STUDY PROTOCOL Open Access

A clinical and cost-effectiveness trial of aparent group intervention to managechallenging restricted and repetitivebehaviours in young children with autismspectrum disorder study protocol for arandomised controlled trialVictoria Grahame1 Linda Dixon1 Sue Fletcher-Watson2 Deborah Garland3 Magdalena Glod4 Jane Goodwin4Zoe Grayson5 Saoirse Heron6 Emma Honey1 Rebecca Iversen6 Adetayo S Kasim7 Ashleigh Kernohan8Ehsan Kharatikoopaei9 Ann Le Couteur4 Leila Mackie5 Ayesha Mathias10 Helen Probert11 Deborah Riby11Priyanka Rob4 Leanne Rogan1 Sarah Thompson11 Luke Vale8 Eamonn Walls6 Elspeth Imogen Webb12Christopher Weetman10 Faye Wolstenhulme10 Ruth Wood10 and Jacqui Rodgers4

Abstract

Background Restricted and repetitive behaviours vary greatly across the autism spectrum and although not all areproblematic some can cause distress and interfere with learning and social opportunities We have alongsideparents developed a parent group based intervention for families of young children with autism which aims tooffer support to parents and carers helping them to recognise understand and learn how to respond to theirchildrsquos challenging restricted repetitive behaviours

Methods The study is a clinical and cost-effectiveness multi-site randomised controlled trial of the ManagingRepetitive Behaviours (MRB) parent group intervention versus a psychoeducation parent group Learning AboutAutism (LAA) (n = 250 125 intervention125 psychoeducation ~ 83site) for parents of young children aged 3ndash9years 11 months with a diagnosis of autismAll analyses will be done under intention-to-treat principle The primary outcome at 24 weeks will use generalisedestimating equation (GEE) to compare proportion of children with improved RRB between the MRB group and the LAAgroup The GEE model will account for the clustering of children by parent groups using exchangeable workingcorrelation All secondary outcomes will be analysed in a similar way using appropriate distribution and link function(Continued on next page)

copy The Author(s) 2021 Open Access This article is licensed under a Creative Commons Attribution 40 International Licensewhich permits use sharing adaptation distribution and reproduction in any medium or format as long as you giveappropriate credit to the original author(s) and the source provide a link to the Creative Commons licence and indicate ifchanges were made The images or other third party material in this article are included in the articles Creative Commonslicence unless indicated otherwise in a credit line to the material If material is not included in the articles Creative Commonslicence and your intended use is not permitted by statutory regulation or exceeds the permitted use you will need to obtainpermission directly from the copyright holder To view a copy of this licence visit httpcreativecommonsorglicensesby40The Creative Commons Public Domain Dedication waiver (httpcreativecommonsorgpublicdomainzero10) applies to thedata made available in this article unless otherwise stated in a credit line to the data

Correspondence VictoriaGrahamecntwnhsuk1Cumbria Northumberland Tyne and Wear NHS Foundation Trust ComplexNeurodevelopmental Disorder Service (CNDS) Walkergate Park BenfieldRoad Newcastle upon Tyne NE6 4QD UKFull list of author information is available at the end of the article

Grahame et al Trials (2021) 22240 httpsdoiorg101186s13063-021-05175-y

(Continued from previous page)

The economic evaluation will be conducted from the perspective of both NHS costs and family access to localcommunity services A lsquowithin trialrsquo cost-effectiveness analysis with results reported as the incremental cost per additionalchild achieving at least the target improvement in CGI-I scale at 24 weeks

Discussion This is an efficacy trial to investigate the clinical and cost-effectiveness of a parent group basedintervention designed to help parents understand and manage their childrsquos challenging RRB If found to be effectivethis intervention has the potential to improve the well-being of children and their families reduce parental stressgreatly enhance community participation and potential for learning and improve longer-term outcomes

Trial registration Trial ID ISRCTN15550611 Date registered 07082018 Sponsor and Monitor CumbriaNorthumberland Tyne and Wear NHS Foundation Trust RampD Manager Lyndsey Dixon Address St Nicholas HospitalJubliee Road Gosforth Newcastle upon Tyne NE3 3XT lyndseydixoncntwnhsuk Tel 0191 246 7222

Keywords Autism spectrum disorder Randomised controlled trial Restricted and repetitive behaviours Parent-mediated intervention

