In Singapore, Autism Spectrum Disorders (ASD) is ranked number one in disease burden for children 0-14 years of age. The Child Development Unit at the National University Hospital serves 3000 children annually, of which 25-30% of children have been diagnosed with ASD. Therapist roles are to provide interim therapy for these children before entry into community-based Early Intervention Centres (EIPIC), which currently have waiting times of 6-9 months. Current limitations with interim care includes long wait times, high cost for families, lack of manpower and space to serve the patients, poor parental involvement due to their work commitments, parental difficulties attending frequent, needed, in-hospital therapy and difficulty generalizing patient treatment to the home/community setting (decreasing effectiveness). The proposed Telerehabilitation (also called Telerehab) initiative involves the use of video conferencing technology to help address the aforementioned deficits. Offering early intervention through Telerehab will enable previously unattainable benefits such as seeing the child in their home environment, allowing multiple caregivers to have access to the early intervention training, more frequent contact with families and the ability to trouble shoot real life difficulties in real time. The important advantages to the caregivers include less financial burden arising from time off from work and travel, more access to treatment over a longer period of time and ability to access a multidisciplinary team. An additional benefit for the children is they need not travel to unfamiliar environments, which is frequently distressing for children with ASD. Lastly, Telerehab is a sustainable initiative allowing for less manpower to cover the growing number of patients, and the possibility to be implemented in other government run hospitals and clinics facing similar challenges. Elaboration of benefits:1) Importance of parent and caregiver empowerment. Early Intervention in the current model has been predominantly centre based with initiatives to increase caregiver education. A large body of literature suggests that early intervention is highly successful when provided at the age of diagnosis, with younger children yielding better outcomes. Caregiver involvement is vital to long-term success, as they spend a significant amount of time with their child; they can support the generalizations of new skills. National Research Council identifies parent training to be the key component for successful intervention for children with autism. Parent training improves quality of life by reducing parental stress and increasing optimism.2) Addressing nationally identified gaps. The Enabling Master plan recommendations for 2012-2016 (under Ministry of Family and Social Development) identifies gaps in family involvement and support in acquiring necessary skills and knowledge to be competent in helping their children make developmental gains. Child Development Unit (CDU) envisions that Telerehab is a viable avenue for supporting parents in learning EI skills.3) Improving existing parent training programmes. CDU has successfully piloted a parent-training program for children with ASD called SPEECCH. In our study of the impact of this parent-training program, children made measurable progress in all four skill areas assessed (p<0.001). Focus on achievable and observable family- centred developmental goals showed evidence for increased parental understanding of children's learning and behaviour amp; effective use of strategies for facilitating communication and interactions to support their child's development (p<0.001). However this intervention service could not be sustained due to high caseload demands and insufficient manpower. Parent interviews during review visits identified having sustained contact with therapists and parent coaching to be key areas of need. Currently the service provides intervention for 24 children with ASD weekly for one hour across 12 weeks, and continued support for up to 20 weeks (maximum of 16 hours of intervention). Of the new referrals of 150 children with ASD, if a sustained service is to be provided, only a small group of children will receive intervention. In order to address the demand, the frequency and intensity of intervention has had to be sacrificed to be able to provide some service to all patients. Hence to maximize the impact of early intervention, a sustainable model of service delivery using technology through videoconferencing is being proposed.

