The telehealth coaching intervention was informed by the Community-Early Start Denver Model (C-ESDM) materials. Twelve session-specific visuals with simple-text (for caregivers) and session scripts (for non-specialist coaches) were developed by the research team. Over 12, 1-hour asynchronous caregiver coaching sessions that are delivered by non-specialists, caregivers were coached in strategies to increase child attention to their social world, increase child communication, create and build joint activity routines, and understand and use the ABCs to support the development of new behaviors. The session itself being delivered via phone. Prior to the coaching session staff shared session materials with the family via WhatsApp. Prior to the session the caregiver shared a 6-min video of an interaction with their child. During the session, the caregiver and coach reviewed the video, and the caregiver completed a self-reflection checklist.

Caregiver-child dyadic exchanges measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates no interaction between the dyad, a rating of 7 indicates a balanced dyadic exchange that flows naturally back and forth.

Caregiver-child dyadic exchanges measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates no interaction between the dyad, a rating of 7 indicates a balanced dyadic exchange that flows naturally back and forth.

Caregiver strategies measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates that the caregiver rarely uses the caregiver strategy to support the child's activity and language. A rating of 7 indicates that the caregiver frequently uses appropriate strategies to support and expand their child's activity and language.

Caregiver strategies measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates that the caregiver rarely uses the caregiver strategy to support the child's activity and language. A rating of 7 indicates that the caregiver frequently uses appropriate strategies to support and expand their child's activity and language.

Caregiver strategies measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates that the caregiver rarely uses the caregiver strategy to support the child's activity and language. A rating of 7 indicates that the caregiver frequently uses appropriate strategies to support and expand their child's activity and language.

Joint attention skills will be measured using the Early Social Communication Scales (ESCS) in the ASD group; the typically developing group; and the developmental delay group. The ESCS assesses nonverbal communication skills, is normed for typically developing children 18 to 30 months of age in addition to children with developmental delay expressive language level is in approximately the same range. Frequency scores for initiation of joint attention (IJA) will be derived from behavioral observations during a series of tasks presented by an examiner blind to child diagnosis. IJA score ranges from 0 to 100 where a higher value indicates greater initiation of joint attention.

Joint attention skills will be measured using the Early Social Communication Scales (ESCS) in the ASD group; the typically developing group; and the developmental delay group. The ESCS assesses nonverbal communication skills, is normed for typically developing children 18 to 30 months of age in addition to children with developmental delay expressive language level is in approximately the same range. Frequency scores for response to joint attention (RJA) will be derived from behavioral observations during a series of tasks presented by an examiner blind to child diagnosis. RJA score ranges from 0 to 20 where a higher value indicates greater response to joint attention.

The Griffiths Scales of Child Development, Third Edition (Griffiths III) is a comprehensive, developmental measure for continuous use from birth to 5 years 11 months (71 months). Although not standardized in South Africa (standardization samples are from the United Kingdom and Ireland), this developmental assessment is widely used in South Africa. The Griffiths III provides an overall measure of a child's development, as well as an individual profile of strengths and needs across five areas: Foundations of Learning; Language and Communication; Eye and Hand Coordination; Personal-Social-Emotional; and Gross Motor. Developmental quotients (DQs) were calculated by (Developmental Age/Chronological Age) *100. Griffiths III DQs range from 1 to 100 where higher scores indicate that the child performs at or near expected chronological age as compared with a normative sample of children the same age.

The Griffiths Scales of Child Development, Third Edition (Griffiths III) is a comprehensive, developmental measure for continuous use from birth to 5 years 11 months (71 months). Although not standardized in South Africa (standardization samples are from the United Kingdom and Ireland), this developmental assessment is widely used in South Africa. The Griffiths III provides an overall measure of a child's development, as well as an individual profile of strengths and needs across five areas: Foundations of Learning; Language and Communication; Eye and Hand Coordination; Personal-Social-Emotional; and Gross Motor. Developmental quotients (DQs) were calculated by (Developmental Age/Chronological Age) *100. Griffiths III DQs range from 1 to 100 where higher scores indicate that the child performs at or near expected chronological age as compared with a normative sample of children the same age.

