Thesis

Michelle Snijder Early identification of and pre-emptive intervention for very young children with a vulnerable social-communicative development Michelle Snijder

Early identification of and pre-emptive intervention for very young children with a vulnerable social-communicative development Michelle Irma Johanna Snijder

Colophon ISBN: 978-94-6473-692-2 Cover design and layout: © evelienjagtman.com Printing: Ipskamp Printing The research presented in this thesis was carried out at Karakter Child and Adolescent Psychiatry and the Radboudumc, Donders Institute for Brain, Cognition and Behaviour and was funded by the Korczak Foundation (Grant No. 2018-01) and Karakter. The studies were further supported by the 1) the EU-AIMS (European Autism Interventions) and AIMS2-TRIALS programs which receive support from Innovative Medicines Initiative Joint Undertaking Grant No. 115300 and 777394, the resources of which are composed of financial contributions from the European Union’s FP7 and Horizon2020 Programs, and from the European Federation of Pharmaceutical Industries and Associations (EFPIA) companies’ inkind contributions, and AUTISM SPEAKS, Autistica and SFARI, and (2) the EU Horizon2020 program CANDY (Grant No. 847818). © M.I.J. Snijder, 2025. All rights reserved. Any unauthorized print or use of this material is prohibited. No part of this thesis may be reproduced, stored or transmitted in any form or by any means, without prior written permission of the author.

Early identification of and pre-emptive intervention for very young children with a vulnerable social-communicative development Proefschrift ter verkrijging van de graad van doctor aan de Radboud Universiteit Nijmegen op gezag van de rector magnificus prof. dr. J.M. Sanders, volgens besluit van het college voor promoties in het openbaar te verdedigen op maandag 17 maart 2025 om 14. 30 uur precies door Michelle Irma Johanna Snijder geboren op 10 mei 1994 te Roermond

Promotor: Prof. dr. J. Buitelaar Copromotor: Dr. I.J. Servatius-Oosterling Manuscriptcommissie: Prof. dr. W.J.J. Assendelft Prof. dr. H.J.A. Van Bakel (Tilburg University) Prof. dr. S.M. Begeer (Vrije Universiteit)

Table of Contents Chapter 1 General introduction 7 Chapter 2 Early detection of young children at risk of autism spectrum disorder at well-baby clinics in the Netherlands: Perspectives of preventive care physicians 27 Published in Autism Chapter 3 Parental Experiences with Early Identification and Initial Care for their Child with Autism: Tailored Improvement Strategies 51 Published in Journal of Autism and Developmental Disorders Chapter 4 Pilot testing and first effects of a new, pre-emptive intervention: Blended E-health for children at eArly Risk. 73 Presented as a poster presentation at INSAR Annual meeting Stockholm, Sweden Chapter 5 Social COmmunication Program supported by E-health (SCOPE) for infants and toddlers at elevated likelihood of autism spectrum disorder: study design of a cluster randomized controlled trial 91 Published in BMC Psychiatry Chapter 6 Short-term and six-month effects of BEAR: A blended, preemptive intervention for infants and toddlers at elevated likelihood for autism. 111 Submitted Chapter 7 Summary and general discussion 151 Appendices Nederlandse samenvatting (summary in Dutch) 167 References 181 Research Data Management 195 Donders Graduate School for Cognitive Neuroscience 201 Portfolio 205 Dankwoord (acknowledgements in Dutch) 211 About the author 219

Chapter 1 General introduction

1 9 General introduction General introduction Over the course of the last few years, there has been a strong increase in scientific interest in the potential of early interventions to alter the developmental trajectories of young children with autism spectrum disorder (ASD) or at elevated likelihood of ASD (French & Kennedy, 2017; Fuller & Kaiser, 2020). Multiple studies show that early intervention of ASD can have a positive impact on the child’s development, but also on family and society functioning, and that early interventions are cost-effective (e.g. Dawson et al., 2012; Fuller & Kaiser, 2020; Horlin et al., 2014; Tinelli et al., 2023). In line with this, the aim of this thesis is to enhance early detection and, most importantly, ensure timely access to suitable interventions for all infants and toddlers experiencing difficulties in their social-communicative development, play, and flexibility, as well as for their families. This introductory chapter starts with a general description of ASD and its early clinical manifestations, followed by the importance and challenges of early detection and intervention. Next, we focus specifically on multiple strategies to improve early detection and intervention of ASD in the Netherlands, including a screening guideline, the founding of an interdisciplinary expertise network and the development of an innovative early intervention program called BEAR (Blended E-health for Children at eArly Risk). This chapter concludes with the specific aims and outline of this thesis. Autism spectrum disorder – it’s definition and prevalence Autism Spectrum Disorder (ASD) – often simply referred to as ‘autism’ in daily practice – is an early onset neurodevelopmental disorder characterized by A) deficits in social communication and interaction and B) restricted repetitive patterns of behaviors and interests (American Psychiatric Association, 2013). In Box 1.1 the current diagnostic criteria for ASD as defined in in the DSM-5 are presented. ASD is a heterogeneous disorder at multiple levels of analysis: phenotypically in terms of the expression of symptoms and clinical severity, indicating a wide variety in how it manifests; etiologically in that different sets of common and rare genes together with environmental factors play a role in causing the disorder; and biologically in that different parts of the brain have been implicated in ASD (Lombardo, Lai & Baron-Cohen, 2019). For example, verbal children with milder symptoms of ASD and (above) average intelligence often show a different presentation of ASD than non-verbal children with more severe symptoms and cognitive impairments (Oosterling et al., 2010). ASD affects about 1 in 100 children (or 100 in 10,000) worldwide (Zeidan et al., 2022). This is a slight increase compared to the previous global prevalence report conducted in 2012 where prevalence was estimated to be 62 in 10,000 children (Elsabbagh et al., 2012). It is important to note that there is a wide variation in the prevalence of ASD in populations worldwide. The

