Skip to main content
NCBI home page
Sage Choice logoLink to Sage Choice
. 2026 Sep 22;30(10):2465–2480. doi: 10.1177/13623613261471808

Rapid Review of the Brief Observation of Social Communication Change (BOSCC) Shows It Can Detect Behavioural Changes in Social Communication in Autism

Claire Howlin 1,✉, Anna Kovalova 2, Artur Jaschke 2, David M Greenberg 2, Beth Mason 1, Irene Pallás-Ferrer 2, Eve Merriam 2, Catherine Brushwood 2, Jonathan Pool 3, Carrie Allison 2, Helen Odell-Miller 3, Karen Hayden 4, Carmel Simone Moore 4, Rebecca Atkinson 3, Ramarni Treyvone Wilfred 2, Simon Baron-Cohen 2
PMCID: PMC13624351  PMID: 42773560

Abstract

Reliable measurement of social communication change remains a major challenge in autism research. The Brief Observation of Social Communication Change (BOSCC) was developed to detect subtle changes in communication. This rapid review synthesised current evidence on the BOSCC’s responsiveness, validity, and suitability as a clinical trial endpoint. Systematic searches of PubMed, PsycINFO, CINAHL, and Web of Science identified 19 eligible studies (n = 789). Within-group meta-analyses estimated pre–post standardised mean change (Hedges’ g) for BOSCC and ADOS (Autism Diagnostic Observation Schedule) outcomes, using random-effect (RE) models. The BOSCC total score for intervention arms showed a moderate, significant change (g = −0.32, 95% confidence interval [CI] = [−0.42, −0.22]), with a smaller, non-significant change in control arms (g = −0.15). The Social Communication subscale yielded consistent effects on the intervention arm (g = −0.34, 95% CI = [−0.49, −0.19]), indicating sensitivity to improvements in reciprocal interaction. The restricted and repetitive behaviour (RRB) subscale also showed a moderate significant effect on the intervention arm but with wide confidence intervals indicating a large degree of heterogeneity in this result (g = −0.40, 95% CI = [−0.74, −0.06]). ADOS scores showed a similar pattern of results for the intervention arms, based on a sub-analysis of four studies that included both BOSCC and ADOS scores. Findings indicate that the BOSCC captures clinically meaningful behavioural change across diverse contexts. Variability in RRB outcomes may arise because they are not routinely targeted by supports for social communication, in line with neuro-affirmative approaches.

Lay Abstract

Autism research often aims to understand how different supports can help autistic children develop skills that are important to them and their families, such as social communication. However, changes in social communication are very difficult to measure. The Brief Observation of Social Communication Change (BOSCC) was developed to help address this. It focuses on observing natural, everyday interactions to capture subtle shifts in social communication and patterns of behaviour, and restricted and repetitive behaviours. In this review, we gathered and summarised all available studies up to March 2025 that used the BOSCC with autistic children, including both intervention studies and studies examining how well the BOSCC works. Across 19 studies including 789 children, we found that the BOSCC was consistently able to detect changes that occurred during a wide range of supports and programmes. The Social Communication subscale was particularly good at picking up changes, but the scale designed to measure repetitive behaviours was more inconsistent. The lack of changes in repetitive behaviours may reflect the diversity and important self-regulatory roles that repetitive behaviours can have, as well as the fact that many short-term interventions do not focus on changing them. The BOSCC was also found to be better at picking up changes in social communication compared to a diagnostic tool called the Autism Diagnostic Observation Schedule (ADOS). Overall, the evidence suggests that the BOSCC, particularly the Social Communication subscale, may be a useful way to detect changes in autistic communication over time. Future work should acknowledge that repetitive behaviours are not routinely targeted by supports intended to enhance communication and may serve a self-regulatory function, to manage sensory input, or to express emotions. This will help ensure that the BOSCC is used in a way that aligns with neurodiversity-affirming principles and supports ethically grounded research.

Keywords: autism, clinical, diagnoses, meta-analysis, methods, outcome measurement, school-age children, populations, interventions, supportive, social interaction, repetitive behaviours and interests

Introduction

Autism is a neurodevelopmental condition characterised by differences in social communication and interaction, as well as patterns of behaviour including restricted interests and sensory sensitivities (American Psychiatric Association, 2013). Supporting autistic children in ways that enhance social communication, participation, and autonomy are central goals in autism research and practice. However, capturing such change, particularly in naturalistic, everyday interactions, remains a methodological challenge. Many commonly used outcome measures, such as the Autism Diagnostic Observation Schedule (ADOS-2), were originally developed for diagnosis rather than for detecting short-term change. These tools often lack sensitivity to subtle behavioural shifts and may be confounded by cognitive or language ability (Grzadzinski et al., 2016; Pijl et al., 2018). Reliable and sensitive outcome measures are essential for determining the effectiveness of interventions in clinical and research settings.

Alongside these measurement challenges, autism research has increasingly shifted towards neurodiversity-affirming perspectives, which understand autism as a form of neurodevelopmental diversity rather than solely in terms of deficit or disorder (Botha et al., 2024; Doyle, 2020). From this perspective, the evaluation of outcomes requires careful consideration of what constitutes meaningful change, for whom, and under what circumstances. In particular, changes in social communication should not be interpreted as movement towards normative behaviour for its own sake but in relation to whether supports enhance autonomy, mutual understanding, functional communication, participation, and well-being (Kapp et al., 2019; Pellicano et al., 2014). Similarly, some behaviours classified as restricted or repetitive may serve important self-regulatory, sensory, or communicative functions and may not always represent appropriate targets for intervention (Charlton et al., 2021; Kapp et al., 2019; I. F. Morris et al., 2025). Accordingly, while this study examines the degree to which the Brief Observation of Social Communication Change (BOSCC) can detect change, it raises a broader question about the appropriateness of including restricted and repetitive behaviours (RRBs) in outcome measures for social communication.

The BOSCC is a structured, video-based observational coding system developed to detect subtle, intervention-related changes in social communication and RRBs among autistic children, particularly within the context of early intervention (Grzadzinski et al., 2016). The tool was developed in response to long-standing challenges in autism research, notably the lack of outcome measures that are both sensitive to behavioural change and applicable across diverse populations and settings (Fletcher-Watson et al., 2016; Pijl et al., 2018). The BOSCC has potential advantages over parent report questionnaires (e.g., the Autistic Impact Measure; Mazurek et al., 2020) used to assess autistic social communication because it is grounded in naturalistic, play-based interactions, typically involving a caregiver and child, but can also be administered by someone else who is not the caregiver. There are several different versions of the BOSCC, tailored for different levels of language development: minimally verbal, phrase speaker, Fluent Speaker 1, and Fluent Speaker 2. The age and language level of the child are used to determine the version of the BOSCC that is used with the child. The current study focuses on evaluating the minimally verbal (MV) version of the BOSCC. To administer the BOSCC, the administrator asks the child to play with them for about 15 min using a number of different toys, that have been selected because they can stimulate conversation, or elicit surprise responses. The play interactions are recorded, and coded offline by trained, blinded raters to minimise expectancy and observer bias (Reszka et al., 2024; Settanni et al., 2024).

The BOSCC assessment requires coding by individuals who have no prior knowledge of the child being evaluated, the specific time point of the assessment, or the intervention group to which the child belongs (Grzadzinski et al., 2016). This masking process is crucial, as it helps to prevent any unconscious influence on the coding that might arise from familiarity with the child’s background or knowing the type of care that the child has received (Byrne et al., 2024). The BOSCC total score is a composite measure that reflects observable behaviours associated with autism, including social communication differences and repetitive patterns of behaviour, as they present in naturalistic interactions. It does not assess level of functioning or ability but rather offers a structured way to document behavioural characteristics and potential changes over time. There are four different versions of the BOSCC, and the assessor considers both the child’s language level and age when deciding which version of the BOSCC to use (Byrne & Lord, 2023). BOSCC MV is used for people of any age who use single words or use mainly rote, or non-generative phrases, and is the most widely used and validated (Byrne & Lord, 2023). BOSCC Phrase Speech/Young Fluent (PSYF) is used for children who use flexible phrase speech and fluent speakers under 8 years of age. BOSCC Fluent 1 (F1) is used for fluent-speaking children aged 6–10 years old and includes developmentally appropriate conversational game-based tasks (Byrne & Lord, 2023). BOSCC Fluent 2 (F2) is used for people aged 11 through adulthood and includes conversational and interactive tasks (Byrne & Lord, 2023).