The Vineland Adaptive Behavior Scales, Third Edition (Vineland-3) is an individually-administered measure of adaptive behavior that is widely used to assess individuals with intellectual, developmental, and other disabilities. The Vineland-3 contains 5 domains each with 2-3 subdomains. The main domains are: Communication, Daily Living Skills, Socialization, Motor Skills, and Maladaptive Behavior. The Caregiver Interview Form uses the Vineland semistructured interview technique to elicit information about the examinee's adaptive functioning from a parent or caregiver. Item responses are collected on a 3-point Likert scale with values representing 0 (never), 1 (sometimes), and 2 (usually or often) to capture frequency of target behavior. Some items require binary responses (yes/no). The VABS-3 Socialization subscale score ranges from 20-140 where a higher score indicates greater frequency of target behavior.

The Vineland Adaptive Behavior Scales, Third Edition (Vineland-3) is an individually-administered measure of adaptive behavior that is widely used to assess individuals with intellectual, developmental, and other disabilities. The Vineland-3 contains 5 domains each with 2-3 subdomains. The main domains are: Communication, Daily Living Skills, Socialization, Motor Skills, and Maladaptive Behavior. The Caregiver Interview Form uses the Vineland semistructured interview technique to elicit information about the examinee's adaptive functioning from a parent or caregiver. Item responses are collected on a 3-point Likert scale with values representing 0 (never), 1 (sometimes), and 2 (usually or often) to capture frequency of target behavior. Some items require binary responses (yes/no). The VABS-3 Communication subscale score ranges from 20-140 where a higher score indicates greater frequency of target behavior.

A 5-item scale that measures the extent to which a new treatment, or an innovation, can be successfully used or carried out within a given agency or setting. Scores are indicated on a scale of 1-5 with higher scores indicating greater appropriateness.

A 5-item scale that measures the extent to which a new treatment, or an innovation, can be successfully used or carried out within a given agency or setting. Scores are indicated on a scale of 1-5 with higher scores indicating greater feasibility.

ESDM Fidelity Rating System uses a 13 item rating scale that includes ratings of performance on core intervention strategies from 1 to 5 (1 represents a lack of an effective display of the practices specified, and 5 represents the best possible example of this teaching behavior). Fidelity of implementation of Naturalistic Developmental Behavioral Intervention (NDBI) strategies will be manually coded by certified ESDM therapists. Mean fidelity scores (with SD) will be calculated across study participants in order to assess change in NDBI strategies. ESDM Fidelity Rating System scores range from 0 to 100 with a higher score reflecting interactions closer to ESDM principles.

SenseToKnow is an app developed for mobile devices that is designed to elicit autism risk behaviors. In order to administer the app, children sit on the parent's lap with the mobile device at eye level for the child. The app is comprised of a set of stimuli (videos and games) that are designed to elicit autism spectrum disorder (ASD) symptoms, such as facial expressions. Multi-modal time-series video-frame-level data streams from the computer vision analysis output of SenseToKnow will be summarized with descriptive statistics and visualizations. We extract landmarks and determine the correct features from them to yield precise measurements of attention, gaze, affect, and posture. Each computer vision analysis feature will be transformed into one or more binary, count, or continuous participant-level summary variable (for e.g. number of times orienting to name).

SenseToKnow is an app developed for mobile devices that is designed to elicit autism risk behaviors. In order to administer the app, children sit on the parent's lap with the mobile device at eye level for the child. The app is comprised of a set of stimuli (videos and games) that are designed to elicit autism spectrum disorder (ASD) symptoms, such as facial expressions. Multi-modal time-series video-frame-level data streams from the computer vision analysis output of SenseToKnow will be summarized with descriptive statistics and visualizations. We extract landmarks and determine the correct features from them to yield precise measurements of attention, gaze, affect, and posture. Each computer vision analysis feature will be transformed into one or more binary, count, or continuous participant-level summary variable (for e.g. number of times orienting to name).