10 Chapter 1 highest prevalence of ASD can be found in the United States, with studies reporting numbers varying from 1:36 (Maenner et al., 2023) to 86 per 10,000 (Anorson et al., 2021). Elsewhere in the world, the estimated prevalence of ASD is lower, with the disorder affecting 59 per 10,000 people in Europe and 47 per 10,000 in Oceania (Anorson et al., 2021). ASD is more commonly diagnosed in boys than in girls. The male-to-female ratio is often assumed to be 4:1, although more recent research considered the ratio closer to 3:1 (Loomes et al., 2017). It is presumed that that girls may present less overt symptoms, may be more likely to mask their social deficits and/or that a gender-based interpretation bias of social difficulties lead to a higher risk for girls being misdiagnosed, diagnosed at a later age or being overlooked (Lai et al., 2015; Hodges, Fealko & Soares, 2020). Individuals who are diagnosed with ASD often cope with lifelong negative consequences on individual, family, and society level which emphasizes the need for global, suitable, and easy access interventions. The neurodiversity perspective The neurodiversity concept has garnered more attention in recent years. It emphasizes the wide natural variation of the human brain, and as has been stated by Baron-Cohen (2017) that “there is no single way for a brain to be normal, as there are many ways for the brain to be wired up and reach adulthood.” Within the concept of neurodiversity, autism is considered as a variation of the human brain with its owns strengths in thinking and behaving (Silberman, 2015). Following this discussion, Baron-Cohen (2017) offers a different approach towards psychiatric conditions such as autism by introducing the four D’s: difference, disability, disease, and disorder. “Difference” meaning that people with autism just differ from the norm set by neurotypical people, whereas “disease” refers to a medical condition with known underlying causal mechanism and pathophysiology. When a parent expresses concerns regarding their child’s development with the child facing challenges in one or more areas of development, autism can be conceived as a “disability”. When there are significant impairments in functioning and there is need for treatment, we speak of autism as a “disorder”. Please note, in this thesis, we focus on the latter two ‘D’s i.e. disability and disorder. Precursors and early clinical manifestations of autism in infants and toddlers As there is no single diagnostic biomarker for ASD, the clinical diagnosis is based on standardized assessments of both observable behaviors and information gathered from developmental history. As defined by the DSM-5 criteria (see Box 1.1.) symptoms of ASD must be present during the early developmental period (American Psychiatric Association, 2013). Research provides solid evidence that the clinical manifestations of ASD can emerge during the first two years of life (Cleary et al., 2023; Zwaigenbaum et al., 2019). In addition to the early clinical manifestations outlined in the DSM-5, extensive research has been conducted on early behavioral precursors through both retrospective and prospective studies.

1 11 General introduction Retrospective studies primarily depend on parents’ recollections of their child’s behavior during infancy, while prospective studies in the field of ASD research involve tracking the development of baby siblings deemed at a high likelihood for ASD due to having an older sibling diagnosed with the disorder. Prospective studies offer the field pivotal insights in how ASD unfolds from birth to later in life by identifying early markers and precursors (Lord et al., 2022; Jones et al., 2014). A summary of both early precursors and clinical indicators of ASD in the first years of life is presented below (see Box 1.2. for clinical case illustrations) 1. Social communication and interaction After the first birthday deficits in social communication and interaction tend to become more pronounced compared to the first twelve months. Social attention and social communication behaviors that can indicate development of ASD include reduced responsiveness to one’s name being called, less frequent use of communicative gestures (for example: waving, pointing), limited social smiling, less variety of facial expressions and limited shared enjoyment (Jones et al., 2014; Miller et al., 2017; Volkmar et al., 2005; Zwaigenbaum et al., 2015). One of the first and most frequent initial parental concerns are delays in the child’s speech and communicative development (Estes et al., 2015; Zwaigenbaum et al., 2005). Additionally, toddlers who were subsequently diagnosed showed a reduced attention towards socially meaningful stimuli (Dawson et al., 2004). Joint attention refers to the development of specific skills that enables one to share attention with others through pointing, showing, and coordinating looks between objects and people (triadic joint attention). Various research demonstrates that toddlers with ASD show fewer joint attention behaviors (Franchini et al., 2019; Sullivan et al., 2007). © P. Brouwers Note. These pictures illustrate a reduced interest in socially meaningful stimuli and limited social smiling