Since its development, the BOSCC has been increasingly employed in both high- and low-resource settings. It has been used to evaluate a wide range of interventions, including caregiver-mediated programmes such as the World Health Organization (WHO) Caregiver Skills Training (Settanni et al., 2024) and structured clinical trials of naturalistic developmental behavioural interventions (Freitag et al., 2025; Gengoux et al., 2019). Its flexibility and sensitivity could make it a valuable tool for detecting nuanced, ecologically valid changes in communication behaviour, changes that may be both statistically significant and clinically meaningful.

Nevertheless, findings have been mixed. In some studies, such as the Autism Frankfurt Early Intervention Programme (A-FFIP) trial, a large multi-centre observer-blind Randomised Controlled Trial (RCT) of a low-intensity individualised Naturalistic Developmental Behavioural Intervention (NDBI), the BOSCC did not detect significant group differences in social communication, possibly due to contextual factors such as the COVID-19 pandemic (Freitag et al., 2025). In other trials, BOSCC changes aligned with parent-reported improvements in RRBs and executive function, underscoring its value in triangulating across outcome types (Freitag et al., 2025). The BOSCC has also been used to measure social communication in other contexts, such as Fragile X syndrome (Shaffer et al., 2022). Other work has examined the BOSCC’s use across different settings (e.g., home vs. school), highlighting not only its adaptability but also variability in coding reliability across domains like RRBs (Reszka et al., 2024).

Although the BOSCC is increasingly used in autism intervention research, key uncertainties remain regarding its suitability as a primary outcome measure in clinical trials. While it was developed to address limitations in traditional diagnostic tools, particularly their insensitivity to subtle, short-term change, there is still limited consensus about when and how the BOSCC can reliably serve as an endpoint in studies evaluating intervention efficacy (Byrne et al., 2024; Freitag et al., 2025; Grzadzinski et al., 2016). As the BOSCC continues to be adopted in a growing range of intervention types, from caregiver-mediated models in community settings (Settanni et al., 2024) to structured NDBIs in RCTs (Freitag et al., 2025), there is a pressing need to evaluate whether it performs consistently across contexts. Despite promising psychometric evidence, the literature is fragmented, with studies varying in design, outcome focus, and reporting practices.

Given these gaps and the increasing reliance on BOSCC in both academic and applied research, a review and synthesis of previous results is warranted. The current study aims to synthesise and review the previous literature using a rapid review. A rapid review was used instead of a systematic review for three reasons: (a) ongoing and new RCTs need clarity on whether the BOSCC is sensitive to change; (b) a full systematic review could take years, which would potentially delay much-needed advances in autism supports; and (c) clarity on the utility of the BOSCC may encourage more researchers to use it, which will lead to a more robust evidence base, and the level of studies needed for a more robust systematic review. Accordingly, the intended users for this review are researchers or research groups who need to determine the validity of the BOSCC as a research endpoint. This approach offers a methodologically robust synthesis of current evidence, enabling researchers, clinicians, and policymakers to make informed decisions about the tool’s strengths, limitations, and appropriate contexts for use. The specific rationale for this review is to determine if the BOSCC is a suitable endpoint measure for ongoing and newly developed RCTs and also aligns with the broader need for accessible, evidence-based guidance in autism research, particularly in supporting standardised, reliable outcome measurement for clinical trials and public health interventions.

This rapid review aims to evaluate the current evidence on the efficacy of the BOSCC in capturing change in social communication among autistic children, while interpreting responsiveness and domain-level change within a neurodiversity-affirming framework. Specifically, the review sought to:

  1. Summarise the use of the BOSCC across intervention types.

  2. Assess evidence of the BOSCC’s sensitivity to intervention-related change and the suitability of the BOSCC as a clinical trial endpoint.

  3. Compare levels of change observed with the BOSCC to levels of change observed with the ADOS-2.

Methods

Design

For this rapid review, a systematic search string was used in five databases, and a pre-defined inclusion and exclusion criteria were used to determine the eligibility of articles. Two reviewers screened the titles and abstracts, and 20% of full texts were reviewed by two reviewers. A formal risk of bias assessment (which would usually be conducted in a full systematic review) was not conducted to expedite the completion time of the review. The design and reporting of this rapid review followed guidance on Rapid Review Methodology (Garritty et al., 2024; King et al., 2022; Tricco et al., 2015) and the interim guidance on the reporting of rapid reviews (Stevens et al., 2025). A rapid review was selected to efficiently synthesise emerging evidence on the use of the BOSCC as an outcome measure in autism intervention research, with a particular focus on its validity, sensitivity to change, and suitability as a clinical trial endpoint.

Eligibility Criteria

For this review, studies were included if they: (a) included autistic children aged under 12 years of age, (b) used the MV version of the BOSCC with caregiver ratings (originally the search was extended to all versions, but only one paper was identified that used the Phrase Speaker version of the BOSCC and one for fluent speaker making separate analyses infeasible), (c) were peer-reviewed studies published in English, and (d) reported pre- and post-intervention BOSCC data or BOSCC ratings in the absence of an intervention, for example, in BOSCC validation studies.

Search Strategy

A comprehensive search was conducted across PubMed, PsycINFO, CINAHL, Google Scholar, and Web of Science that covered the inception of the databases until March 26, 2025. The search strategy combined keywords and MeSH terms related to autism (“autis*” OR “ASD” OR “autistic”), social communication (“social communication” OR “social interaction”), and the BOSCC (“Brief Observation of Social Communication Change” OR “BOSCC”). Additional articles were identified by contacting the authors of the BOSCC, who were aware of several additional validation studies.

Study Selection

Retrieved citations were imported to Covidence software. Following removal of duplicates, two reviewers independently (C.H. and B.M.) screened titles and abstracts for eligibility. Full-text articles were reviewed against inclusion criteria, and discrepancies were resolved through consensus, with 20% examined by a second reviewer (B.M.). Disagreements were resolved through discussion. Finally, 19 articles were selected for inclusion. Figure 1 shows a flowchart of the study selection process.

Figure 1.

Flowchart tracking literature screening process, 120 studies narrowed down to 20 for review.

Open in a new tab

Flow chart for screening process of included articles.

Data Extraction

Data were extracted using a bespoke data extraction form developed within Covidence, designed to capture key study characteristics, BOSCC-specific implementation, and ADOS scores. All included studies had BOSCC outcomes reported. In each study, BOSCC scores were available for caregiver administration, and these were extracted; for consistency, no clinician-administered scores were included. Only six studies included information on ADOS scores (Carruthers et al., 2021; Fletcher-Watson et al., 2016; Freitag et al., 2025; Green et al., 2022; Pijl et al., 2018). In line with rapid review methodology (Garritty et al., 2024; King et al., 2022; Tricco et al., 2015), data extraction was completed by one reviewer and independently verified for a 20% random sample by a second reviewer to ensure accuracy and consistency while maintaining timeliness. Discrepancies were discussed and resolved between the two reviewers who extracted the data (C.H. and B.M.). Some studies provided details on what they considered as Care As Usual (CAU), which included co-occurring music therapy, play therapy, occupational therapy, or speech and language therapy. These details were extracted where available. Table 1 shows the characteristics of included studies.

Table 1.

Characteristics of Included Studies.