IntroductionBackgroundAutism spectrum disorder (hereafter lsquoautismrsquo) is a life-long neurodevelopmental condition affecting 1ndash2 ofthe population with profound impact on diagnosedindividuals families and society [1 2] Restricted and re-petitive behaviours (RRB) are one of two key behaviouraldomains required for a diagnosis of autism [3] Theyinclude repetitive motor mannerisms rigid adherence tospecific routines highly circumscribed interests and ex-treme responses to everyday sensory experiences SomeRRBs are reported by autistic adults to be enjoyablefunctional and helpful They may provide a basis forfriendship and can also build areas of strength support-ing skill development and yielding employment oppor-tunities [4 5] However RRB may also be outward signsof anxiety or distress deserving of attention and carefrom others [6] In some cases RRB in children can havea distressing and negative impact for the child and theirfamily members [7] RRB can interfere with learningopportunities community participation aspects ofhealth such as sleep and nutrition and deplete childparent and sibling wellbeing [8] Examples observed bypractitioners include an elaborate and rigid bedtimeroutine lasting 3 h or more a restricted diet of only 5different foods and a repetitive pattern of pinchingmembers of the publicHere we use the term lsquochallenging RRBrsquo to describe be-

haviours that fit the diagnostic description of a restrictedand repetitive behaviour and that cause significantchallenges for the child and in daily family life A recentmeta-analysis indicates that parents of autistic childrenexperience higher levels of stress than parents ofchildren with other disabilities and that stress is highlycorrelated with the presence of challenging RRB [9] Theimpact is greater if parentscarers are not able to accessappropriate advice and support [6] Challenging RRBmay also incur direct risk of injury for the autistic child

[6] They can take up large amounts of time interferewith the childrsquos ability to engage in everyday living activ-ities (eg personal care mealtimes) and reduce access tolearning [10ndash14] Challenging RRBs are associated withincreased rates of disruptive behaviours and aggressionto others which can further isolate the child and familyand contribute to negative stigma [15 16]Parents report they do not receive specific professional

advice on how to recognise or understand their childrsquoschallenging RRB [17] It is therefore critical that parentsare supported to understand the different formsfunctions and impact of RRB They need to be able todistinguish possible underlying reasons for RRB and toidentify those RRB that have the potential for deleteriousimpact on their child They require evidence-basedeffective and efficient strategies to specifically addresschallenging RRB Such interventions if found to beeffective have the potential to improve the well-being ofautistic children and their families reduce parentalstress greatly enhance community participation and po-tential for learning and improve longer-term outcomesIn the UK the National Service Framework for

Disabled Children and Young People and those withComplex Health Needs highlights the potential impactof care for parents of children with autism and the needfor effective and efficient evidence-based parent traininginterventions [18] The evidence base for the effective-ness of parent-mediated interventions for young childrenwith autism has been reported [19] However to datemost autism-specific early intervention programmesfocus on social communication and rarely considerchallenging repetitive behaviours despite this area beinga priority for parents [20ndash23] Furthermore a systematicreview on effectiveness of treatments for RRB in ASDhas established that strategies that involve understandingthe behaviour were promising but lacked a sufficient evi-dence as the majority of studies used single case designsand focused solely on stereotypy [21] Stereotypy is one

Grahame et al Trials (2021) 22240 Page 2 of 13

specific type of RRB that is also less likely to beperceived as challenging A review of trial databases(USNIH and UKCTG) indicates no current behaviouralintervention trial registered for RRB (2020) There is onerecently completed trial (completed Feb 2019) that aimsto use mobile technology to reduce stereotypy (repetitivevocal and motor behaviours) in children with autismand is thus not relevant to the current trial [24]The Managing Repetitive Behaviours (MRB) parent

group intervention and study protocol were developedjointly in partnership with parents of young childrenwith autism initially through two developmental parentgroups The intervention was designed to help parentsrecognise understand and manage their childrsquos RRB[25] The development work was followed by a pilotfeasibility and acceptability randomised controlled trial(RCT) The results of the feasibility and acceptabilitypilot indicated that the MRB intervention was bothacceptable to parents and feasible to deliver throughroutine clinical services If effective this intervention hasthe potential to extend the range of early interventionsavailable to meet the needs of young children with aut-ism and their families ensuring best use of therapeuticresources and reducing the risk that challenging RRBpersist with significant long-term consequences for thechild and family However before recommending thatthis parent group intervention is included within localcommunity early intervention services a fully poweredclinical- and cost-effectiveness RCT is required to ruleout harmful effects of the intervention and quantify anybenefit This study is a phase 3 superiority trial with aparallel group design in which each family is randomised11 to one of the two parent group interventions