12 Chapter 1 2. Restrictive and repetitive behavior Further potential precursors of early onset ASD include more pronounced repetitive behavior with objects and atypical body movement (Barber et al., 2012; Bryson et al., 2007; Lim et al., 2021; Matson et al., 2009; Miller et al., 2021; Patterson et al., 2022). However, it is important to note that restricted and repetitive behaviors are also part of typical development, especially in the first year of life (Arnott et al., 2010; Leekam et al., 2007). Later, stereotypical, restrictive and repetitive behaviors become more prominent. Toddlers with ASD may display hand and finger mannerisms, show repetitive interests, or play behavior, engage in self-injury behaviors (for example head banging), self-stimulatory use of objects, insistence on sameness, and/ or unusual sensory behaviors (for example hyper- or hyposensitivity to sounds, textures or other sensory stimuli). Non-specific markers, such as early motor impairments in the 0-24 month age range, encompassing differences in fine, gross, and generalized motor functions, have been identified in infants later diagnosed with autism (Lim et al., 2021). © P. Brouwers Note. These pictures illustrate a specific interest in toys and a fascination with spinning objects Importance of early detection and timely intervention of autism Over the last years there has been an increased scientific and clinical interest in the potential of early interventions changing developmental trajectories in young children with or at elevated likelihood of autism spectrum disorder (ASD) (French & Kennedy, 2018). Multiple studies show that intervening at an early age can have a positive impact on child development, on family and society functioning and cost-effectiveness (e.g. Dawson et al., 2012; Fuller & Kaiser, 2020; Horlin et al., 2014; Tinelli et al., 2023), although results of very early interventions vary amongst studies (Hamptom & Rodriguez, 2022; McGlade et

1 13 General introduction al., 2023). It is hypotized that, due to the dynamic and plastic nature of the brain during preschool years, (very) early interventions have the possibility to alter the course of brain and behavioral development in children with ASD (Webb et al., 2014). Additionally, some studies report positive changes in both parent and child behaviors, parent-child interaction and improving parental responsiveness after early intervention (Green et al., 2017; Jones et al., 2017; Kasari et al., 2014; Watson et al., 2017). These findings suggest that starting early intervention in the earliest years of life can have a positive impact on both child, parent and dyadic interactions. 1. From early concerns to early detection Early detection is the first step in receiving timely intervention, not only for children with ASD but also for their parents. Early detection starts with the recognition of initial concerns related to a child’s development, indicating the possibility of developmental delay or disorder in a young child (Buitelaar et al., 2009). After recognizing concerns, the next step would be screening: asking clarifying questions to parents, possibly using screening tools, to assess the risks in a more quantitative manner (Buitelaar et al., 2009) and determine if further steps are needed. After screening, a next step can be referral to a specialized healthcare center for diagnostic assessments. Research indicates that even at a very early age it is possible to provide an ASD diagnosis that is relatively stable over time, in many cases (Cleary et al., 2023; Guthrie et al., 2013; Lord et al., 2022; Zwaigenbaum et al., 2018). Currently, the global average age at which a child receives an ASD diagnosis lies around 5 years (van ‘t Hof et al., 2021). Typically, it is the (one of the) parents who are the first to express concerns about the child’s development. Such concerns often arise during the first 2 years of life (Landa, 2008; Zwaigenbaum et al., 2015) or, in some cases, even before the first birthday. In other cases, it is a professional, for example a preventive care physician or nurse, who expresses initial concerns regarding the child’s development. Unfortunately, the entire diagnostic procedure from concerns to clinical assessment, diagnosis and intervention can be lengthy and time consuming with delay varying between 1.5 and 3.5 years (Crais et., 2020; Crane et al., 2016; Zuckerman et al., 2015). Such delay is undesirable as, a) parents experience high levels of stress during the process of obtaining an ASD diagnosis for their child, and b) delay in early detection and diagnostics results in later access to early intervention services. 2. From early detection to early intervention After diagnostic assessment early interventions can start. In recent years, there has been an uptake in interest in pre-emptive interventions: intervening at a very early age and even before diagnosis. This field has gained traction over the years with mixed but promising results. Previous research shows positive effects on child outcomes such as improvement in social communicative development and language development. (Anderson et al., 2014; Lord et al., 2018; Estes et al., 2015; Fuller & Kaiser, 2020; Pickles et al., 2016). On the other hand, two recent systematic reviews and

14 Chapter 1 meta-analysis found limited effects of very early pre-emptive intervention on child outcomes (such as autism symptoms, cognitive or expressive language outcomes) (Hamptom & Rodriguez, 2022; McGlade et al., 2023). Although these recent reviews found limited effects of pre-emptive interventions on child developmental outcomes, there is one specific intervention to be found more promising: iBASIS-Video Interaction to Promote Positive Parenting (iBasisVIPP). This intervention used video feedback techniques to enhance parental awareness of the infant’s individual social communication skills and needs and offered tailored guidance for parental responses to promote social engagement and interaction (Green, Wan, Guiraud et al; BASIS Team, 2013). Results of an intervention such as iBASIS-VIPP are very promising in that they demonstrate that pre-emptive interventions have the capacity to a) reduce ASD symptom severity and lower the odds of an ASD diagnosis in early childhood; b) elicit a (positive) change in sensitive parental responses; c) improve parent-child interaction; and d) improve the child’s development (Charman,2019; Green & Garg,2018; Kasari,2019; McGlade et al., 2023; Whitehouse et al., 2021). However, further research on the effects of pre-emptive interventions is still needed, not only on documenting the short-term but the long-term effects as well. Early interventions of children with ASD or at increased likelihood of ASD can vary immensely across the world, even within countries and regions (Elsabbagh & Johnson, 2016). This is also the case for the Netherlands, where up until now there is a very limited availability of pre-emptive interventions for infants and toddlers at high likelihood of ASD, as well as their families. Although there is greater global awareness regarding the significance of early detection and intervention, in practice children and their parents receive access to suitable interventions at a later stage than preferred. What makes early detection and access to early intervention so difficult? Both previous research and concerning signals from the clinical field tell us that early detection and access to early intervention for infants and toddlers at elevated likelihood for ASD can be a complex and challenging process. But what makes this process so difficult? As ASD is a heterogeneous disorder with a wide variety in expression of symptoms and clinical severity, there is elevated risk that a subgroup of infants and toddlers receive later care than preferred. For example, children with more subtle and milder ASD symptoms, (above) average IQ and language skills, girls and/or children from ethnic minorities are at a risk of late identification (Mandell et al., 2009; Rutherford et al., 2016). Next, from parents’ perspectives, it is sometimes difficult to acknowledge and accept that their child might develop differently or even has a delayed development. Therefore, parents might be reluctant to be referred to specialized mental healthcare when their child is still very young (Boshoff et al., 2018). Furthermore, it can be difficult for parents to discuss