SenseToKnow is an app developed for mobile devices that is designed to elicit autism risk behaviors. In order to administer the app, children sit on the parent's lap with the mobile device at eye level for the child. The app is comprised of a set of stimuli (videos and games) that are designed to elicit autism spectrum disorder (ASD) symptoms, such as facial expressions. Multi-modal time-series video-frame-level data streams from the computer vision analysis output of SenseToKnow will be summarized with descriptive statistics and visualizations. We extract landmarks and determine the correct features from them to yield precise measurements of attention, gaze, affect, and posture. Each computer vision analysis feature will be transformed into one or more binary, count, or continuous participant-level summary variable (for e.g. number of times orienting to name).

Participants in this arm will be children with ASD. Early Start Denver Model (ESDM): Twelve one hour sessions, skill is introduced by "Help is in your hands" Community ESDM videos. Session skills include specific strategies caregivers can use to increase their child's attention to people and their child's communication; strategies to identify, create and use joint activity routines to engage with and teach their child; in addition to ways to use the ABCs of learning to teach their child new behaviors. "Help is in your hands" Community ESDM materials also provide video examples of other caregivers using the session skill with their child. The early childhood development practitioner then coaches the caregiver in the session skill, in at least two activities. The goal of coaching was to provide enough support for the caregiver to experience success in implementing the new skill with their child. Coaching is followed by caregiver reflection and a discussion of generalization of the new skill across various daily activities, using a daily activities visual aid.

Participants in this arm will be caregivers of children with ASD who began the intervention in-person and moved to telehealth due to COVID-19 restrictions.

In-person: Twelve 1-hour sessions, skill is introduced by "Help is in your hands" Community Early Start Denver Model (C-ESDM) videos. Skills include specific strategies caregivers can use to increase their child's attention to people and their child's communication; strategies to identify, create and use joint activity routines to engage with and teach their child; in addition to ways to use the ABCs of learning to teach their child new behaviors. The ECD practitioner then coaches the caregiver in at least two activities. Coaching is followed by caregiver reflection and a discussion of generalization of the new skill using a daily activities visual aid.

Telehealth coaching: The telehealth coaching intervention was informed by the C-ESDM materials. Twelve session-specific visuals with simple-text (for caregivers) and session scripts (for non-specialist coaches) were developed by the research team. Over 12, 1-hour asynchronous caregiver coaching sessions that are delivered by non-specialists, caregivers were coached as above (via phone). Prior to the coaching session staff shared session materials with the family via WhatsApp and the caregiver shared a 6-min video of an interaction with their child. During the session, the caregiver and coach reviewed the video, and the caregiver completed a self-reflection checklist.

Participants in this arm will be children with ASD who began the intervention in-person and moved to telehealth due to COVID-19 restrictions.

In-person: Twelve 1-hour sessions, skill is introduced by "Help is in your hands" Community Early Start Denver Model (C-ESDM) videos. Skills include specific strategies caregivers can use to increase their child's attention to people and their child's communication; strategies to identify, create and use joint activity routines to engage with and teach their child; in addition to ways to use the ABCs of learning to teach their child new behaviors. The ECD practitioner then coaches the caregiver in at least two activities. Coaching is followed by caregiver reflection and a discussion of generalization of the new skill using a daily activities visual aid.

Telehealth coaching: The telehealth coaching intervention was informed by the C-ESDM materials. Twelve session-specific visuals with simple-text (for caregivers) and session scripts (for non-specialist coaches) were developed by the research team. Over 12, 1-hour asynchronous caregiver coaching sessions that are delivered by non-specialists, caregivers were coached as above (via phone). Prior to the coaching session staff shared session materials with the family via WhatsApp and the caregiver shared a 6-min video of an interaction with their child. During the session, the caregiver and coach reviewed the video, and the caregiver completed a self-reflection checklist.

Participants in this arm will be caregivers of children with ASD receiving telehealth intervention (adaptation due to COVID-19 restrictions).