In Singapore, Autism Spectrum Disorders (ASD) is ranked number one in disease burden for children 0-14 years of age. The Child Development Unit at the National University Hospital serves 3000 children annually, of which 25-30% of children have been diagnosed with ASD. Therapist roles are to provide interim therapy for these children before entry into community-based Early Intervention Centres (EIPIC), which currently have waiting times of 6-9 months. Current limitations with our interim care includes long wait times, high cost for families, lack of manpower and space to serve the patients, poor parental involvement due to their work commitments, parental difficulties attending frequent, needed, in-hospital therapy and difficulty generalizing patient treatment to the home/community setting (decreasing effectiveness). Currently, clinic-based environment does not include natural environments for learning development and hence spontaneous generalization is a challenge and impacts child outcomes. Standard interventions do not allow therapists to observe or facilitate a child's behaviour and skills during natural routines. Previous work on telerehab show that it ensures the best outcome for the child. It provides the best modality for therapists to observe the child during these routines in home environments and provide meaningful and relevant strategies in real time. Despite standard therapy, expected significant outcomes of parent training have not been achieved. The key factor has been the need to provide parents with more instruction and practice with the content inside real-life moments and interactions with children. Telerehab facilitates the provision of more sustained opportunities for parent coaching within the home setting (natural environment). A clinic based model of intervention requires transition of children into new environments. The intolerance of children with Autism to changes in their routine/environment is well known; typically presented with acute symptoms of agitation or anxiety. They are frequently distressed in clinical environments, making it difficult for the parent and professionals to continue with the early intervention program. Telerehab provides early intervention in safe and secure natural home contexts of the child thus averting the challenges of transition. Early intervention services provided in the clinical setting are often provided with only one caregiver present. Logistics and travel costs prevent participation of multiple caregivers in early intervention coaching programs. Telerehab makes it feasible for multiple caregivers to participate in early intervention and build their capacities through teleconferencing support from professionals. Previous models of home-based interventions have found to incur high costs for the provider with limited reach. Home visits by professionals incur expense in the form of travel time, travel costs and hence the number of families covered through this is limited. Telerehab maximizes impact of home based interventions without high costs and with wider reach. The cost of early intervention(EI) through Telerehab will be similar to the current costs of clinic-based interventions paid by families. Cost incurred is essentially technology investment and professional manpower investment. Current duration of EI is 16 hours, which is sustainable up to 20 weeks for each family based on current CDU caseloads of 150 new referrals per year. Telerehab proposes to sustain the core intervention across 47 weeks for a duration of 16 hours leading to better sustainability, which also contributes to gains in productivity. This demonstrates the scalability of the Telerehab program.

The current standard program for early intervention treatment is in-clinic therapy based on the Foundational Skills Curriculum (FSC): a framework for early intervention developed from outcomes of an Autism research project conducted in the UK. This framework provides a clear and systematic approach to understanding the child's functioning in 3 core areas of development (across 141 items): Play, Social Interaction, and Communication.

The telerehab program will provide parent coaching on EI through video conferencing using the FSC. The telerehab program aims to improve access to services and facilitate intervention in the natural environment of children and their families by developing video conferencing (VC) as a feasible and acceptable medium to build parent capabilities in EI. The specific service model objectives are to realize positive gains in service efficiency (i.e. reducing the total time and resource utilization per completed intervention course relative to current practice); reduce total costs and non-financial barriers for patients (i.e. access) as well as the total cost to the health system per completed course (i.e. sustainability) and realize positive gains in provider productivity and hence cost-effectiveness (i.e. reduced cost per unit outcome achieved).

Implementation Strategies and Patient Flow CDU receives referrals from polyclinics for children at risk of ASD. These referrals will be scheduled for a developmental assessment by the developmental paediatrician. Baseline assessments will be conducted under the program as well as under standard care. Children who are enrolled into the evaluation study will then continue with either the standard program (clinic-based therapy) or the telerehab program. The program schedule for the patient in terms of type of visits and intervention blocks for both the standard program and telerehab is shown in Annex 1: Program Schedule. The duration of clinic-based intervention sessions will be 60 min while the duration of VC based intervention sessions will be 45 min.

Baseline Assessments Visit 0/1: Child found to be at risk for ASD on clinical evaluation by the developmental paediatrician based on the DSM-5 criteria will be referred for this study. The informed consent will be obtained prior to undergoing any research-related activities.