1 15 General introduction concerns regarding their child’s development due to the apprehension of being dismissed by a professional (Johnson et al., 2020; Locke et al., 2020). From parents’ point of view, the aforementioned might be reasons to not pursue follow-up steps (for example diagnostic assessment) which contributes to the delay of early detection and intervention. From a professional and organizational perspective, previous studies show that limited knowledge of ASD in infants and toddlers, limited use of screening instruments and limited capacity of suitable intervention options are main components in late detection (Pinto-Martin et al., 2005). Within young children, developmental concerns can still be diverse and seen as falling within the broad variation of “average” development, making it difficult to determine whether development is “delayed” or “deviant.” For this reason, preventive care professionals often opt for a “wait and see” approach. This approach is sometimes fitting, but often not. As is elsewhere in the world, the process of early detection and access to early intervention in the Netherlands can be complex and lengthy. In 2018, when the current SCOPE project (as presented in this thesis) started, there was limited knowledge about how the aforementioned obstacles related to the Dutch context. Therefore, one of the objectives of this thesis was, through qualitative research, to identify and gain a deeper understanding on how these barriers relate to the Dutch healthcare system, with the ultimate aim of proposing strategies for improvement (see Chapter 2 and Chapter 3). Previous research of early detection and screening in the Netherlands Over the course of the past few decades, significant research in the Netherlands has been conducted with the aim to improve early detection and intervention for infants and toddlers with (possible) ASD. The SCOPE (Social Communication Program supported by E-health) project in Nijmegen builds upon the knowledge, expertise, and findings of previous research, specifically the SOSO study (initiated in the late 1990’s, Utrecht) and the DIANE project (initiated in 2003, Nijmegen). The SOSO project and the DIANE-project brought the clinical field pivotal insights into how early detection and screening of infants and toddlers at elevated likelihood for ASD could be improved. Both projects led to the development and implementation of important innovative strategies, such as the development of a Dutch screening guideline and the founding of the Dutch interdisciplinary expertise network called ‘Stichting Autisme Jonge Kind’. First, we offer a summary of both research projects and their impact below. Finally, we discuss how the current SCOPE project as described in this thesis relates to the previous work. 1. SOSO Study The SOSO study brought the scientific and clinical field a screening instrument to detect ASD at an early age (14 months), under the name: ESAT (Early Screening of Autistic Traits Questionnaire, Dietz et al. 2006). Nowadays, the ESAT is known as the CoSoS

16 Chapter 1 (Communication and Social development Signs). It is a questionnaire consisting of 14-items to be answered dichotomously as ‘yes’ or ‘no’. A score equal or higher than 3 indicates a positive screen, meaning the child is at an elevated likelihood for ASD. The CoSoS was evaluated in both a population sample and a high-likelihood sample. When compared to other screening instruments such as the M-CHAT, the CoSoS identified fewer children with a screen-positive result. However, the CoSoS (rather than the M-CHAT) is associated with more clinical referrals as most children who received a false positive on the CoSoS were diagnosed with other developmental disorders (81%) (Beuker et al., 2014; Dietz et al., 2007). The CoSoS seemed most effective when applied during two-stage screening: specifically screening for ASD in children who showed one or more “red flags” (i.e. early signs of ASD) during routine developmental surveillance (Dietz et al., 2006). These “red flags” are presented in Box 1.3. 2. DIANE project After the SOSO study, a successful program known as the DIANE project was implemented to improve early detection of ASD. In this project, physicians and nurses working at wellbaby clinics and members of special infant-toddler development teams were trained to a) recognize early signs of autism, and b) use a specially designed referral protocol consisting of a two phase screening approach for ASD, which included administering of the CoSoS questionnaire after concerns were raised during routine developmental screening. The strategies and results of the DIANE project were successful in lowering the age of diagnosis, generally for children with severe ASD and intellectual disabilities (Oosterling, 2010). However, to date there remains a significant limitation in that multiple groups of children are still being overlooked. These groups include children with milder ASD and/or above (average) intelligence, and girls and children from ethnic minorities. The current SCOPE study builds on these findings, with the aim of offering timely intervention for all children who could benefit from it (see Chapter 5). 3. Mountain of success After the initial successful effects of the DIANE project, Pijl et al. (2018) conducted a 8 year follow-up study to evaluate the sustainability of the integrated early detection program as introduced in the DIANE project. The findings of the follow-up study indicate a “mountain of success”. Figure 1.1 shows that, during the DIANE project, there was an increase in children with ASD being diagnosed before age 3. After the project had ended the number of children diagnosed decreased again, showing no difference before versus after the program. Thus, although the objectives of the early detection program were fulfilled and led to earlier referral of children with, the effects disappeared over the years when active investment had faded out (Pijl et al., 2018).