Telehealth coaching: The telehealth coaching intervention was informed by the Community-Early Start Denver Model (C-ESDM) materials. Twelve session-specific visuals with simple-text (for caregivers) and session scripts (for non-specialist coaches) were developed by the research team. Over 12, 1-hour asynchronous caregiver coaching sessions that are delivered by non-specialists, caregivers were coached in strategies to increase child attention to their social world, increase child communication, create and build joint activity routines, and understand and use the ABCs to support the development of new behaviors. The session itself being delivered via phone. Prior to the coaching session staff shared session materials with the family via WhatsApp. Prior to the session the caregiver shared a 6-min video of an interaction with their child. During the session, the caregiver and coach reviewed the video, and the caregiver completed a self-reflection checklist.

Participants in this arm will be children with ASD receiving telehealth intervention (adaptation due to COVID-19 restrictions).

Telehealth coaching: The telehealth coaching intervention was informed by the Community-Early Start Denver Model (C-ESDM) materials. Twelve session-specific visuals with simple-text (for caregivers) and session scripts (for non-specialist coaches) were developed by the research team. Over 12, 1-hour asynchronous caregiver coaching sessions that are delivered by non-specialists, caregivers were coached in strategies to increase child attention to their social world, increase child communication, create and build joint activity routines, and understand and use the ABCs to support the development of new behaviors. The session itself being delivered via phone. Prior to the coaching session staff shared session materials with the family via WhatsApp. Prior to the session the caregiver shared a 6-min video of an interaction with their child. During the session, the caregiver and coach reviewed the video, and the caregiver completed a self-reflection checklist.

Participants in this arm will be children with typical development. No intervention.

Participants in this arm will be children with developmental delay. No intervention.

The ECD worker and/or school supervisor are 1) employed by our participating recruitment partners (Western Cape Education Department Schools) and 2) involved in delivery of the caregiver coaching sessions, either in person or remotely.

The local supervisors are 1) trained ESDM therapist, 2) supervise weekly coaching sessions in the schools, and/or 3) supervise remote intervention delivery sessions.

Total of all reporting groups

Children only. Data not collected on ECD Practitioners and ECD Practitioner School Supervisors or ESDM Supervisors groups.

Child engagement states measured by the Joint Engagement Rating Inventory (JERI) in the ASD group; the typically developing group; and the developmental delay group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates a very low rating and no time spent in that engagement state, 4 is the midpoint indicating the moderate time and quality of episodes in an engagement state, and 7 indicates a very high rating and high-quality engagement.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD, Children with typical development, and Children with developmental delay groups. Data only collected at baseline for Children with Typical Development and Children with Developmental Delay.

Child engagement states measured by the Joint Engagement Rating Inventory (JERI) in the ASD group; the typically developing group; and the developmental delay group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates a very low rating and no time spent in that engagement state, 4 is the midpoint indicating the moderate time and quality of episodes in an engagement state, and 7 indicates a very high rating and high-quality engagement.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD, Children with typical development, and Children with developmental delay groups. Data only collected at baseline for Children with Typical Development and Children with Developmental Delay.

Caregiver-child dyadic exchanges measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates no interaction between the dyad, a rating of 7 indicates a balanced dyadic exchange that flows naturally back and forth.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD, Children with typical development, and Children with developmental delay groups. Data only collected at baseline for Children with Typical Development and Children with Developmental Delay.

Caregiver-child dyadic exchanges measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates no interaction between the dyad, a rating of 7 indicates a balanced dyadic exchange that flows naturally back and forth.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD, Children with typical development, and Children with developmental delay groups. Data only collected at baseline for Children with Typical Development and Children with Developmental Delay.

Caregiver strategies measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates that the caregiver rarely uses the caregiver strategy to support the child's activity and language. A rating of 7 indicates that the caregiver frequently uses appropriate strategies to support and expand their child's activity and language.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD, Children with typical development, and Children with developmental delay groups. Data only collected at baseline for Children with Typical Development and Children with Developmental Delay.