Visit 2 to 3: Upon parent consent in Visit 0 or 1, the child will be assessed on the MSEL and ADOS. However, MSEL assessment may be conducted before informed consent as part of clinical routine practice. Participants will also be assessed on the Vineland Adaptive Behaviour Scales (VABS-III) and Parenting Stress Index-Short Form (PSI-SF). Children who meet the criteria for Autism on the ADOS and do not have any of the exclusion criteria will be offered to be part of the study.

Visit 4: On completion of the above assessments, the paediatrician will share the MSEL and ADOS report with parents and confirm enrolment into the study if the child meets the inclusion criteria. If the child does not, they will be referred to standard care outside the study. Children whose parents consent to be on the study will be randomly assigned to either the telerehab program (experimental arm) or standard program (control arm) using block randomisation. The paediatrician will inform parents of the study arm that they are assigned to.

Visit 5a: Initial assessments based on the FSC, Joint Engagement Rating Inventory (JERI), child's routines (parent interview) and Family Early Intervention Quality of life (FEIQoL) will be completed for all. Parents will also participate in a short demographic and cost survey. If parents are unable to complete the surveys during their clinic-based session, they will be sent the surveys through email or asked to answer the questions over a phone call. Parents may also be invited to participate in an (semi structured, in-depth) interview for evaluating the perceived effectiveness of standard care.

Visit 5b: For the telerehab group, initial assessments and observations will be collected through 2 settings, one clinic-based, as well as one using VC. The latter is done in order to observe parent-child interactions and skills demonstrated by the child in the context of the natural home environment. The VC assessment and intervention will be conducted using a MOHH commissioned VC platform. This platform will allow the therapist and parents to see, hear, and communicate with one another live in real-time. The therapist will access the program from an office computer with a web-camera in the clinic, whereas parents will access the program from their home. Parents who do not have the necessary equipment for VC will be provided with notebook computers and needed accessories. An orientation to the use and care of VC equipment will be provided.

Visit 6: Parents of children in both programs will attend a parent education workshop. The workshop will discuss the role of parents and therapists in early intervention, the key areas that children have difficulties with during development, the use of routines to facilitate the child's learning and strategies that can be embedded in routines. If parents choose to not attend the workshop or are unable to attend their scheduled workshop, parents can still proceed with the intervention phase. Parents who are unable to attend the workshop will be reminded to reschedule the workshop to the next available date and will have training slides and materials sent to them through email for their reading. Parents will have also have opportunities to seek further clarification during subsequent visits with their therapist.

Intervention: Standard Program The profile of the child in terms of their developmental status based on initial evaluation results and an individualised intervention plan (IIP) will be discussed by the therapist with parents. The therapist will coach parents in the context of clinic-based play activities and help parents identify contexts and activities at home where the parents could follow up on the program at home. Children and their parents will receive the standard program which consists of 16 clinic-based intervention sessions of 60 min each, separated into 3 intervention blocks, with breaks in between so that parents will have opportunities to practise at home. This arrangement also allows parents who had missed their scheduled sessions to have make-up sessions during the breaks. Intervention sessions empower and equip parents with skills and strategies to engage with their child at home. Since the program involves behavioural intervention and parent coaching, active participation of parent and child are key to each session. However, realistic challenges in terms of child/parent or therapist being unwell, or when the parent has work related commitments, may require rescheduling or cancellation of some visits. These are not anticipated to have any adverse impact on the child's health or the study.

Intervention: Telerehab Program The intervention will commence with 2 clinic-based intervention sessions of 60 min each to provide adequate opportunity to engage in relational family-centred practices, joint decision and planning of the IIP based on initial evaluations and explore effective strategies as foundation to the subsequent interactions and coaching through VC.

Setting: Prior to intervention sessions, the therapist and parent will discuss and identify a suitable and feasible location to place the devices of their choice that would provide clear unobstructed views of parent-child interactions and activities. For 16 sessions, the therapist and parent will log onto the MOHH VC portal for the VC call.