Methods and designStudy designThis study is a UK three-site two-group RCT Afterconsent families will be randomised at each site toreceive either the 8-week MRB parent group interven-tion or the 8-week Learning About Autism (LAA) parentgroup equivalent to best current practice and will act asan attentional control for time and attention Assess-ments are administered on entry (baseline) to the trialat the end of the eight parent group sessions (10 weeks)at 24 weeks primary endpoint and at the 52 weeksfollow-up

Study aimsThe aim of this study is to evaluate the clinical and cost-effectiveness of the Managing Repetitive Behaviours(MRB) parent group intervention for young childrenwith autism compared with a parent autism psychoedu-cation group at 10 24 and 52 weeks follow-up

The clinical effectiveness of this intervention will beassessed by measuring whether children show an overallimprovement in global functioning (primary outcome) at24 weeks (primary endpoint) after the MRB interventioncompared to the children whose parents had attendedthe LAA group There are a number of secondaryoutcomes (measured at the end of the intervention andat 24-week follow-up) these include whether (i) parentsincrease their knowledge and skills in understanding andmanaging their childrsquos challenging RRB (ii) childrenshow a reduction in challenging RRB and change in par-ticipation in everyday activities parent stress and impacton family life (iii) the intervention provides value formoney for the NHS and (iv) these improvements aremaintained at 1 year from baselineWe have carefully selected end of group and follow-up

(24 and 52 weeks) outcome measures to capture bothindependent (blinded researcher) teacher and parent-reported changes These include blinded ratings ofimprovement in childrsquos overall functioning (assessedusing Clinical Global ImpressionsndashImprovement (CGI-I)) and change in challenging RRB Parent questionnaireswill also provide information on child RRB child partici-pation and daily living skills parent stress self-efficacyand impact on family life Parents will be aware of groupstatus but teachers and the primary outcome assessorwill be blind to group allocationWe will evaluate maintenance of effect at the 52-week

follow-up appointment which will also allow us to investi-gate any potential longer-term (downstream) impact onchild and family functioning (eg social participation)

Study settingThe MRB and the LAA parent groups will all take placein community settings in different geographical locationsacross the three UK sites in Tyne and Wear TheLothians and Teesside This is in line with the UK Na-tional CAMHS Review (2008) [26] which indicatedyoung people and families want accessible services inconvenient places

Managing Repetitive Behaviour (MRB) interventionMRB is a manualised intervention designed to help par-ents of young children with ASD to recognise under-stand and learn how to manage their childrsquos challengingRRB It is an 8-week manualised intervention designedto be delivered by community-based professionals withknowledge and experience of working with young chil-dren with ASD and their families who have been trainedto deliver the MRB intervention Each weekly sessionlasts for approximately 2 h (total duration of group inter-vention is ~ 16 h)Each parent will be provided with a manual related

weekly materials and individualised support to identify

Grahame et al Trials (2021) 22240 Page 3 of 13

strategies to address challenging RRBs lsquoAt homersquo activ-ities will be set each week for parents and children tocomplete between sessions The opportunity to identifyand practice new ideas outside of the clinic setting isimportant for investigating how best to generalise theapproach into different settings for each child andfamilyParents have the opportunities for individual support

and group learning In weeks 2 and 7 they meet thegroup leaders individually to select a target challengingRRB to focus on during the group and to practise thenew skills they are learning thus ensuring that strategiesare individually tailored for each child and are develop-mentally appropriate

Learning About Autism (LAA) parent groupLAA is an 8-week manualised parent psychoeducationgroup that will act as an attentional control LAA isdesigned for parents of young children with autism andfocuses on understanding what autism is and providessupport for parents of young children with autism It isdesigned to be run in the community with professionalswho have experience of working with young childrenwith autism and their families and have been trained bya UK-based autism charity the National Autistic Societyto deliver the intervention Each weekly session lasts forapproximately 2 h (total duration of group interventionis ~ 16 h)The groups will offer parents and carers psychoeduca-

tion on understanding autism and what that means fortheir individual child such as helping them understandtheir childrsquos social communication difficulties and behav-iour in different environments It will also provide adviceand guidance on strategies and approaches for dealingwith behaviour problems in young children with ASDThere will be opportunities for mutual support and shar-ing of ideas with other parents Thus increasing parentsrsquoand carersrsquo understanding confidence and responsive-ness to their childrsquos patterns of communication andinteraction This is equivalent to current best practice Itwill not include any specific information about the roleand functions of RRB functional analysis and tailoredstrategies to manage challenging RRB