Caregiver strategies measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates that the caregiver rarely uses the caregiver strategy to support the child's activity and language. A rating of 7 indicates that the caregiver frequently uses appropriate strategies to support and expand their child's activity and language.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD, Children with typical development, and Children with developmental delay groups. Data only collected at baseline for Children with Typical Development and Children with Developmental Delay.

Caregiver strategies measured by the Joint Engagement Rating Inventory (JERI) in the ASD group. Joint Engagement Rating Inventory contains eighteen 7-point Likert scale items that characterized various aspects of joint engagement. The items were designed to span the range of possibilities likely to be observed during interactions with 18- to 30-month-old TD children as well as similarly aged and older children with developmental difficulties, including ASD. A rating of 1 indicates that the caregiver rarely uses the caregiver strategy to support the child's activity and language. A rating of 7 indicates that the caregiver frequently uses appropriate strategies to support and expand their child's activity and language.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD, Children with typical development, and Children with developmental delay groups. Data only collected at baseline for Children with Typical Development and Children with Developmental Delay.

Joint attention skills will be measured using the Early Social Communication Scales (ESCS) in the ASD group; the typically developing group; and the developmental delay group. The ESCS assesses nonverbal communication skills, is normed for typically developing children 18 to 30 months of age in addition to children with developmental delay expressive language level is in approximately the same range. Frequency scores for initiation of joint attention (IJA) will be derived from behavioral observations during a series of tasks presented by an examiner blind to child diagnosis. IJA score ranges from 0 to 100 where a higher value indicates greater initiation of joint attention.

Only applicable to Caregiver-Child Dyads With ASD, Children with typical development, and Children with developmental delay groups. This measure was introduced later in the study and therefore not administered to all participants. Data not collected on the Caregiver-Child Dyads With ASD (Telehealth Intervention) group.

Joint attention skills will be measured using the Early Social Communication Scales (ESCS) in the ASD group; the typically developing group; and the developmental delay group. The ESCS assesses nonverbal communication skills, is normed for typically developing children 18 to 30 months of age in addition to children with developmental delay expressive language level is in approximately the same range. Frequency scores for response to joint attention (RJA) will be derived from behavioral observations during a series of tasks presented by an examiner blind to child diagnosis. RJA score ranges from 0 to 20 where a higher value indicates greater response to joint attention.

Only applicable to Caregiver-Child Dyads With ASD, Children with typical development, and Children with developmental delay groups. This measure was introduced later in the study and therefore not administered to all participants. Data not collected on the Caregiver-Child Dyads With ASD (Telehealth Intervention) group.

The Griffiths Scales of Child Development, Third Edition (Griffiths III) is a comprehensive, developmental measure for continuous use from birth to 5 years 11 months (71 months). Although not standardized in South Africa (standardization samples are from the United Kingdom and Ireland), this developmental assessment is widely used in South Africa. The Griffiths III provides an overall measure of a child's development, as well as an individual profile of strengths and needs across five areas: Foundations of Learning; Language and Communication; Eye and Hand Coordination; Personal-Social-Emotional; and Gross Motor. Developmental quotients (DQs) were calculated by (Developmental Age/Chronological Age) *100. Griffiths III DQs range from 1 to 100 where higher scores indicate that the child performs at or near expected chronological age as compared with a normative sample of children the same age.

Participants who completed the study. Only applicable to child groups. Data not collected on Caregiver-Child Dyads With ASD (Telehealth Intervention) - Children group. Data only collected at baseline for Children with Typical Development and Children with Developmental Delay.

The Griffiths Scales of Child Development, Third Edition (Griffiths III) is a comprehensive, developmental measure for continuous use from birth to 5 years 11 months (71 months). Although not standardized in South Africa (standardization samples are from the United Kingdom and Ireland), this developmental assessment is widely used in South Africa. The Griffiths III provides an overall measure of a child's development, as well as an individual profile of strengths and needs across five areas: Foundations of Learning; Language and Communication; Eye and Hand Coordination; Personal-Social-Emotional; and Gross Motor. Developmental quotients (DQs) were calculated by (Developmental Age/Chronological Age) *100. Griffiths III DQs range from 1 to 100 where higher scores indicate that the child performs at or near expected chronological age as compared with a normative sample of children the same age.