VC sessions are provided remotely by the child's therapist from the clinic. Parents who do not have the necessary equipment for VC will be provided notebook computers and omnidirectional microphones. The MOHH commissioned VC platform will be used for observing the home session and coaching parents. Parents will orient the computer's web cam to the selected activity area in the home while using the wireless headset for communication. An omnidirectional portable microphone will be placed in the vicinity of the routine home activity so that the therapist can look, listen and understand the dynamics of parent child interactions clearly during a given daily routine involving the child.

Telerehab sessions will follow the ethical and practice guidelines of computer-mediated intervention including online communication, consent, confidentiality amp; privacy and security issues, verbal and nonverbal feedback.

Strategies: Parents will be trained in using evidence-based strategies (e.g., follow their child's lead, imitation; modelling, communicative temptations, playful obstructions) during playtime activities and daily routine.

Approach: Intervention sessions will be provided by therapists trained in the FSC and the parent coaching protocol. Parent-child interactions during activities at home will be observed through the VC platform comissioned by MOHH. Parents will be encouraged to interact with their child using strategies aimed at increasing their child's attention and motivation, turn-taking routines, initiating and responding to joint attention, communication, etc. Sessions will follow the collaborative practice protocol. The protocol provides an outline for the session structure (i.e. checking in, reviewing, observation, reflection, explanation and coaching of new topics and strategies, setting goals and follow-up planning in the home context and closing).

At the start of the session, the parents' progress from the past topic and any other updates will be reviewed and discussed for 5-10 minutes. This is then followed by a 10-minute parent-child play activity. The activity allows the therapist to observe the parent's competency in implementing intervention (i.e. the goals and effective use of strategies) from the previous session's coaching. Based on observations made, the therapist will coach further to strengthen parent's confidence and competency in using strategies to facilitate the attainment of specific learning objectives in the IIP. During the final 10-15 minutes of each session, the therapist will seek feedback from the parent to understand parent's clarity, comfort and confidence in implementing the intervention plan and provide further input and information if required. The therapist will also address any challenges faced or anticipated in the context of implementing the intervention in the home context. The therapist will discuss with parents to implement the intervention plan using at least two play or daily routines (naturally occurring opportunities) identified by the parent. Adult learning principles will be adopted by the therapist to facilitate active participation in planning and intervention implementation. For clinic-based sessions, an additional 15 minutes is allocated to allow the child and parents to settle down and transit into and out of the therapist's room.

Data collection: For all subjects, video recordings of parent child interactions will be taken at 4 specific time points, namely Baseline, two Midterm reviews and Final review. Specific to the children in the telerehab arm, an additional video recording of parent child interaction will be done through VC for each on-site review session, totalling 8 video recording time points throughout the program's duration. Recorded videos will be downloaded and transferred to a password encrypted hard drive authorized by the institution data protection and security teams. These recordings can only be accessed by authorized team members with institutional accounts and external raters. They will be used for evaluation on the JERI, NDBI-Fi and for inter-rater reliability assessment.

Parent-delivered therapy (Home-based) At the end of each intervention session and at the end of each intervention block, a joint discussion will be held to craft a plan for parents to implement. This will include the functional goals to be embedded, specific strategies to be used that the parent has been coached in and the home routines that the parent plans to implement. Intervention related resources such as description of strategies and examples of embedding within routines and documents on the plan will be shared with parents for their reference.

Program Timeline While set timeframes for assessments, intervention and reviews are important, this study will be implemented under routine operational conditions of the clinical service. The standard practice of having the same therapist for each session for continuity will be adopted. Scheduling challenges are an unavoidable field condition. This is compounded by the need for rescheduling of sessions at times, such as when a child/family member or therapist is unwell. For these reasons, the duration of time to complete each intervention block may vary slightly for each child. It may also vary between the intervention and control groups.

In general, where delays from scheduling occur, participants will move to the next phase of the study (i.e. intervention and follow-up reviews) only when they have completed all planned sessions in the previous phase (e.g., intervention will only begin after all baseline assessments have been completed). These are recognised as limitations of the study design and will be considered when interpreting the study findings.