Recruitment and eligibility criteriaFamilies will be identified via local community NationalHealth Service (NHS) clinical services Clinicians areasked to introduce the study to families fitting the inclu-sion criteria discuss the study with them and if inter-ested give them study packs For some clinical serviceswhere children may not be seen as frequently postalstudy packs will be sent to identified families Interestedparents are asked to complete and return to the researchteam an expression of interest form The research team

will then contact parents they are interested in hearingmore about the study to arrange a face-to-face meetingto discuss the study what participation involves answerany questions they may have and obtain written in-formed consentParents are eligible for study entry if their child meets

the following criteria (1) aged 3ndash9 years 11 months atthe time of consent with a clinical diagnosis of autism orautism spectrum disorder (2) experience challengingRRB Parents themselves will have (1) sufficient spokenand written English to provide written informed consentcomplete assessments and participate in the interven-tion (2) are willing to be randomised and attend allgroup sessions for the allocated arm of the study and(3) agree to maintain their childrsquos current medication re-gime until the primary 24-week end point (clinicianchanges to medication are permitted) Parents are noteligible for study entry if their child meets the followingcriteria (1) no clinical diagnosis of autism or ASD (2)no challenging RRB (3) is currently taking part in an-other parent group based intervention trial or (4) has asibling already taking part in this study Furthermorechildren of parents who have significant mental healthdifficulties will not be eligible to participate There willbe no special criteria for discontinuing or modifying al-located interventions

Training and fidelityAll parent group sessions in both arms of the trial willbe recorded both to facilitate opportunities for supervi-sion of the group leaders and to provide access to arandom sample of recorded sessions across all sites forthe independent evaluation of (i) the fidelity to thedelivery of the intervention and (ii) the therapeuticcompetence of the individual group leadersrsquo delivery ofthis manualised intervention Fidelity will be formallyassessed by independent raters watching 10 ofrandomly selected tapes and utilising the MRB FidelityRating Scale as developed during the pilot feasibility andacceptability study with adaptations for the content ofthe LAA parent groupsAt the start of the research the group leaders and site

clinical leads will attend a 1 day MRB training courselearning how to use the training manual and the mate-rials for each 2 h weekly session and planning for home-work tasks Each group leader will learn about groupprocesses and gain skills to deliver the manualised inter-vention in a participatory style in combination withstrategies to personalise the programme for each parentThe senior trainer (LD) alongside the chief investigator(VG) will deliver the introductory training course LDwill also visit each site as each new group is started Fur-thermore to ensure the intervention is being deliveredto the families as intended with high fidelity to the

Grahame et al Trials (2021) 22240 Page 4 of 13

manual the senior trainer will alongside the local siteclinical lead will undertake weekly supervision of groupleaders at each site Group Leaders delivering the LAAparent group will attend a 1-day training course specific-ally designed for this study to ensure they are trained inthe skills needed to deliver the programme This coursehas been designed by the National Autistic Society andwill be delivered by DGParent attendance at group sessions will be recorded

and monitored Parents who miss a parent group sessionwill be contacted by a group leader to provide a briefcatch up of session and handouts for that session will beposted out Therapy compliance and receipt of otherinterventions outside of the protocol will be monitored

Avoidance of contaminationThere will be separate clinical and research leads at eachsite and separate training and supervision structures Re-searchers will be located separately from staff involvedin delivery of the MRB and LAA intervention Researchinterviews and assessments will be conducted mostly athome visits with the Autism Diagnostic ObservationSchedule-2 (ADOS-2) being undertaken at Universitysites All parents will be reminded prior to every contactwith research staff not to talk about group allocationFurthermore we will also consider issues around poten-tial contamination of the active attentional control LAAgroup by ensuring the LAA families are not on the caseload of the MRB intervention group leaders The clinicaland research leads in the trial are experienced in studiesof this type and therefore able to take steps to discernany potential risk of contamination bias and avoid it