Study Country Design Intervention n Comparator n Age (months) Sex (%) BOSCC version
M SD Range Male Female
Carruthers et al. (2021) The United Kingdom RCT PACT 60 CAU 57 44 — 24–60 95 5 ADOS-BOSCC MV
Frankel Heffler et al. (2022) The United States Open-label pilot study PACT 9 None — 28.2 7.7 18–40 88.9 11.1 BOSCC-MV
Frost et al. (2019) The United States Secondary analysis of RCT Online Project ImPACT (Ingersoll et al., 2016) 50 None — 46.05 14.50 — 74 26 BOSCC-MV
Geoffray et al. (2025) France Non-randomised experimental study PACT 9 CAU 27 37.7 — 30–46 88.9 11.1 BOSCC-MV
Green et al. (2022) The United Kingdom RCT PACT-G 121 CAU 127 67.02 12.9 24–132 79 21 BOSCC-MV
Freitag et al. (2025) Germany RCT A-FFIP 68 CAU a 66 50 10 24–66 80.6 19.4 BOSCC-MV
Husu et al. (2024) Finland RCT PMI 9 CAU b 10 49.2 12 30–66 90 10 ADOS-BOSCC MV
Nordahl-Hansen et al. (2016) Norway Secondary analysis of RCT CMI (Kaale et al., 2014) 33 CAU 26 48.75 8.4 24–48 80.2 19.8 BOSCC-MV
Pijl et al. (2018) The Netherlands Secondary analysis of RCT FPT or CAU c (Oosterling et al., 2010) 44 None — 31.2 5.3 — 77.3 22.7 BOSCC-MV
Divan et al. (2018) India RCT PASS-Plus 19 CAU 21 65 21 — 90.48 9.52 BOSCC-MV
Grzadzinski (2016) The United States and the United Kingdom BOSCC-MV Validation Study Various 56 None — 28.9 10.5 — 78.57 21.42 BOSCC-MV
Klein et al. (2021) The United States RCT NDBI + VBF 8 NDBI 7 26.84 5.18 — 66.67 33.33 BOSCC-MV
Pellecchia et al. (2024) The United States RCT (a) ImPACT4,
(b) ImPACT1 (based on NDBI)		 19
23		 CAU d 18 23.2 4.7 0–30 65 35 BOSCC-MV
Settanni et al. (2024) The United States, the United Kingdom, Italy Secondary analysis of RCT CST (based on NDBI; Salomone et al., 2022) 42 None — 44.8 9.52 24–60 77.9 22.1 BOSCC-MV
Swain et al. (2024) The United States Secondary analysis of 3 RCTs (a) ESI Individual (Wetherby et al., 2014) 17 ESI Group 16 19.1 2.7 — 76.8 24.2 BOSCC-MV
(b) ESDM (Rogers et al., 2019) 42 CAU 32 24 3.61 — 72 28
(c) CMM (Kasari et al., 2014) 49 CEM 43 40.5 9.48 — 81.8 18.2
Woodard (2024) The United States Secondary analysis ESDM 12 None — — — 24–60 — — BOSCC-MV
Reszka et al. (2024) The United States BOSCC Validation Study ASAP intervention
(Boyd et al., 2018)		 13 None — 51.24 — 36–60 61.54 38.46 BOSCC-MV
Fletcher-Watson et al. (2016) The United Kingdom RCT iPad 27 CAU 27 49.5 12 0–72 80 20 BOSCC-MV
Frost et al. (2020) The United States Observational study Online Project ImPACT (Ingersoll et al., 2016) 59 None 59 46.11 15.97 17–93 76 24 BOSCC-MV
Open in a new tab

Note. n = sample size, M = mean, SD = standard deviation, BOSCC-MV Brief Observation of Social Communication Change Minimally Verbal Version (Grzadzinski et al., 2016), ADOS-BOSCC MV (Kim et al., 2019), A-FFIP = Parent-Mediated Training; Advancing Social-Communication and Play Intervention; CEM = Caregiver Education Module (active control condition); CMM = Caregiver Mediated Module (treatment condition); CMI = Caregiver-Mediated Intervention; CST = Caregiver Skills Training; ESDM = Early Start Denver Model; ESI = Early Social Interaction; FPT = Focussed Parent Training; ImPACT = Improving Parents as Communication Teachers; NDBI = Naturalistic Developmental Behavioural Intervention; Online Project ImPACT = a caregiver-mediated intervention; PACT = Paediatric Autism Communication Training; PASS-Plus = Parent-mediated intervention for Autism Spectrum Disorder Plus; PISI = Parent-Implemented Social Intervention; PACT-G = adaptation into home and education contexts of the original clinic-delivered PACT; PMI = Parent-Mediated Intervention; RCT = randomised controlled trial.

a

Individual or group-based interventions with an intensity of ⩽10 hr/week. Additional speech and language, occupational or other therapeutic interventions, individual support at kindergarten, and stable psychotropic medication were allowed and continuously documented during the study period. bSpecial early learning curriculum and speech therapy for every child, occupational therapy for 80%, and music therapy for 15% of the children. cResults of the intervention and CAU were pooled in this study. CAU involved individual speech and language therapy, motor therapy, music therapy, and play therapy. Support for parents can range from low-frequency sessions with a psychologist (e.g., 1 hr per month) to intensive practical support set up in the home environment (e.g., twice a week for 90 min). dAverage of 1 hr per week integrating developmentally focused intervention strategies, behavioural intervention strategies, or strategies focused on improving communication or play skills.

Distinguishing the Rapid Review From a Systematic Review

This review deviates in several ways from Standard Cochrane methodology. Our review was limited to articles that clearly reported data on BOSCC outcomes, so we did not follow up with authors if the data reported were incomplete. We also excluded articles in non-English languages, which may have resulted in the exclusion of potentially relevant articles. In addition, we took steps to reduce the time spent screening by only dually screening 20% of full texts. We also carried out data collection in an expedited manner by using a single review author with checks by a second review author for data extraction. A formal risk of bias assessment was not completed. This decision was made to prioritise timeliness and feasibility, consistent with rapid review methodology guidelines (King et al., 2022; Tricco et al., 2015). Given that a full systematic review could take years, which would potentially delay much-needed advances in autism supports, the research team felt it was important to interrogate the available data to gain clarity on the utility of the BOSCC, which may influence whether more researchers use it. If the number of studies using the BOSCC increases, a full systematic review with a risk of bias assessment could be warranted. In the interim, the authors acknowledge that the omission of a formal risk of bias assessment limits the robustness of the findings.

A Priori-Defined Iterative Methods

This rapid review was primarily interested in evaluating the sensitivity of the BOSCC. The decision to include data from the ADOS was dependent on the volume of relevant literature identified to make the analyses feasible. The finalisation of the list of outcomes evaluated in the meta-analyses was made by the research team.

Participatory Methods

Knowledge users were not directly involved in the development of this RR; however, the research team regularly meets with an autism community engagement group to gain greater insight into the autism community’s needs and research priorities. The community engagement group includes six people comprised of three males and three females, four White people and two Black people. Four members had autism diagnoses; two of these autistic people also had autistic children. One member was an occupational therapist who specialised in autism, and one was a teacher and foster parent of autistic children. The community engagement group supports the concept of developing additional supports for autistic children with communication difficulties, that are neuro-inclusive, and respect the uniqueness of each individual. We have implemented the suggestions of the community engagement group directly on a larger music therapy intervention study (Jaschke et al., 2024) by including measures of anxiety and well-being alongside measures of social communication.

Meta-Analysis

Where sufficient data were available, a meta-analysis was conducted to estimate the standardised mean difference (SMD) in BOSCC and ADOS scores pre- to post-intervention. Where available, means, standard deviations, and group size were used to compute g scores. Some studies reported Cohen’s d scores, which we converted to Hedges’ g to correct for small sample bias using the formula g = d × J, where J = 1 − 3 / (4N − 9). Confidence intervals were adjusted by applying the same correction factor to the reported bounds. For one study (Klein et al., 2021), effect size r values from Wilcoxon signed-rank tests were converted to approximate Cohen’s d using the formula d=2r/(1−r2). Hedges’ g, standard errors, and 95% confidence intervals (CIs) for g were computed using the method outlined above.