Follow-up and Review (Standard Program and Telerehab Program) Midterm Progress Review: Visit 7 to Visit 11 Paediatrician Review (two visits): The paediatrician will review the child's general developmental profile, progress and child and family needs. The clinician will make appropriate recommendations and referrals as needed.

Therapist Review (two visits): Therapists will review the child's developmental profile on the FSC and JERI to track progress on the intervention plan and for program evaluation purposes. A cost survey will also be conducted with parents during these reviews by a member of the study team to identify any changes to household income and cost burden. If parents are unable to complete the survey during their session, they will be sent the survey through email or asked to answer the questions over a phone call.

Midpoint IIP review: Therapists will review the child on the individualized intervention plan (IIP) developed after baseline assessment and plan for further intervention.

Final Reviews and Assessments: Visit 12 to 14 The final reviews and assessments will be conducted across three sessions, which can take place in any order depending on the participant's availability and therapist's schedule.

Final Review with Therapist: The child will be evaluated on the baseline assessment measures. Parents will also be given a parent satisfaction survey, cost survey and FEIQoL survey to complete. If parents are unable to complete the surveys during their clinic-based session, they will be sent the surveys through email or asked to answer the questions over a phone call. For children in the telerehab group, there will be a VC review in addition to the clinic-based assessment.

Nurse Assessment: The child will be evaluated on MSEL. The Parental Stress Index will be administered as part of this assessment.

Psychologist Assessment: VABS-III will be administered to the parent for this assessment.

Final Paediatrician review: Visit 15 This last paediatrician visit is for the paediatrician to share the reports on the evaluations, discuss progress and further needs of child and family.

Transfer to Early Intervention Programs CDU provides interim therapy for children pending their enrolment into an EIPIC (or private early intervention service). Under normal circumstances, the child does not receive any intervention from EIPIC in the first 10 weeks on enrolment as the child is being assessed. In this transition period, the child continues to receive CDU intervention per standard clinical practice. As such, children in this study will continue to receive intervention for 10 weeks after EIPIC enrolment. Similarly, participants who have been transferred to a private intervention service will continue to receive intervention per the study protocol as long as they have not started receiving intervention at the private service.

Children also continue to remain as CDU patients under the care of their doctors until age 7, regardless of whether they have been enrolled to an EIPIC or private service. As such, study participants who have been transferred to an EIPIC will still be assessed per protocol for Midterm Progress Reviews and Final Reviews and Assessments. As far as possible, these assessments will be scheduled to coincide with their doctor's follow-up at the CDU as part of standard care.

(i) The study is designed as a non-inferiority study to compare the changes in MSEL between 2 programs.

The investigators expected that the change in MSEL score for the subjects in telerehab program arm will be slighter better than that in the standard program arm by 1.5 units. Assuming SD of changes in MSEL score be 15 units, and 5 units as equivalence range, it is sufficient to recruit 132 subjects (66 in each arm) with 80% power and 5% significant level. The investigators decided to recruit 200 subjects in total (100 in each arm) after considering withdrawals and lost to follow up. Block randomisation will be used to allocate the recruited subjects into one of the interventions. A randomisation list will be generated by the study statistician and envelopes will be prepared. 10 parents will be recruited from both groups for semi-structured in-depth interviews. The investigators aim to recruit a total of 20 parents for the interview as it is expected to be a sufficient number to reach qualitative thematic saturation. Up to 5 therapists involved in delivering therapy will be involved in a Focus Group Discussion. The investigators will conduct in-depth interview with the program staffs who are involved in coordinating and/or overseeing the care of patients recruited to the study (Paediatrician, Study Coordinator, Patient Service Associate). The investigators will recruit up to a maximum of 10 staff. (ii) As the study is conducted as part of routine care and operations, there are practical constraints that may cause protocol deviations to occur. Rescheduling of appointments by the parents (e.g. due to work constraints, child's sickness, other events) or defaulting appointments is inevitable. Hence, for per-protocol analyses, only participants with major protocol deviation, which will be defined as participants who do not complete 80% of the intended therapy sessions allocated for their study arm, will be excluded. This means that a control arm participant must attend at least 8 out of 10 clinic-based sessions. For the remaining 2 sessions, participants can either be attending it through video-conferencing (due to the COVID-19 pandemic) or miss the appointment. For the randomized trial, since the study is a non-inferiority study, statistical analysis will be carried out on a per-protocol basis. Subjects with any major protocol deviation as defined above will be excluded from the final analysis. All the demographic, clinical and cost characteristics will be analyzed descriptively. Mean (SD) or median (range), whichever is more appropriate will be reported for numerical variables, while N (%) will be reported for categorical variables.2-sample T test, or Mann-Whitney U test will be used to compare the change of MSEL score between 2 arms. Mean difference and 95% CI will be estimated. If the lower boundary of the 95% CI is more than -5, non-inferiority will be declared. Changes in the quality of parent child engagement, parental stress index score, and adaptive score will be compared by using either 2-sample T test, or Mann-Whitney U test, whichever is more appropriate. Linear regression will be used to adjust for any co-variates.