1 17 General introduction The end of investment terminated the collaborative relationship between preventive and specialist care. This may have led to a potential decrease in awareness about early detection among preventive care professionals. Additionally, the absence of postprogram training resulted in a potential loss of expertise when previously trained staff members departed. Considering the overall pressure on funding of mental healthcare, it is of great importance that effective strategies with minimal financial demands are explored to ensure a lasting impact (Pijl et al., 2018). Furthermore, the results of the sustainability research conducted by Pijl et al. a) underlines the need for continued and active investment in early detection and b) emphasizes the importance of ongoing study and the development of improvement strategies for early detection and early intervention in infants and toddlers at elevated likelihood for ASD. The SCOPE project addresses this recommendation by examining a new strategy called the BEAR intervention, which will be briefly discussed at the end of this chapter and explored in greater detail in Chapters 4, 5 and 6 of this thesis. 0 5 10 15 20 25 30 35 2003 2004-2006 2009-2011 Percentage of children diagnosed with Autism Spectrum Disorder (ASD) under 36 months. Figure 1.1. Mountain of succes of the DIANE early detection program, as evaluated by Pijl et al. (2018). Graph first presented by de Korte (2014). Dutch guideline: Screening for Autism Spectrum Disorders at well-baby clinics In the Netherlands, a routine and systematic observation of infants and toddlers takes place at a well-baby office where preventive healthcare professionals such as physicians and nurse regularly engage with young children and their parents for vaccinations, medical care, and developmental screening. This provides preventive care physicians (PCPs) with a crucial role in surveillance, facilitating early detection and guiding children towards early interventions when necessary. Amongst others, the results of both the SOSO study and the

18 Chapter 1 DIANE project led to the development of the Dutch national guideline “Autism Spectrum Disorders” by van Berckelaer-Onnes et al. in 2015 (Nederlands Centrum Jeugdgezondheid, 2015). A summary of the guideline is presented in Figure 1.2. For a schematic representation of the guideline, please see Chapter 2. The Dutch guideline “Autism Spectrum Disorders” offers action-oriented advice to PCPs regarding early detection and screening of ASD. It advocates the use of a general health surveillance tool known as the Van Wiechen Scheme (Laurent de Angulo et al., 2008) to monitor the development of all children from birth to 4 years. The surveillance tool holds the eight “red flags” of Dietz et al. (2007, see Box 1.3.). Should one or more behavioral red flags be identified during general surveillance with the Van Wiechen Scheme, the Dutch guideline encourages PCPs to employ the CoSoS screening questionnaire. In the Netherlands, there are high standards of primary and preventive healthcare and an extensive scientific and clinical knowledge about the importance of early detection and intervention of ASD. However, signals from the clinical field gave the impression that the Dutch ASD guideline was not applied in PCPs daily work, a few years after its introduction (2018/2019). This perhaps contributed to a later than preferred detection and receiving later than desired healthcare of children at elevated likelihood of ASD. However, research on this “gut feeling” was limited. Therefore, one of the goals of this thesis was to explore possible barriers experienced amongst PCPs regarding early detection and to offer possible solutions to alleviate these identified barriers (see Chapter 2). The professional and/or parent is concerned. There is a suspicion of ASD. The professional completes a screening questionnaire based on their own observations, information from other sources (such as daycare/school), and input from the parent(s)/caregiver(s) If the score on the screening questionnaire indicates the need, and the parents have a request for support, the child can be referred for diagnostic assessment and treatment. “Gut feeling” Screening questionnaire Referral for diagnostic assessment Figure 1.2. Summary of the Dutch guideline “Autism Spectrum Disorders” (van Berckelaer-Onnes et al., 2015).

1 19 General introduction Dutch Expertise Network “Autisme Jonge Kind” (AJK) Another important strategy designed to improve early detection and timely intervention of ASD in the Netherlands was the founding of a national interdisciplinary expertise network. This network, called the Dutch Expertise Network Autism in Young Children (in Dutch: Stichting Autisme Jonge Kind, AJK), was initiated by dr. Iris Servatius-Oosterling, dr. Claudine Dietz and prof. dr. Jan Buitelaar in 2013. The network brings together the expertise of specialized clinicians, scientific knowledge of researchers and parental experiences in the early detection, diagnosis, and treatment of ASD in children aged 0-6 years. AJK aims to improve early detection and, more importantly, timely access to intervention. To this end three innovations were introduced. First, an informative online platform for parents and professionals was launched in 2018 (www.autismejongekind.nl). This website, available in Dutch, English, Turkish and Arabic, offers easy and accessible information about the early indicators of ASD to parents and professionals. At the same time, the website spreads awareness about the importance of early detection and intervention. Second, a free e-learning platform for preventive care professionals regarding the recognition of the early signs of ASD and use of the CoSoS was developed. Lastly, the third component consisted of a relatively short and acceptable homebased early intervention offered to parents of infants and toddlers at elevated likelihood for ASD. This early intervention is called ‘BEAR’, short for ‘Blended E-health for children at eArly Risk’ and was developed by Dietz & Oosterling (2018) with vital input from other ASD specialists. The BEAR intervention will be further described below. The SCOPE project focusses on studying the outcomes of the BEAR intervention applied to families with very young children at elevated likelihood for ASD (see Chapter 6). BEAR intervention The BEAR intervention is a pre-emptive intervention offered by a first line healthcare professional, ideally supervised by a specialized mental healthcare professional. The primary objective of BEAR is to assist parents in understanding their child’s behavior, fostering sensitivity to the child’s needs, and, consequently, motivating the child to engage socially. The intervention is built on two well-studied principles, namely 1) high synchrony between parent and child is assumed to be related to decreasing the severity of autism symptoms, and 2) improved joint attention and joint engagement skills are associated with improved communicative abilities (Kaale et al., 2012; Kasari et al., 2012; Murza et al., 2016). Like other early interventions (for example: Pivotal Response Treatment, Floorplay, iBasisVIPP and JASPER-training), BEAR intervention techniques are rooted in evidence-based naturalistic developmental behavioral intervention principles (NDBI’s) (Bruinsma et al., 2020). The innovative value of BEAR lies in the combination of the a) very early start of intervention, b) easy accessibility with no waiting lists and the intervention conducted at home, and