Participants who completed the study. Only applicable to child groups. Data not collected on Caregiver-Child Dyads With ASD (Telehealth Intervention) - Children group. Data only collected at baseline for Children with Typical Development and Children with Developmental Delay.

The Vineland Adaptive Behavior Scales, Third Edition (Vineland-3) is an individually-administered measure of adaptive behavior that is widely used to assess individuals with intellectual, developmental, and other disabilities. The Vineland-3 contains 5 domains each with 2-3 subdomains. The main domains are: Communication, Daily Living Skills, Socialization, Motor Skills, and Maladaptive Behavior. The Caregiver Interview Form uses the Vineland semistructured interview technique to elicit information about the examinee's adaptive functioning from a parent or caregiver. Item responses are collected on a 3-point Likert scale with values representing 0 (never), 1 (sometimes), and 2 (usually or often) to capture frequency of target behavior. Some items require binary responses (yes/no). The VABS-3 Socialization subscale score ranges from 20-140 where a higher score indicates greater frequency of target behavior.

Participants who completed the study. Only applicable to the Caregiver-Child Dyads With ASD groups.

The Vineland Adaptive Behavior Scales, Third Edition (Vineland-3) is an individually-administered measure of adaptive behavior that is widely used to assess individuals with intellectual, developmental, and other disabilities. The Vineland-3 contains 5 domains each with 2-3 subdomains. The main domains are: Communication, Daily Living Skills, Socialization, Motor Skills, and Maladaptive Behavior. The Caregiver Interview Form uses the Vineland semistructured interview technique to elicit information about the examinee's adaptive functioning from a parent or caregiver. Item responses are collected on a 3-point Likert scale with values representing 0 (never), 1 (sometimes), and 2 (usually or often) to capture frequency of target behavior. Some items require binary responses (yes/no). The VABS-3 Communication subscale score ranges from 20-140 where a higher score indicates greater frequency of target behavior.

Participants who completed the study. Only applicable to the Caregiver-Child Dyads With ASD groups.

A 5-item scale that measures the extent to which a new treatment, or an innovation, can be successfully used or carried out within a given agency or setting. Scores are indicated on a scale of 1 - 5 with higher scores indicating greater acceptability.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD (Telehealth Intervention) - Caregivers, Early Childhood Development (ECD) Practitioners and ECD Practitioner School Supervisors, and Early Start Denver Model (ESDM) Supervisors groups.

A 5-item scale that measures the extent to which a new treatment, or an innovation, can be successfully used or carried out within a given agency or setting. Scores are indicated on a scale of 1-5 with higher scores indicating greater appropriateness.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD (Telehealth Intervention) - Caregivers, Early Childhood Development (ECD) Practitioners and ECD Practitioner School Supervisors, and Early Start Denver Model (ESDM) Supervisors groups.

A 5-item scale that measures the extent to which a new treatment, or an innovation, can be successfully used or carried out within a given agency or setting. Scores are indicated on a scale of 1-5 with higher scores indicating greater feasibility.

Participants who completed the study. Only applicable to Caregiver-Child Dyads With ASD (Telehealth Intervention) - Caregivers, Early Childhood Development (ECD) Practitioners and ECD Practitioner School Supervisors, and Early Start Denver Model (ESDM) Supervisors groups.

ESDM Fidelity Rating System uses a 13 item rating scale that includes ratings of performance on core intervention strategies from 1 to 5 (1 represents a lack of an effective display of the practices specified, and 5 represents the best possible example of this teaching behavior). Fidelity of implementation of Naturalistic Developmental Behavioral Intervention (NDBI) strategies will be manually coded by certified ESDM therapists. Mean fidelity scores (with SD) will be calculated across study participants in order to assess change in NDBI strategies. ESDM Fidelity Rating System scores range from 0 to 100 with a higher score reflecting interactions closer to ESDM principles.

Participants who completed the study. Only applicable to the Caregiver-Child Dyads With ASD Caregiver groups.

