retardation, 105, 269–285. doi: 10.1352/0895-8017(2000)105<0269:RTOIEI>2.0.CO;2 [DOI] [PubMed] [Google Scholar]
  51. Trembath D, Gurm M, Scheerer NE, Trevisan DA, Paynter J, Bohadana G, Roberts J, Iarocci G (2019). Systematic review of factors that may influence the outcomes and generalizability of parent-mediated interventions for young children with autism spectrum disorder. Autism Research, 12, 1304–1321. 10.1002/aur.2168 [DOI] [PubMed] [Google Scholar]
  52. Venker CE, McDuffie A, Ellis Weismer S, & Abbeduto L (2011). Increasing verbal responsiveness in parents of children with autism: A pilot study. Autism, 16, 568–585. 10.1177/1362361311413396 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Vibert BA, Dufek S, Klein CB et al. (2020). Quantifying caregiver change across early Autism interventions using the Measure of NDBI Strategy Implementation: Caregiver Change (MONSI-CC). Journal of Autism and Developmental Disorders, 50, 1364–1379. https://doi-org.libproxy.lib.unc.edu/10.1007/s10803-019-04342-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Vivanti G, Prior M, Williams K, Dissanayake C (2014). Predictors of outcomes in autism early intervention: Why don’t we know more? Frontiers in Pediatrics, 2, 58. doi: 10.3389/fped.2014.00058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Warren SF, Brady N, Sterling A, Fleming K, & Marquis J (2010). Maternal responsivity predicts language development in young children with fragile X syndrome. American Journal on Intellectual and Developmental Disabilities, 115, 54–75. DOI: 10.1352/1944-7558-115.1.54 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Watson LR, Crais ER, Baranek GT, Turner-Brown L, Sideris J, Wakeford L, Kinard J, Reznick JS, Martin KL, & Nowell SW (2017). Parent-mediated intervention for one-year-olds screened as at-risk for autism spectrum disorder: A randomized controlled trial. Journal of Autism and Developmental Disorders, 47, 3520–3540. DOI: 10.1007/s10803-017-3268-0 [DOI] [PubMed] [Google Scholar]
  57. Wetherby AM, Guthrie W, Woods J, Schatschneider C, Holland RD, Morgan L, & Lord C (2014). Parent-implemented social intervention for toddlers with autism: An RCT. Pediatrics, 134, 1084–1093. doi: 10.1542/peds.2014-0757 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Wetherby AM, & Prizant BM (2002). Communication and Symbolic Behavior Scales: Developmental Profile, (1st normed ed.). Paul H Brookes Publishing Co. https://psycnet-apa-org.pallas2.tcl.sc.edu/record/2002-17060-000 [Google Scholar]
  59. Whitehouse AJO, Varcin KJ, Alvares GA, Barbaro J, Bent C, Boutrus M, … & Hudry K (2019). Pre-emptive intervention versus treatment as usual for infants showing early behavioural risk signs of autism spectrum disorder: a single-blind, randomised controlled trial. The Lancet Child and Adolescent Health, 3, 605–615. 10.1016/S2352-4642(19)30184-1 [DOI] [PubMed] [Google Scholar]
  60. Yoder PJ, Stone WL, Edmunds SR (2021a). For which younger siblings of children with autism spectrum disorder does parent-mediated intervention work? Autism: the International Journal of Research and Practice, 25, 58–69. 10.1177/1362361320943373 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Yoder PJ, Stone WL, Edmunds SR (2021b). Parent utilization of ImPACT intervention strategies is a mediator of proximal then distal social communication outcomes in younger siblings of children with ASD. Autism: the International Journal of Research and Practice, 25, 44–57. 10.1177/1362361320946883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Yoder P, Watson LR, & Lambert W (2015). Value-added predictors of expressive and receptive language growth in initially nonverbal preschoolers with autism spectrum disorders. Journal of Autism and Developmental Disorders, 45, 1254–1270. doi: 10.1007/s10803-014-2286-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Zachor DA, & Ben-Itzchak E (2017). Variables affecting outcome of early intervention in autism spectrum disorder. Journal of Pediatric Neurology, 15, 129–133. doi: 10.1055/s-0037-1601444. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supinfo
NIHMS1754809-supplement-supinfo.doc (237KB, doc)
fS1
NIHMS1754809-supplement-fS1.tiff (25.7MB, tiff)
tS1
NIHMS1754809-supplement-tS1.pdf (20.2KB, pdf)
fS2
NIHMS1754809-supplement-fS2.tiff (23.2MB, tiff)

RESOURCES