20 Chapter 1 c) strong collaboration between first line (executer) and specialized mental healthcare (supervisor). Without the need of an ASD diagnosis, BEAR is presumably more acceptable and better-received by parents who have relatively minor concerns, or who are somewhat ambivalent about getting help or those who dread diagnostic assessment. Additionally, it is seen as a suitable healthcare option for children who exhibit unclear clinical signals/ “red flags”. Chapter 5 and 6 will provide a more detailed description of the BEAR intervention, including its development and results. Aims and outline of the thesis This thesis aims to investigate ways in which early detection can be improved, and, most importantly, to facilitate timely access to early interventions for infants and toddlers (along with their families) at elevated likelihood for ASD. The three specific aims of this thesis are: 1. To identify barriers in the early detection and screening of infants and toddlers at high likelihood for ASD and to offer improvement strategies for the clinical practice; 2. To develop and pilot test a new, innovative early intervention named BEAR (Blended E-health for children at eArly Risk); 3. To study short-term and six-month effects of the BEAR intervention. The data described within this thesis have been collected from three study samples. Samples are collected in the “real world”, consisting of preventive care physicians, parents, and young children. See Box 1.4. for more details on the study samples. To address the concerns regarding late detection and intervention of infants and toddlers at an elevated likelihood for ASD, we explored the experienced barriers in early detection and access to (early) interventions and gathered perspectives on improvement strategies from two key stakeholders: preventive care physicians (Chapter 2) and parents (Chapter 3). We summarize these two perspectives in the discussion of Chapter 3 and present a chart that may be helpful in thinking of solutions for improving early detection and intervention of ASD. As one of these solutions, we propose an innovative intervention named BEAR. In Chapter 4, the BEAR intervention and SCOPE study protocol are tested in a pilot study to assess the feasibility of the research design and address any potential research challenges. After pilot testing, adjustments to the protocol were made. The final study protocol is presented in Chapter 5. In Chapter 6 the short-term and six month results of the BEAR intervention are presented. Finally, the general discussion in Chapter 7 presents a summary and analysis of the major findings, discusses limitations of the conducted research, explores directions for future research and concludes with implications for clinical practice.

1 21 General introduction Box 1.1. DSM-5 diagnostic criteria for ASD A. Persistent deficits in social communication and social interaction across multiple contexts, as manifested by the following, currently or by history (examples are illustrative, not exhaustive, see text): 1. Deficits in social-emotional reciprocity, ranging, for example, from abnormal social approach and failure of normal back-and-forth conversation; to reduced sharing of interests, emotions or affect; to failure to initiate or respond to social interactions. 2. Deficits in nonverbal communicative behaviors used for social interaction, ranging, for example, from poorly integrated verbal and non-verbal communication; to abnormalities in eye contact and body language or deficits in understanding and use of gestures; to a total lack of facial expressions and nonverbal communication. 3. Deficits in developing, maintaining, and understanding relationships, ranging, for example, from difficulties adjusting behavior to suit various social contexts; to difficulties in sharing imaginative play or in making friends; to absence of interest in peers. B. Restricted, repetitive patterns of behavior, interests or activities, as manifested by at least two of the following, currently or by history (examples are illustrative, not exhaustive; see text): 1. Stereotyped or repetitive motor movements, use of objects, or speech (e.g. simple motor stereotypies, lining up toys or flipping objects, echolalia, idiosyncratic phrases). 2. Insistence on sameness, inflexible adherence to routines, or ritualized patterns or verbal nonverbal behavior (e.g. extreme distress at small changes, difficulties with transitions, rigid thinking patterns, greeting rituals, need to take same route or eat food every day). 3. Highly restricted, fixated interests that are abnormal in intensity or focus (e.g. strong attachment to or preoccupation with unusual objects, excessively circumscribed or perseverative interest). 4. Hyper-or hypo reactivity to sensory input or unusual interests in sensory aspects of the environment (e.g. apparent indifference to pain/temperature, adverse response to specific sounds or textures, excessive smelling or touching of objects, visual fascination with lights or movement). Specify current severity: Severity is based on social communication impairments and restricted, repetitive patterns of behavior. C. Symptoms must be present in the early developmental period (but may not become fully manifest until social demands exceed limited capacities, or may be masked by learned strategies in later life). D. Symptoms cause clinically significant impairment in social, occupational, or other important areas of current functioning E.. These disturbances are not better explained by intellectual disability (intellectual developmental disorder or global developmental delay. Intellectual disability and autism spectrum disorder frequently co-occur; to make comorbid diagnosis of autism spectrum disorder and intellectual disability, social communication should be below that expected for general developmental level.