Random-effect (RE) models were used to account for between-study variability. To evaluate the sensitivity of each BOSCC outcome (Total, Social Communication, RRB) and ADOS outcome (Raw total, (Calibrated Severity Score) CSS total, Raw Social Affect, CSS Social Affect, Raw RRB, CSS RRB) to detect change across diverse early intervention contexts, we conducted separate within-group meta-analyses of standardised mean change (Hedges’ g) for intervention and control arms. This approach allows for the estimation of responsiveness by assessing the degree of pre–post change within each arm type, independent of between-group treatment effects. An RE model accounts for the fact that we might expect to see different effect sizes depending on the age of the sample or the intensity of the intervention. Intervention arms reflect the measure’s ability to detect treatment-related change, while control arms serve as a reference for stability over time. This analytic strategy is commonly used when interventions vary in content and dosage but share a common outcome measure (Sandbank et al., 2020; Velikonja et al., 2019), and it aligns with recommendations for evaluating responsiveness using pre–post effect sizes in meta-analysis (Borenstein et al., 2010). Separate pooling of arms also mitigates the confounding effects of study design variability while providing insight into whether the BOSCC can reliably detect clinically meaningful change. Heterogeneity was assessed using the I2 statistic. Analyses were performed in Python.

Results

A series of RE meta-analyses were conducted to evaluate the responsiveness of the BOSCC to capture meaningful behavioural change following a range of autism interventions, including Paediatric Autism Communication Therapy (PACT), Early Start Denver Model (ESDM), Improving Parents as Communication Teachers (ImPACT), and Caregiver-Mediated Intervention (CMI).

BOSCC Total

Across 18 intervention arms that included a total of 642 participants, the BOSCC total score showed a moderate and statistically significant change, with a pooled effect size of Hedges’ g = −0.32, 95% CI = [−0.42, −0.22], p < .001 (see Figure 2). This suggests that the BOSCC was sensitive to meaningful changes in how children engaged and interacted over time. Heterogeneity was relatively low (I2 = 22.15%), which indicates relatively stable change across the intervention arms. In contrast, 11 control arms that included a total of 443 participants showed a much smaller, non-significant change (g = −0.15, 95% CI = [−0.29, 0.00], p = .05), with larger heterogeneity (I2 = 52.58%), reflecting diversity in CAU programmes. The smaller but consistent change in the BOSCC for the control group likely reflects a combination of developmental changes and the impacts of CAU.

Figure 2.

A forest plot showing the effect sizes of various studies on BOSCC total scores with pooled intervention, control, and no effect clusters.

Open in a new tab

Forest plot of effects for BOSCC total scores.

BOSCC Social Communication Subscale

Analysis of the BOSCC Social Communication (SC) subscale across 13 intervention arms with a total of 447 participants also showed a statistically significant effect (g = −0.34, 95% CI = [−0.49, −0.19], p < .001), with moderate heterogeneity (I2 = 44.3%; see Figure 3). The SC subscale captures observable aspects of reciprocal communication and social engagement, including eye gaze, gestures, shared affect, and verbal and nonverbal initiations, within naturalistic interactions. The significant pooled effect suggests that many children showed growth in observable social communication behaviours following intervention. Control groups (k = 10) with a total of 311 participants also showed a smaller but statistically significant change (g = −0.21, 95% CI = [−0.42, −0.01], p = .04), and a higher level of heterogeneity across studies (I2 = 62.3%). Again, the smaller but consistent change in the BOSCC for the control group likely reflects a combination of developmental changes and the impacts of CAU.

Figure 3.

Forest plot showing BOSCC, comparing pooled control, intervention, and no effects for various studies.

Open in a new tab

Forest plot of effects for BOSCC Social Communication.

BOSCC RRB Subscale

The BOSCC RRB subscale, analysed across seven intervention arms with a total of 211 participants, showed a moderate effect, with relatively wide confidence intervals and moderate heterogeneity, indicating a high level of variability in this result (Hedges’ g = −0.40, 95% CI = [−0.74, −0.06], p = .02; I2 = 49.16%; see Figure 4). This domain includes observable patterns such as repetitive motor behaviours, echolalia, and sensory interests. The direction of change suggests that there was a decrease in RRBs over time in response to intervention, and effect sizes were consistent with changes on the SC subscale. In contrast, control arms (k = 4) with 140 participants showed no significant change in RRB scores and low heterogeneity across studies (g = −0.14, 95% CI = [−0.39, 0.11], p = .26; I2 = 4.4%). These findings reflect the relative stability of RRBs in short-term contexts and highlight the importance of interpreting change in these behaviours with nuance: many repetitive behaviours serve valuable self-regulatory, sensory, or communicative functions for autistic individuals and may not represent targets for change in all interventions.

Figure 4.

Clinical study results on BOSCC behaviours indicated varying effects of interventions, control, and no effect.

Open in a new tab

Forest plot of effects for BOSCC restricted and repetitive behaviours.

Comparison to ADOS Scores

To contextualise BOSCC responsiveness, parallel analyses were conducted on ADOS scores, which is more commonly used for diagnostic purposes than for capturing short-term developmental change. A sub-group of six studies (Carruthers et al., 2021; Fletcher-Watson et al., 2016; Freitag et al., 2025; Green et al., 2022; Pijl et al., 2018) was used for this analysis that contained scores for both the ADOS and the BOSCC, facilitating direct comparisons on the same datasets, with the same raters. Outcome measures that had less than four studies available for analysis are not reported. Given the small number of studies included for each outcome, it is important to interpret this analysis with caution; however, the overall pattern of results indicates that there is a striking similarity between the BOSCC and ADOS outcomes.

Comparing ADOS Total to BOSCC Total Score

For the ADOS total score, four intervention arms with 276 children showed a small but statistically significant change (g = −0.32, 95% CI = [−0.56, −0.07], p = .01; I2 = 46.52%). A similar magnitude of change was found using the BOSCC total score (g = −0.31, 95% CI = [−0.46, −0.14], p < .001), for five intervention arms with 320 children, from the same sub-group of studies. However, it is notable that the confidence intervals for the BOSCC scores are narrower.

Comparing ADOS Social Affect to BOSCC SC Score

For the ADOS Social Affect raw score, four intervention arms with 276 children showed significant change (g = −0.33, 95% CI = [−0.56, −0.11], p < .00; I2 = 66.46%). Similarly, the BOSCC SC decreased with a similar magnitude of change (g = −0.31, 95% CI = [−0.56, −0.07], p = .01), based on four studies with 199 children from the same sub-group of studies.

Comparing ADOS RRB to BOSCC RRB Score

Finally, the ADOS RRB scores showed evidence of intervention-related change. On raw RRB scores, four intervention arms with 276 children showed a small but significant effect (g = −0.26, 95% CI = [−0.43, −0.09], p = .003; I2 = 0.00%). Comparatively, the BOSCC RRB scores also showed a significant decrease (g = −0.29, 95% CI = [−0.55, −0.02], p = .03; I2 = 31.80%). It is important to interpret these findings with caution given the limited nature of the data; the ADOS RRB analysis has an I2 = 0.00, indicating that it is underpowered to compute heterogeneity. The confidence intervals of the BOSCC RRB are very wide and approaching 0, which indicates that the true difference could be much closer to zero.

Sensitivity Analysis

Results were robust to model specification and study exclusion across both BOSCC and ADOS outcomes. For BOSCC Total, intervention effects were small to moderate (RE g = −0.32, 95% CI = [−0.42, −0.22]) and consistent with fixed-effect estimates (FE g = −0.33, 95% CI = [−0.43, −0.23]); no single study altered the pooled estimate by more than 0.03. Similar stability was observed for BOSCC SC (RE g = −0.34, 95% CI = [−0.49, −0.19]; FE g = −0.36, 95% CI = [−0.49, −0.24]) and for Repetitive Behaviours effects were moderate but more variable (RE g = −0.40, 95% CI = [−0.74, −0.06]; FE g = −0.32, 95% CI = [−0.52, −0.12]), with leave-one-out shifts ⩽0.12.