To examine cost-effectiveness, the investigators will compare resource utilization between treatment and control groups to estimate resource and cost differences in absolute terms from the direct medical cost perspective as well as the societal perspective (inclusive of the direct costs of travel and the indirect cost of parental time). As the trial focuses on non-inferiority, the assessment of cost-effectiveness will rely on whether the intervention is cost-saving relative to the standard of care. The investigators will also perform modelling estimates to quantify the uncertainty surrounding the incremental cost-effectiveness ratio.

For the interviews, applied thematic analysis would be used to analyse the qualitative data gathered to understand the concerns and experience of the programme from the perspectives of the parents. A professional transcriptionist will complete data transcription. A standard operating guideline and format of the verbatim will be established jointly by the principal researcher and the transcriptionist. Subsequently, the qualitative analysis research team will perform an iterative and inductive process of identifying new themes and sub-themes derived from participants' narratives. A codebook that best describe the data would be developed and used through coding and constant comparison to establish inter-coder reliability by the team for the analysis of all transcripts.

Inclusion Criteria:

1. Children aged 15-48 months
2. Children meet cut off score for Autism Spectrum Disorders (ASD) on Autism Diagnostic Observation Schedule-2 (ADOS): The Autism Diagnostic Observation Schedule (ADOS) is a semi-structured assessment of communication, social interaction, and play (or imaginative use of materials) for individuals suspected of having autism or other pervasive developmental disorders. The ADOS consists of a toddler module and four other modules, each of which is appropriate for children and adults of differing developmental and language levels, ranging from nonverbal to verbally-fluent. The ADOS consists of standardized activities that allow the examiner to observe the occurrence or non-occurrence of behaviours that have been identified as important to the diagnosis of autism and other pervasive developmental disorders across developmental levels and chronological ages.
3. Parent(s) is/are willing and able to give informed consent
4. Families with at least one parent who is digitally literate with the home use of the internet and access to Wi-Fi
5. The same parent(s) or caregiver(s) in attendance for most intervention sessions and all review sessions in order to monitor performance across outcome measures

Exclusion Criteria:

1. Participants not having access to the internet will be excluded
2. Received or receiving other treatment or interventions (Note: this is an exclusion criterion but not a withdrawal criterion)
3. Children with genetic and other associated auditory or visual impairment and/or seizure disorders

Computer-generated randomisation will be used to randomly assign the participants to one of the intervention groups. Block randomisation will be used to allocate the recruited subjects into one of the interventions. A randomisation list will be generated by the study statistician and envelopes will be prepared. Although the allocation of intervention will be concealed and the study team will not know in advance which subject will receive which intervention, the blinding will not be possible once the intervention is assigned. Assessors of the initial and final MSEL were not told of the intervention status of the patient to avoid bias in outcome assessment. Blinding of the parents is not possible in the context of this study.