22 Chapter 1 Box 1.2. Clinical reports William is 13 months old. His parents have concerns regarding his development. They observe that William displays limited interest in social interaction with them, does not smile at them or does not respond when spoken to. Furthermore, William is a notably quiet infant, making no babbling sounds. Parents note that these behaviors differ from their experience with their first child, prompting them to express their concerns to the preventive care physician at the well-baby clinic. The preventive care physician considers William to be still very young and opts to monitor his development. A follow-up appointment is scheduled in three months. Amir, who is 24 months old, accompanies his parents to the well-baby clinic for a routine check-up. During the visit, the preventive care nurse observes that Amir is not speaking and avoids making eye contact. Additionally, the nurse notices certain hand mannerisms that she perceives to be atypical. She communicates her concerns to Amir’s parents. Amir’s parents react surprised. While they share the nurse’s concern about Amir’s language development, they do not fully recognize and share the other observations the nurse mentions. Amir’s parents agree with a referral to a speech therapist. Emma is a 30 month old girl who visits kindergarten. Over time, the staff starts noticing some behaviors that stand out to them. For instance, when her parents drop her off, Emma walks into the kindergarten without waving goodbye. Once inside, she prefers playing alone in a corner with her dolls. If another child attempts to join her play, Emma shows visible distress and is difficult to console. The staff also noticed that Emma does not respond to her name being called and exhibits a fascination with turning lights on and off. Upon discussing these behaviors with her parents, they acknowledge observing similar behaviors at home. Concerned about Emma’s development, her parents conduct independent research on the internet, leading them to suspect that Emma may have autism. They decide to discuss these concerns during the upcoming well-baby clinic visit.

1 23 General introduction Box 1.3. Red flags or signs of ASD (Dietz, 2007) · No babbling by 12 months · No interest in other people by 12 months · No smiling at others by 12 months · No reaction when spoken to by 12 months · No gesturing by 12 months (pointing, waving bye bye) · No functional use of single words by 18 months · No 2-word spontaneous phrases (not echoic) by 24 months · Any loss of any language or social skills at any age

24 Chapter 1 Box 1.4. Study samples PRE-SCOPE project part 1| Chapter 2 The sample of the first part of the PRE-SCOPE project, as presented in Chapter 2, consisted of 12 preventive care physicians who worked at well-baby clinics spread across the 12 different regions of the Netherlands (one per province). Most participants (all but one) were woman. A broad scope of work experienced was covered, with work experience varying between 0.5-30 years (M=9.04). Qualitative analyses based on Grounded Theory led to the conceptualization of themes and relevant categories regarding barriers and improvement strategies. PRE-SCOPE project part 2| Chapter 3 The second part of the PRE-SCOPE project, as presented in Chapter 3, holds a mixed-method study where the total sample can be divided into two sub-samples. First, an online survey was completed by 45 parents (39 mothers and 6 fathers) recruited via partners from the Autisme Jonge Kind expertise network. All participants reported in name of both parents, with information about one father missing in total (N=89). Participants reported on their experiences with the early identification process and diagnostic assessment trajectory of in total 45 children. Most of these children was male (82%). All children were diagnosed with ASD, with 11.1% of the children having a comorbid disorder. Second, an additional focus group was held, consisting of 10 parents (nine mothers, one father). Their age varied between 37 and 52 years old (M=43.5). Parents participating in the focus group reported on 12 children: 10 boys and 2 girls. Most parents shared their experiences with one child diagnosed with ASD. Two parents shared experiences with two children, one of which concerned twins. Children received an ASD diagnosis between the ages of three and six years. SCOPE project | Chapter 6 The sample of the SCOPE projected consisted in total of 54 children, divided between two groups. One group called the experimental group, received the BEAR intervention (N=40) whereas the other group, the control group (N=14) received care as usual (CAU). Our sample consisted of infants and toddlers (aged 12-30 months) who were referred to our study by preventive care professionals working at well-baby clinics based on clinical concerns and a positive screen on the CoSoS questionnaire (≥3). Participants were divided into one of the two groups (BEAR or CAU) based on cluster randomization. Assessments were performed at three time points (baseline, endpoint and 6-month follow-up) using the Joint Engagement Rating Inventory (JERI) and Brief Observation of Social Communication Change (BOSCC) as primary outcome measures, both rated on the same parent-child interaction videos. Secondary outcome measures include ADOS-2, global levels of adaptive functioning (Vineland Screener), parental skills (video-observations) and parental well-being (OBVL, WEMWBS).

Chapter 2 Early detection of young children at risk of autism spectrum disorder at well-baby clinics in the Netherlands: perspectives of preventive care physicians Michelle I.J. Snijder12, Shireen P.T. Kaijadoe1, Maarten van ’t Hof34, Wietske A. Ester345, Jan K. Buitelaar12 & Iris J. Oosterling1 Autism, 2021 Oct;25(7): 2012-2024. https://doi.org/10.1177/13623613211009345 1 Karakter Child and Adolescent Psychiatry University Centre, Nijmegen, The Netherlands 2 Department of Cognitive Neuroscience, Donders Institute for Brain, Cognition and Behaviour, Radboudumc, Nijmegen, The Netherlands 3 Sarr Expert Centre for Autism, Lucertis Child and Adolescent Psychiatry, Rotterdam, The Netherlands 4 Parnassia Psychiatric Institute, The Hague, The Netherlands 5 Department of Child and Adolescent Psychiatry, Curium-LUMC, Leiden University, Leiden, The Netherlands