For ADOS, results were likewise stable. ADOS Total, intervention effects were small to moderate (RE g = −0.27, 95% CI = [−0.46, −0.08]; FE g = −0.30, 95% CI = [−0.48, −0.12]), with leave-one-out shifts ⩽0.08. ADOS Social Affect intervention effects were similarly robust (RE g = −0.33, 95% CI = [−0.56, −0.11]; FE g = 0.35, 95% CI = [−0.52, −0.17]), with leave-one-out shifts ⩽0.12. ADOS RRB yielded smaller effects (RE g = −0.26, 95% CI = [−0.43, −0.09]; FE g = −0.26, 95% CI = [−0.43, −0.09]), with minimal influence from individual studies (⩽0.03). Control arms showed smaller effects in the same direction across outcomes.

Small-Study Effects

Small-study effects were assessed for BOSCC outcomes where sufficient studies were available (k ⩾ 10). There was no evidence of small-study or publication bias in either intervention or control arms. For BOSCC SC, neither Egger’s test (control: p = .40; intervention: p = .26) nor Begg’s test (control: p = .38; intervention: p = .31) indicated asymmetry. Similarly, for BOSCC Total, no evidence of small-study effects was observed (Egger’s test: control p = .25; intervention p = .37; Begg’s test: control p = .54; intervention p = .11). For other outcomes, including BOSCC Repetitive Behaviours and all ADOS measures, the number of studies per arm was insufficient (k < 10) to permit reliable assessment of small-study effects.

Discussion

This meta-analysis provides preliminary evidence that the BOSCC-MV subscales are sensitive to meaningful change following a range of autism interventions. Across intervention arms, the BOSCC Total and SC scales captured moderate improvements in reciprocal communication and social engagement, with effects that were both statistically significant and robust to sensitivity analyses. This is particularly encouraging given long-standing concerns about the inability of traditional diagnostic tools to detect short-term developmental or behavioural shifts (Grzadzinski et al., 2016; Pijl et al., 2018). The BOSCC RRB subscale also demonstrated that it is sensitive to change in a short time, showing a moderate decrease in RRBs at the group level; however, the very wide confidence intervals approaching zero indicate a high level of variability in the underlying effect and should be interpreted with caution. In addition, this meta-analysis provides a small but direct comparison, which indicates that there is a striking similarity between the BOSCC and ADOS in terms of their ability to detect change in intervention arms. Taken together, these results indicate that while the BOSCC may be useful for detecting intervention-specific change, this meta-analysis does not provide evidence to suggest that it is more sensitive than the ADOS measures. Additional studies are needed to confirm this finding more robustly.

Where individual studies do not show substantial differences between the intervention and control arm, it is possible that either the intervention is not effective, or perhaps the CAU is already robust, and additional interventions are not required. For example, in Husu et al. (2024), the lack of a detectable difference between groups may be attributable to the high intensity and quality of the CAU arm, which already included access to music therapy and occupational therapy and speech therapy and a specialised early learning curriculum. This underscores the importance of contextualising BOSCC findings in relation to baseline services received.

Given long-standing debates about whether RRBs should be considered intervention targets, this finding warrants closer consideration. Although a moderate pooled effect for the BOSCC RRB was observed, this result should be interpreted with caution. First, it is based on only seven studies, raising concerns of publication bias because null or negative findings in RRB may simply remain unpublished. Indeed, several studies in our rapid review noted limited or no change in repetitive behaviours, but they did not provide further details, so these results could not be formally included in the analysis. Second, where restrictive repetitive behaviours are present, they may not always be clinically relevant (Charlton et al., 2021; Kapp et al., 2019; I. F. Morris et al., 2025). Repetitive behaviours such as “stimming” are increasingly recognised as adaptive strategies that support self-regulation and sensory processing in many autistic individuals (Cho, 2024). This raises a broader question about the appropriateness of including RRBs in outcome measures for social communication. Many naturalistic developmental approaches do not target these behaviours, so it is unsurprising that little change was observed. Also, attempts to suppress stimming without recognising its adaptive value may conflict with neuro-affirmative principles and risk undermining well-being (Kapp et al., 2019; Pellicano et al., 2014).

Finally, RRBs are not unique to autistic populations (Kim & Lord, 2010), and many autistic children do not demonstrate restrictive repetitive behaviours, further undermining their usefulness both as a diagnostic marker and treatment outcome. The heterogeneity of presentation in RRBs leads to a skewed distribution (Kim et al., 2019; Kim & Lord, 2010), which increases the likelihood of floor effects; for example, if children do not engage in RRBs at baseline, reductions cannot be detected. Together, these considerations underscore the need for nuanced clinical interpretation that distinguishes between behaviours that warrant intervention and those that reflect neurodivergent identity, self-regulation, or adaptive coping. Repetitive behaviours may in fact be linked to markers of mental health such as Compulsivity and Intrusive Thought (McCabe and Howlin, under review), which is an indicator of distress. Approaches that support autistic individual to engage in repetitive behaviours that align with their goals or interests, for example, learning and practising music; crafting, baking or gardening, may be a more constructive way to manage repetitive behaviours rather than trying to bluntly reduce them.

An important point of comparison in this review is the performance of the BOSCC relative to the ADOS. The ADOS remains one of the most widely used instruments in autism assessment and research. Although raw ADOS scores have generally been unsuccessful in identifying changes in behaviour (Owley et al., 2001), more recent studies using the ADOS have been able to capture change over time that is statistically significant (Chung et al., 2025). The current meta-analysis provides a small but direct comparison, which indicates that there is a striking similarity between the BOSCC and ADOS in terms of their ability to detect change in intervention arms. This directly contradicts previous studies that indicate that the BOSCC is more sensitive to change than the ADOS (Byrne et al., 2024; Byrne & Lord, 2023; Grzadzinski et al., 2016); however, it is important to interpret this result with caution given the small number of studies included for each outcome. Nonetheless, the relatively stable effect sizes observed for the BOSCC suggest it is a potentially useful outcome measure that can generalise across implementation contexts.

Despite these encouraging findings, several limitations of this review must be acknowledged. First, although meta-analytic methods were used, not all studies reported sufficient data to estimate effect sizes for each subscale, and outcome reporting was inconsistent. Second, there was considerable heterogeneity across studies in terms of intervention type and intensity and huge variation in what was provided to children under CAU, which is not standardised. This diversity, while reflective of real-world practice, introduces variability that may influence effect estimates despite the use of RE models. For instance, in Husu et al. (2024), the lack of detectable difference between groups may be attributable to the high intensity and quality of the CAU arm. Third, in line with rapid review methodology, we did not conduct a formal risk of bias assessment. This means that some of the included studies have extremely small samples and wide confidence intervals, which limits the robustness of the findings. However, our sensitivity analyses indicated that the results are not impacted by any individual study, which indicates good consistency across studies. Although we found limited evidence of small-study bias in BOSCC analyses, the relatively small number of included studies, especially for the RRB domain, warrants cautious interpretation. Finally, the meta-analysis relies on within-group pre–post change rather than between-group comparisons. While this approach is useful in the current context because several studies do not have direct comparisons between the intervention group and control group (S. B. Morris & DeShon, 2002) and because the research studies are heterogeneous in design (Wilson & Lipsey, 2001), it limits the ability to draw conclusions about intervention effects and may inflate effect size estimates.

Together, these findings underscore the importance of domain-specific reporting when using the BOSCC. Composite scores may obscure critical intervention effects, especially when repetitive behaviours, is less responsive or is not a focus of the intervention. The SC subscale offers a psychometrically robust, ecologically valid measure of intervention-related change, while the RRB subscale may still provide useful insight in targeted trials but should be interpreted cautiously, especially when used in broader composites.

Future Directions

This review highlights several priority areas for future research and refinement in the use of the BOSCC as an outcome measure. Broader psychometric validation is needed across more diverse populations, including non-English-speaking groups and low-resource settings, to establish the generalisability of the tool and ensure that newer versions of the BOSCC (e.g., Phrase Speaker/Young Fluent PSYF, F1, F2) are appropriately tailored. A remaining concern is that inter-rater reliability, a critical component of the tool’s validity, is typically established within individual research teams or study contexts. As a result, it remains unclear how well BOSCC scoring practices transfer across research sites, raters, or cultural settings (Pijl et al., 2018; Reszka et al., 2024). This variability raises questions about the tool’s generalisability and standardisation, especially when used in multi-site trials or across international studies where coder training, supervision, and implementation may differ substantially. Standardisation of coding protocols and rater training, such as through the development of shared training materials, multi-site calibration processes, and reliability benchmarks, would support consistency and improve cross-study comparisons.