SenseToKnow is an app developed for mobile devices that is designed to elicit autism risk behaviors. In order to administer the app, children sit on the parent's lap with the mobile device at eye level for the child. The app is comprised of a set of stimuli (videos and games) that are designed to elicit autism spectrum disorder (ASD) symptoms, such as facial expressions. Multi-modal time-series video-frame-level data streams from the computer vision analysis output of SenseToKnow will be summarized with descriptive statistics and visualizations. We extract landmarks and determine the correct features from them to yield precise measurements of attention, gaze, affect, and posture. Each computer vision analysis feature will be transformed into one or more binary, count, or continuous participant-level summary variable (for e.g. number of times orienting to name).

SenseToKnow is an app developed for mobile devices that is designed to elicit autism risk behaviors. In order to administer the app, children sit on the parent's lap with the mobile device at eye level for the child. The app is comprised of a set of stimuli (videos and games) that are designed to elicit autism spectrum disorder (ASD) symptoms, such as facial expressions. Multi-modal time-series video-frame-level data streams from the computer vision analysis output of SenseToKnow will be summarized with descriptive statistics and visualizations. We extract landmarks and determine the correct features from them to yield precise measurements of attention, gaze, affect, and posture. Each computer vision analysis feature will be transformed into one or more binary, count, or continuous participant-level summary variable (for e.g. number of times orienting to name).

SenseToKnow is an app developed for mobile devices that is designed to elicit autism risk behaviors. In order to administer the app, children sit on the parent's lap with the mobile device at eye level for the child. The app is comprised of a set of stimuli (videos and games) that are designed to elicit autism spectrum disorder (ASD) symptoms, such as facial expressions. Multi-modal time-series video-frame-level data streams from the computer vision analysis output of SenseToKnow will be summarized with descriptive statistics and visualizations. We extract landmarks and determine the correct features from them to yield precise measurements of attention, gaze, affect, and posture. Each computer vision analysis feature will be transformed into one or more binary, count, or continuous participant-level summary variable (for e.g. number of times orienting to name).

SenseToKnow is an app developed for mobile devices that is designed to elicit autism risk behaviors. In order to administer the app, children sit on the parent's lap with the mobile device at eye level for the child. The app is comprised of a set of stimuli (videos and games) that are designed to elicit autism spectrum disorder (ASD) symptoms, such as facial expressions. Multi-modal time-series video-frame-level data streams from the computer vision analysis output of SenseToKnow will be summarized with descriptive statistics and visualizations. We extract landmarks and determine the correct features from them to yield precise measurements of attention, gaze, affect, and posture. Each computer vision analysis feature will be transformed into one or more binary, count, or continuous participant-level summary variable (for e.g. number of times orienting to name).

Participants in this arm will be children with ASD. Early Start Denver Model (ESDM): Twelve one hour sessions, skill is introduced by "Help is in your hands" Community ESDM videos. Session skills include specific strategies caregivers can use to increase their child's attention to people and their child's communication; strategies to identify, create and use joint activity routines to engage with and teach their child; in addition to ways to use the ABCs of learning to teach their child new behaviors. "Help is in your hands" Community ESDM materials also provide video examples of other caregivers using the session skill with their child. The early childhood development practitioner then coaches the caregiver in the session skill, in at least two activities. The goal of coaching was to provide enough support for the caregiver to experience success in implementing the new skill with their child. Coaching is followed by caregiver reflection and a discussion of generalization of the new skill across various daily activities, using a daily activities visual aid.

Participants in this arm will be caregivers of children with ASD who began the intervention in-person and moved to telehealth due to COVID-19 restrictions.

In-person: Twelve 1-hour sessions, skill is introduced by "Help is in your hands" Community Early Start Denver Model (C-ESDM) videos. Skills include specific strategies caregivers can use to increase their child's attention to people and their child's communication; strategies to identify, create and use joint activity routines to engage with and teach their child; in addition to ways to use the ABCs of learning to teach their child new behaviors. The ECD practitioner then coaches the caregiver in at least two activities. Coaching is followed by caregiver reflection and a discussion of generalization of the new skill using a daily activities visual aid.