28 Chapter 2 Abstract To improve early detection of autism spectrum disorder (ASD) in preventive care, a Dutch guideline was developed five years ago. The guideline provides preventive care physicians at well-baby clinics action-oriented advice and describes a step-by-step approach for children identified at an increased risk for ASD during general health care surveillance. This qualitative study evaluated guideline adherence and studied barriers regarding early detection of ASD at well-baby clinics. Semi-structured interviews were undertaken with 12 preventive care physicians (one representative per province). Interviews were analyzed using grounded theory. It was found that the vast majority of participants did not followup general surveillance with an ASD-specific screener as prescribed by the guideline. Six barriers, to be divided in personal, guideline-related and external factors, were found regarding early detection of ASD and adherence to the guideline. The results of this study highlight the importance of an integrative approach, raising awareness of the benefits regarding early detection of ASD in preventive care, the need of continuous investment in easy and accessible training and active screening, and a closer collaboration between preventive care organizations and ASD experts. Keywords: autism spectrum disorder, early detection, preventive care, guideline adherence, screening

29 Perspectives of preventive care physicians on early detection of autism spectrum disorder 2 Autism spectrum disorders (ASD) affect about 1% of the population and are associated with lifelong impairments in functioning, substantial family burden, increased mortality and high medical costs (Buescher et al., 2014; Elsabbagh et al., 2012). Over the last years, the importance of early detection and access to early intervention of ASD is widely recognized (French & Kennedy, 2018). Due to the dynamic and plastic nature of the brain, early interventions can alter the development in children with ASD. Early detection and intervention can increase social communication outcomes and can reduce comorbid problems, negative impact on families and lifetime costs to the individual, family and society (Dawson et al., 2012; Fuller and Kaiser, 2019; Horlin et al., 2014). Despite these benefits, significant delays exist in access to screening, diagnosis and healthcare services for young children with ASD (BishopFitzpatrick & Kind, 2017). Research has established that developmental abnormalities associated with ASD become evident in some children as early as twelve months (Jones et al., 2014) and that a reliable ASD diagnosis can be made before the age of two (Guthrie et al., 2013). Despite increasing awareness of the importance of early detection, in the Netherlands, the average age children with ASD are diagnosed is around five years (Begeer et al., 2017) and thus receiving access to appropriate intervention later than preferred. In the Netherlands, about 15 years ago, a successful program known as the ‘DIANE project’ was implemented in order to improve early detection of ASD. In this program, physicians and nurses at well-baby clinics and members of special infant-toddler development teams were trained to a) recognize early signs of autism and b) use a specially designed referral protocol existing of a two-phase screening approach for ASD which included a screening questionnaire, the CoSoS (Communication and Social development Signs, previously known as ESAT; Dietz et al., 2006) after concerns were raised during routine developmental screening. The strategies of the ‘DIANE project’ mainly led to an improvement of early detection of children with severe ASD (Oosterling et al., 2010). However, eight years after the program ended, effects in early detection were not maintained in clinical practice (Pijl et al., 2017) and highlights the importance to continue studying improvement strategies for early detection of infants and toddlers at risk for ASD. In order to improve early detection of ASD, a Dutch guideline was developed for use within preventive care (van Berckelaer-Onnes et al., 2015). In the Netherlands, all infants and toddlers are systematically observed at well-baby clinics during thirteen visits between 0-4 years (program varies slightly per well-baby clinic). Well-baby clinics are a form of preventive healthcare where physicians and nurses have regular contact with young children and their parents, mostly for routine vaccinations, medical and developmental screening. Preventive care physicians (PCPs) have an important surveillance role in the Dutch healthcare system and can therefore play a critical part in early detection, as well as guiding children to early interventions. The Dutch ASD guideline provides PCPs action-oriented advice in early

30 Chapter 2 detection and screening of ASD. It describes that the development of all children should be monitored by using a general tool for health surveillance, called the Van Wiechen Scheme (Laurent de Angelo et al., 2008). The Van Wiechen Scheme provides a template to monitor developmental milestones for children from birth to four years of age. The surveillance tool holds eight signals, considered to be first behavioral red flags of ASD in infants and toddlers (Dietz et al., 2006). When one or more behavioral red flags are identified during general surveillance with the Van Wiechen Scheme, the Dutch guideline urges PCPs to use an ASDspecific screener (CoSoS). Figure 1 provides a schematic representation of the guideline, as developed by van Berckealer-Onnes et al. (2015). In the Netherlands, there is a general rule that national guidelines must be implemented in preventive healthcare organizations within two years after its introduction. To realize this, the organization that oversees guideline implementation carries out a standard protocol (Dunnink et al., 2016). Although it has been five years since the guideline was introduced and implemented by standard methods, until now, the current state of guideline adherence has not been investigated. Developmental worries expressed by parents, school, kindergarten, and/or healthcare professionals Developmental worries during general healthcare surveillance (missed items on the Van Wiechen Scheme) Likelihood of ASD CoSoS/ESAT Specific anamnesis AND / OR Positive screen? Continue to monitor child development. Follow-up assessment if needed Referral for diagnostic assessment and early intervention YES NO Start other healthcare trajectory (i.e. referral to audiologic centre) YES NO Other causes? Figure 1. Representation of the Dutch national guideline on autism spectrum disorders by van Berckelaer-Onnes et al. (2015, translated). Previous studies show that clinical screening guidelines are often not fully applied (Arunyanart et al., 2012; Fischer et al., 2016; Wallis et al., 2020). A systematic review that proposes a model to improve guideline implementation and adherence (Knowledge-

RkJQdWJsaXNoZXIy MjY0ODMw