Furthermore, future trials should commit to transparent and comprehensive reporting of BOSCC outcomes, including null findings, and how they determine the reliability of coding to minimise reporting bias and support cumulative evidence synthesis. Particular attention should be given to the Repetitive and Restrictive Behaviour subscale, where further study is needed to clarify its external validity and explore how it may be adapted to differentiate between clinically impairing behaviours and those that serve adaptive, self-regulatory functions. Finally, as autism research increasingly embraces neurodiversity-oriented frameworks, it is essential that future work critically examine how BOSCC outcomes are interpreted in light of autistic perspectives. Engaging autistic individuals and advocacy communities in defining meaningful change will be key to ensuring that the BOSCC evolves in ways that are both scientifically valid and ethically aligned.

Footnotes

Ethical Considerations: This Rapid Review utilised only publicly available data extracted from previously published studies. Therefore, this project did not involve human subjects, and no ethical approval was required from the authors.

Consent to Participate: This Rapid Review utilised only publicly available data extracted from previously published studies. Therefore, this project did not involve human subjects, and no informed consent was required from the authors.

Author Contributions: Claire Howlin: Conceptualisation; Data curation; Formal analysis; Methodology; Software; Visualisation; Writing – original draft; Writing – review & editing.

Anna Kovalova: Conceptualisation; Data curation; Investigation; Writing – review & editing.

Artur Jaschke: Conceptualisation; Writing – review & editing.

David M Greenberg: Conceptualisation; Investigation; Writing – review & editing.

Beth Mason: Methodology; Validation.

Irene Pallás-Ferrer: Writing – review & editing.

Eve Merriam: Writing – review & editing.

Catherine Brushwood: Writing – review & editing.

Jonathan Pool: Writing – review & editing.

Carrie Allison: Conceptualisation; Writing – review & editing.

Helen Odell-Miller: Conceptualisation; Writing – review & editing.

Karen Hayden: Project administration.

Carmel Simone Moore: Project administration; Validation.

Rebecca Atkinson: Project administration.

Ramarni Treyvone Wilfred: Writing – review & editing.

Simon Baron-Cohen: Conceptualisation; Supervision; Writing – review & editing.

Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: C.H. received funding from Research Ireland NF/2024/11528. S.B.-C. received funding from the Wellcome Trust 214322\Z\18\Z. For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. The results leading to this publication have received funding from the Innovative Medicines Initiative 2 Joint Undertaking under grant agreement no. 777394 for the project AIMS-2-TRIALS. This Joint Undertaking receives support from the European Union’s Horizon 2020 research and innovation programme and EFPIA and AUTISM SPEAKS, Autistica, and SFARI. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results. S.B.-C. also received funding from the Autism Centre of Excellence, SFARI, the Templeton World Charitable Fund, and the MRC. All research at the Department of Psychiatry in the University of Cambridge was supported by the NIHR Cambridge Biomedical Research Centre (BRC-1215-20014) and the NIHR Applied Research Collaboration East of England. Any views expressed are those of the authors and not necessarily those of the funders, IHU-JU2, the NIHR, or the Department of Health and Social Care. SBC received funding from the Wellcome Trust 214322\Z\18\Z. SBC also received funding from the Innovative Medicines Initiative (IMI) 2 Joint Undertaking under grant agreement No 777394 for the project AIMS-2-TRIALS. This Joint Undertaking receives support from the European Union’s Horizon 2020 research and innovation programme and EFPIA and Autism Speaks, Autistica, and SFARI. SBC also received funding from the Autism Research Trust (now managed by Autism Action) and from SFARI, the European Commission, the Templeton World Charitable Fund, the Stoneygate Trust, the MRC, the Rosetrees Trust, the Eagles Autism Foundation, the K. Lisa Yang Centre for Autism Research at Cambridge/Yang Tan Collective, the Cambridge University Development and Alumni Relations (CUDAR), the NIHR Cambridge Biomedical Research Centre (NIHR203312), and the NIHR Applied Research Collaboration East of England. Any views expressed are those of the author(s) and not necessarily those of the funders (including IHI-JU2, or the NIHR or the Department of Health and Social Care). The funders had no role in the design of the study, in the collection, analyses, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results. For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. DMG was supported during this work by the Evans Lam Professorship in Music and Medicine and associated research startup funds provided through a philanthropic gift from Evans Lam to the Eastman School of Music, University of Rochester, and by the Ernest J. Del Monte Institute for Neuroscience, University of Rochester Medicine. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

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

Data Availability Statement

The data analysed for this rapid review are derived from published research articles and publicly accessible literature. All data, including descriptive statistics, effect sizes, and other quantitative information used to inform the findings and conclusions of this review, are available within the cited primary source articles listed in the reference section of this manuscript. All synthesis data tables created by the authors during the review process and the Python code used to analyse the data are available as a project on the Open Science Repository view-only link for review: https://osf.io/8uake/overview?view_only=9dab6439ccfc46a8a11e73a71e71dd82.