Telehealth coaching: The telehealth coaching intervention was informed by the C-ESDM materials. Twelve session-specific visuals with simple-text (for caregivers) and session scripts (for non-specialist coaches) were developed by the research team. Over 12, 1-hour asynchronous caregiver coaching sessions that are delivered by non-specialists, caregivers were coached as above (via phone). Prior to the coaching session staff shared session materials with the family via WhatsApp and the caregiver shared a 6-min video of an interaction with their child. During the session, the caregiver and coach reviewed the video, and the caregiver completed a self-reflection checklist.

Participants in this arm will be children with ASD who began the intervention in-person and moved to telehealth due to COVID-19 restrictions.

In-person: Twelve 1-hour sessions, skill is introduced by "Help is in your hands" Community Early Start Denver Model (C-ESDM) videos. Skills include specific strategies caregivers can use to increase their child's attention to people and their child's communication; strategies to identify, create and use joint activity routines to engage with and teach their child; in addition to ways to use the ABCs of learning to teach their child new behaviors. The ECD practitioner then coaches the caregiver in at least two activities. Coaching is followed by caregiver reflection and a discussion of generalization of the new skill using a daily activities visual aid.

Telehealth coaching: The telehealth coaching intervention was informed by the C-ESDM materials. Twelve session-specific visuals with simple-text (for caregivers) and session scripts (for non-specialist coaches) were developed by the research team. Over 12, 1-hour asynchronous caregiver coaching sessions that are delivered by non-specialists, caregivers were coached as above (via phone). Prior to the coaching session staff shared session materials with the family via WhatsApp and the caregiver shared a 6-min video of an interaction with their child. During the session, the caregiver and coach reviewed the video, and the caregiver completed a self-reflection checklist.

Participants in this arm will be caregivers of children with ASD receiving telehealth intervention (adaptation due to COVID-19 restrictions).

Telehealth coaching: The telehealth coaching intervention was informed by the Community-Early Start Denver Model (C-ESDM) materials. Twelve session-specific visuals with simple-text (for caregivers) and session scripts (for non-specialist coaches) were developed by the research team. Over 12, 1-hour asynchronous caregiver coaching sessions that are delivered by non-specialists, caregivers were coached in strategies to increase child attention to their social world, increase child communication, create and build joint activity routines, and understand and use the ABCs to support the development of new behaviors. The session itself being delivered via phone. Prior to the coaching session staff shared session materials with the family via WhatsApp. Prior to the session the caregiver shared a 6-min video of an interaction with their child. During the session, the caregiver and coach reviewed the video, and the caregiver completed a self-reflection checklist.

Participants in this arm will be children with ASD receiving telehealth intervention (adaptation due to COVID-19 restrictions).

Telehealth coaching: The telehealth coaching intervention was informed by the Community-Early Start Denver Model (C-ESDM) materials. Twelve session-specific visuals with simple-text (for caregivers) and session scripts (for non-specialist coaches) were developed by the research team. Over 12, 1-hour asynchronous caregiver coaching sessions that are delivered by non-specialists, caregivers were coached in strategies to increase child attention to their social world, increase child communication, create and build joint activity routines, and understand and use the ABCs to support the development of new behaviors. The session itself being delivered via phone. Prior to the coaching session staff shared session materials with the family via WhatsApp. Prior to the session the caregiver shared a 6-min video of an interaction with their child. During the session, the caregiver and coach reviewed the video, and the caregiver completed a self-reflection checklist.

Participants in this arm will be children with typical development. No intervention.

Participants in this arm will be children with developmental delay. No intervention.

The ECD worker and/or school supervisor are 1) employed by our participating recruitment partners (Western Cape Education Department Schools) and 2) involved in delivery of the caregiver coaching sessions, either in person or remotely.

The local supervisors are 1) trained ESDM therapist, 2) supervise weekly coaching sessions in the schools, and/or 3) supervise remote intervention delivery sessions.