References

  1. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th edn). [Google Scholar]
  2. Borenstein M., Hedges L. V., Higgins J. P. T., Rothstein H. R. (2010). A basic introduction to fixed-effect and random-effects models for meta-analysis. Research Synthesis Methods, 1(2), 97–111. 10.1002/jrsm.12 [DOI] [PubMed] [Google Scholar]
  3. Botha M., Chapman R., Onaiwu M. G. K., Kapp S., Stannard A. (2024). The neurodiversity concept was developed collectively: An overdue correction on the origins of neurodiversity theory. Autism, 28(6), 1591–1594. 10.1177/13623613241237871 [DOI] [PubMed] [Google Scholar]
  4. Boyd B. A., Watson L. R., Reszka S. S., Sideris J., Alessandri M., Baranek G. T., Crais E. R., Donaldson A., Gutierrez A., Johnson L., Belardi K. (2018). Efficacy of the ASAP Intervention for Preschoolers with ASD: A Cluster Randomized Controlled Trial. Journal of Autism and Developmental Disorders, 48(9), 3144–3162. 10.1007/s10803-018-3584-z [DOI] [PubMed] [Google Scholar]
  5. Byrne K., Lord C. (2023). The Brief Observation of Social Communication Change (BOSCC): Procedures, strengths, limitations, and future directions. Archives of Paediatrics and Developmental Pathology, 6(1–3), 1026. [Google Scholar]
  6. Byrne K., Sterrett K., Holbrook A., Kim S. H., Grzadzinski R., Lord C. (2024). Extending the usefulness of the Brief Observation of Social Communication Change (BOSCC): Validating the phrase speech and young fluent version. Journal of Autism and Developmental Disorders, 54(3), 1009–1023. 10.1007/s10803-022-05877-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Carruthers S., Charman T., El Hawi N., Kim Y. A., Randle R., Lord C., Pickles A. (2021). Utility of the Autism Diagnostic Observation Schedule and the Brief Observation of Social and Communication Change for Measuring Outcomes for a Parent-Mediated Early Autism Intervention. Autism Research: Official Journal of the International Society for Autism Research, 14(2), 411–425. 10.1002/aur.2449 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Charlton R. A., Entecott T., Belova E., Nwaordu G. (2021). “It feels like holding back something you need to say”: Autistic and non-autistic adults accounts of sensory experiences and stimming. Research in Autism Spectrum Disorders, 89, 101864. 10.1016/j.rasd.2021.101864 [DOI] [Google Scholar]
  9. Cho A. (2024). 11.2 an autistic-led demonstration of strategies to help clinicians put autistic youth at ease and facilitate connection. Journal of the American Academy of Child & Adolescent Psychiatry, 63(10), S16. 10.1016/j.jaac.2024.07.098 [DOI] [Google Scholar]
  10. Chung E. Y., Kuen-fung Sin K., Chow D. H. (2025). Effectiveness of robotic intervention on improving social development and participation of children with autism spectrum disorder – A randomised controlled trial. Journal of Autism and Developmental Disorders, 55(2), 449–456. 10.1007/s10803-024-06236-2 [DOI] [PubMed] [Google Scholar]
  11. Divan G., Vajaratkar V., Cardozo P., Huzurbazar S., Verma M., Howarth E., Emsley R., Taylor C., Green J. (2018). The Feasibility and Effectiveness of PASS Plus, A Lay Health Worker Delivered Comprehensive Intervention for Autism Spectrum Disorders: Pilot RCT in a Rural Low and Middle Income Country Setting. AUTISM RESEARCH, 10.1186/s13063-023-07621–07625. [DOI] [PubMed] [Google Scholar]
  12. Doyle N. (2020). Neurodiversity at work: A biopsychosocial model and the impact on working adults. British Medical Bulletin, 135(1), 108–125. 10.1093/bmb/ldaa021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fletcher-Watson S., Petrou A., Scott-Barrett J., Dicks P., Graham C., O’Hare A., Pain H., McConachie H. (2016). A trial of an iPadTM intervention targeting social communication skills in children with autism. Autism, 20(7), 771–782. 10.1177/1362361315605624 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Frankel Heffler K., Frome L. R., Garvin B., Bungert L. M., Bennett D. S. (2022). Screen time reduction and focus on social engagement in autism spectrum disorder: A pilot study. Pediatrics International, 64(1), 1–10. CINAHL Ultimate. 10.1111/ped.15343 [DOI] [PubMed] [Google Scholar]
  15. Freitag C. M., Kirchner M., Sauer L. D., Kleber S. K., Polzer L., Raji N., Lemler C., Fröhlich U., Jarczok T., Geissler J., Radtke F., Ring M., Roessner V., Taurines R., Noterdaeme M., Teufel K., Kim Z., Kitzerow-Cleven J. (2025). Complex, low-intensity, individualised naturalistic developmental behavioural intervention in toddlers and pre-schoolers with autism spectrum disorder: The multicentre, observer-blind, parallel-group randomised-controlled A-FFIP trial. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 66, 1500–1513. 10.1111/jcpp.14162 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Frost K., Russell K., Ingersoll B. (2020). Using thin-slice ratings to measure social communication in children with autism spectrum disorder. Research in Autism Spectrum Disorders, 74, Article 101550. 10.1016/j.rasd.2020.101550 [DOI] [Google Scholar]
  17. Frost K. M., Koehn G. N., Russell K. M., Ingersoll B. (2019). Measuring child social communication across contexts: Similarities and differences across play and snack routines. Autism Research: Official Journal of the International Society for Autism Research, 12(4), 636–644. 10.1002/aur.2077 [DOI] [PubMed] [Google Scholar]
  18. Garritty C., Hamel C., Trivella M., Gartlehner G., Nussbaumer-Streit B., Devane D., Kamel C., Griebler U., King V. J. (2024). Updated recommendations for the Cochrane rapid review methods guidance for rapid reviews of effectiveness. BMJ, 384, Article e076335. 10.1136/bmj-2023-076335 [DOI] [PubMed] [Google Scholar]
  19. Gengoux G. W., Abrams D. A., Schuck R., Millan M. E., Libove R., Ardel C. M., Phillips J. M., Fox M., Frazier T. W., Hardan A. Y. (2019). A pivotal response treatment package for children with autism spectrum disorder: An RCT. Pediatrics, 144(3), Article e20190178. 10.1542/peds.2019-0178 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Geoffray M.-M., Bourgeois-Mollier M., Maleysson-Baste M., Gallifet N., Dochez S., Bonis G., Jay A., Jurek L. (2025). Feasibility of a videoconferencing-based parent-mediated intervention: A mixed-method pilot study. Frontiers in Psychology, 15, 1450455. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Green J., Leadbitter K., Ellis C., Taylor L., Moore H. L., Carruthers S., James K., Taylor C., Balabanovska M., Langhorne S., Aldred C., Slonims V., Grahame V., Parr J., Humphrey N., Howlin P., McConachie H., Couteur A. L., Charman T., …Pickles A. (2022). Combined social communication therapy at home and in education for young autistic children in England (PACT-G): A parallel, single-blind, randomised controlled trial. The Lancet. Psychiatry, 9(4), 307–320. 10.1016/S2215-0366(22)00029-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Grzadzinski R. (2016). Preliminary reliability and validity of the brief observation of social communication change. Journal of the American Academy of Child and Adolescent Psychiatry, 55(10), S301–S301. 10.1016/j.jaac.2016.07.279 [DOI] [Google Scholar]
  23. 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(7), 2464–2479. 10.1007/s10803-016-2782-9 [DOI] [PubMed] [Google Scholar]
  24. Husu E., Helminen T. M., Muuvila M., Lauttia J., Eriksson K., Nevalainen J., Kylliäinen A. (2024). Parent-mediated eye gaze intervention in young autistic children: ADOS-BOSCC outcome study [Preprint]. Research Square. 10.21203/rs.3.rs-4424436/v1 [DOI]
  25. Ingersoll B., Wainer A. L., Berger N. I., Pickard K. E., Bonter N. (2016). Comparison of a self-directed and therapist-assisted telehealth parent-mediated intervention for children with ASD: A pilot RCT. Journal of Autism and Developmental Disorders, 46(7), 2275–2284. 10.1007/s10803-016-2755-z [DOI] [PubMed] [Google Scholar]
  26. Jaschke A. C., Howlin C., Pool J., Greenberg Y. D., Atkinson R., Kovalova A., Merriam E., Pallás-Ferrer I., Williams S., Moore C., Hayden K., Allison C., Odell-Miller H., Baron-Cohen S. (2024). Study protocol of a randomized control trial on the effectiveness of improvisational music therapy for autistic children. BMC Psychiatry, 24(1), 637. 10.1186/s12888-024-06086-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kaale A., Fagerland M. W., Martinsen E. W., Smith L. (2014). Preschool-based social communication treatment for children with autism: 12-month follow-up of a randomized trial. Journal of the American Academy of Child and Adolescent Psychiatary, 53(2), 188–198. [DOI] [PubMed] [Google Scholar]
  28. Kapp S. K., Steward R., Crane L., Elliott D., Elphick C., Pellicano E., Russell G. (2019). ‘People should be allowed to do what they like’: Autistic adults’ views and experiences of stimming. Autism: The International Journal of Research and Practice, 23(7), 1782–1792. 10.1177/1362361319829628 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kasari C., Lawton K., Shih W., Barker T. V., Landa R., Lord C., Orlich F., King B., Wetherby A., Senturk D. (2014). Caregiver-Mediated Intervention for Low-Resourced Preschoolers With Autism: An RCT. Pediatrics, 134(1), e72–e79. 10.1542/peds.2013-3229 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kim S. H., Grzadzinski R., Martinez K., Lord C. (2019). Measuring treatment response in children with autism spectrum disorder: Applications of the Brief Observation of Social Communication Change to the Autism Diagnostic Observation Schedule. Autism: The International Journal of Research and Practice, 23(5), 1176–1185. 10.1177/1362361318793253 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kim S. H., Lord C. (2010). Restricted and repetitive behaviors in toddlers and preschoolers with autism spectrum disorders based on the Autism Diagnostic Observation Schedule (ADOS). Autism Research, 3(4), 162–173. 10.1002/aur.142 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. King V. J., Stevens A., Nussbaumer-Streit B., Kamel C., Garritty C. (2022). Paper 2: Performing rapid reviews. Systematic Reviews, 11(1), 151. 10.1186/s13643-022-02011-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Klein C. B., Swain D. M., Vibert B., Clark-Whitney E., Lemelman A. R., Giordano J. A., Winter J., Kim S. H. (2021). Implementation of Video Feedback Within a Community Based Naturalistic Developmental Behavioral Intervention Program for Toddlers With ASD: Pilot Study. Frontiers in Psychiatry, 12. 10.3389/fpsyt.2021.763367 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Mazurek M. O., Carlson C., Baker-Ericzén M., Butter E., Norris M., Kanne S. (2020). Construct validity of the Autism Impact Measure (AIM). Journal of Autism and Developmental Disorders, 50(7), 2307–2319. 10.1007/s10803-018-3462-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Morris I. F., Sykes J. R., Paulus E. R., Dameh A., Razzaque A., Esch L. V., Gruenig J., Zelazo P. D. (2025). Beyond self-regulation: Autistic experiences and perceptions of stimming. Neurodiversity, 3, 27546330241311096. 10.1177/27546330241311096 [DOI] [Google Scholar]
  36. Morris S. B., DeShon R. P. (2002). Combining effect size estimates in meta-analysis with repeated measures and independent-groups designs. Psychological Methods, 7(1), 105–125. 10.1037/1082-989X.7.1.105 [DOI] [PubMed] [Google Scholar]
  37. Nordahl-Hansen A., Fletcher-Watson S., McConachie H., Kaale A. (2016). Relations between specific and global outcome measures in a social-communication intervention for children with autism spectrum disorder. Research in Autism Spectrum Disorders, 29–30, 19–29. 10.1016/j.rasd.2016.05.005 [DOI]
  38. Oosterling I., Visser J., Swinkels S., Rommelse N., Donders R., Woudenberg T., Roos S., Van Der Gaag R. J., Buitelaar J. (2010). Randomized Controlled Trial of the Focus Parent Training for Toddlers with Autism: 1-Year Outcome. Journal of Autism and Developmental Disorders, 40(12), 1447–1458. 10.1007/s10803-010-1004-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Owley T., Mcmahon W., Cook E. H., Laulhere T., South M., Mays L. Z., Shernoff E. S., Lainhart J., Modahl C. B., Corsello C., Ozonoff S., Risi S., Lord C., Leventhal B. L., Filipek P. A. (2001). Multisite, double-blind, placebo-controlled trial of porcine secretin in autism. Journal of the American Academy of Child & Adolescent Psychiatry, 40(11), 1293–1299. 10.1097/00004583-200111000-00009 [DOI] [PubMed] [Google Scholar]
  40. Pellecchia M., Ingersoll B., Marcus S. C., Rump K., Xie M., Newman J., Zeigler L., Crabbe S., Straiton D., Chávez E. C. (2024). Pilot randomized trial of a caregiver-mediated naturalistic developmental behavioral intervention in part C early intervention. Journal of Early Intervention, 46(2), 194–216. [Google Scholar]
  41. Pellicano E., Dinsmore A., Charman T. (2014). Views on researcher-community engagement in autism research in the United Kingdom: A mixed-methods study. PLOS ONE, 9(10), Article e109946. 10.1371/journal.pone.0109946 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Pijl M. K., Rommelse N. N., Hendriks M., De Korte M. W., Buitelaar J. K., Oosterling I. J. (2018). Does the Brief Observation of Social Communication Change help moving forward in measuring change in early autism intervention studies? Autism: The International Journal of Research and Practice, 22(2), 216–226. 10.1177/1362361316669235 [DOI] [PubMed] [Google Scholar]
  43. Reszka S. S., Wallisch A., Boyd B. A., Watson L. R., Grasley-Boy N. (2024). Initial examination of use of the Brief Observation of Social-Communication Change (BOSCC) across home and school contexts. Infant and Child Development, 33(6), Article e2547. 10.1002/icd.2547 [DOI] [Google Scholar]
  44. Rogers S., Estes A., Lord C., Munson J., Rocha M., Winter J., Greenson J., Colombi C., Dawson G., Vismara L. A., Sugar C. A., Hellemann G., Whelan F., Talbott M. (2019). A Multisite Randomized Controlled Two-Phase Trial of the Early Start Denver Model Compared to Treatment as Usual. Journal of the American Academy of Child & Adolescent Psychiatry, 58(9), 853–865. 10.1016/j.jaac.2019.01.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Salomone E., Settanni M., McConachie H., Suma K., Ferrara F., Foletti G., Salandin A., Brown F. L., Pacione L., Shire S., Servili C., Adamson L. B. WHO CST Team. (2022). Pilot Randomized Controlled Trial of the WHO Caregiver Skills Training in Public Health Services in Italy. Journal of Autism and Developmental Disorders, 52(10), 4286–4300. 10.1007/s10803-021-05297-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sandbank M., Bottema-Beutel K., Crowley S., Cassidy M., Dunham K., Feldman J. I., Crank J., Albarran S. A., Raj S., Mahbub P., Woynaroski T. G. (2020). Project AIM: Autism intervention meta-analysis for studies of young children. Psychological Bulletin, 146(1), 1–29. 10.1037/bul0000215 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Settanni M., Suma K., Adamson L. B., McConachie H., Servili C., Salomone E. (2024). Treatment mechanism of the WHO caregiver skills training intervention for autism delivered in community settings. Autism Research: Official Journal of the International Society for Autism Research, 17(1), 182–194. 10.1002/aur.3058 [DOI] [PubMed] [Google Scholar]
  48. Shaffer R., Thurman A. J., Ronco L., Cadavid D., Raines S., Kim S. H. (2022). Social communication in fragile X syndrome: Pilot examination of the Brief Observation of Social Communication Change (BOSCC). Journal of Neurodevelopmental Disorders, 14(1), 4. 10.1186/s11689-021-09411-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Stevens A., Hersi M., Garritty C., Hartling L., Shea B. J., Stewart L. A., Welch V. A., Tricco A. C. (2025). Rapid review method series: Interim guidance for the reporting of rapid reviews. BMJ Evidence-Based Medicine, 30(2), 118–123. 10.1136/bmjebm-2024-112899 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Swain D., Li Y., Brown H. R., Petkova E., Lord C., Rogers S. J., Estes A., Kasari C., Kim S. H. (2024). Implementing a Uniform Outcome Measurement Approach for Early Interventions of Autism Spectrum Disorders. Journal of the American Academy of Child and Adolescent Psychiatry, S0890-8567(24)00351-4. 10.1016/j.jaac.2024.06.004 [DOI] [PubMed]
  51. Tricco A. C., Antony J., Zarin W., Strifler L., Ghassemi M., Ivory J., Perrier L., Hutton B., Moher D., Straus S. E. (2015). A scoping review of rapid review methods. BMC Medicine, 13(1), Article 224. 10.1186/s12916-015-0465-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Velikonja T., Fett A.-K., Velthorst E. (2019). Patterns of nonsocial and social cognitive functioning in adults with autism spectrum disorder: A systematic review and meta-analysis. JAMA Psychiatry, 76(2), 135. 10.1001/jamapsychiatry.2018.3645 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wetherby A. M., Guthrie W., Woods J., Schatschneider C., Holland R. D., Morgan L., Lord C. (2014). Parent-Implemented Social Intervention for Toddlers With Autism: An RCT. Pediatrics, 134(6), 1084–1093. 10.1542/peds.2014-0757 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wilson D. B., Lipsey M. W. (2001). The role of method in treatment effectiveness research: Evidence from meta-analysis. Psychological Methods, 6(4), 413–429. 10.1037/1082-989X.6.4.413 [DOI] [PubMed] [Google Scholar]
  55. Woodard C. (2024). Brief Observation of Social Communication Change to Evaluate Change in Autism Symptoms as a Treatment Response in a Semistructured Play Assessment. [Google Scholar]

Associated Data

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

Data Availability Statement

The data analysed for this rapid review are derived from published research articles and publicly accessible literature. All data, including descriptive statistics, effect sizes, and other quantitative information used to inform the findings and conclusions of this review, are available within the cited primary source articles listed in the reference section of this manuscript. All synthesis data tables created by the authors during the review process and the Python code used to analyse the data are available as a project on the Open Science Repository view-only link for review: https://osf.io/8uake/overview?view_only=9dab6439ccfc46a8a11e73a71e71dd82.

Articles from Autism are provided here courtesy of SAGE Publications

RESOURCES