Baseline characterisation and outcome measuresBaseline characterisation measuresThe following baseline measures will be collected

Demographics a bespoke demographics tool hasbeen designed for the study

Autism Diagnostic Observation Schedule-2(ADOS-2 [27]) is an observational assessmentundertaken by a trained researcher It is a semi-structured set of play and social communicationactivities ADOS-2 has demonstrated good inter-rater reliability coefficients (Fleissrsquo ĸ was 38(range 19ndash55) and Cohenrsquos ĸ was 69 (range 61ndash76)for all modules) in the clinical setting [28]

Social Responsiveness Scale ndash Second Edition (SRS-2 [29]) is a 65-item questionnaire measure of socialand communication features The pre-school andschool-age form has demonstrated adequate inter-rater reliability across parent and teacher ratingsCorrelations were 77 and 61 respectively There isstrong consistency across items on the SRS-2 school

age form which yielded a total reliability coefficientof 095 [30]

Child outcome measuresPrimary child outcome measure

Clinical Global Impression-Improvement Scale(CGI-I) [31] provides a standardised framework toassess how much behaviour has improved orworsened relative to the childrsquos baseline state using a7-point scale Research demonstrates that theinterclass correlation coefficient (ICC) betweentherapist and patients rating of CGI-I was 065indicating a moderate to high agreement betweenthe therapists CGI-I rating and patientrsquos perspectiveon their condition [32]

Secondary child outcome measures

Target Behaviour Vignette [33] As part of thebaseline assessment we require parents to undertakea detailed semi-structured interview during whichthey are required to identify two challenging RRBThe protocol was developed by the Research Unitson Paediatric Psychopharmacology (RUPP) AutismNetwork [33] and has demonstrated high levels ofagreement between expert raters Arnold et al [33]reported an Intraclass Correlation Coefficient (ICC)of 0895 across a panel of five raters

Repetitive Behaviour Questionnaire - 2 (RBQ-2) [34]The RBQ-2 is a 20-item questionnaire that measuresthe frequency and intensity of RRB RBQ-2demonstrates good internal consistency for RRBs forchildren with autism from 2 to 17 years Theinternal consistency was high for the Total RBQ-2scale (Cronbachrsquos alpha = 86 for 20 items and 86 for19 items) [35]

Teacher Repetitive Behaviour Questionnaire 2(Teacher RBQ-2) [36] This is the correspondingversion of the parent RBQ-2 for completion byteachersteaching staff

Vineland Adaptive Behaviour Scales3 (VABS 3) [37]The VABS 3 measures aspects of the childrsquos level ofadaptive functioning VABS 3 has demonstratedexcellent internal consistency (Cronbachrsquos alpha 090to 098) and content construct and concurrentvalidity [37] It has been used in numerous autismstudies

Economic outcome measures

Primary economic outcome measure Incremental costs to achieve target difference in the

CGI-I at 24 weeks The improvement scores from

Grahame et al Trials (2021) 22240 Page 5 of 13

the CGI-I will be taken from each randomised armof the trial to inform the efficiency of theintervention A cost per incremental improvementof CGI-I scale will be calculated in each pathway

Secondary economic outcome measures Costs to the family Cost to the family related to

MRB will be estimated Resources questionnairesand time and travel cost questionnaires are used toaid the estimation of these costs Total cost to thefamily will be reported at 24 and 52 weeks

Incremental cost per QALY gained for the childThe CHU9D [38] will be measured in both arms ofthe trial to measure quality of life in relation to thechild This will be measured at baseline 24 and 52weeks The CHU9D proxy version will be used [38]The CHU9D is a paediatric generic preference-basedmeasure of health-related quality of life that issuitable for use in this particular patient group Thismeasure has an acceptable level of internalconsistency (Cronbach alpha 0781) [39] Followingrecommended practice parentscaregivers will beasked to complete the CHU9D with the child at thethree time points The responses from thisinstrument will be used to create utility valueswhich will be incorporated in QALY outcomes Thiswill be expressed an in an average incremental costsper QALY ratio for the children in each arm

QALYs for the caregiver The EQ-5D-5L [40] will becompleted at baseline 24 and 52 weeks by thecaregiver for the child The EQ-5D-5L is astandardised instrument for use as a measure ofhealth outcome It is applicable to a wide range ofhealth conditions and treatments The EQ-5D-5Lhealth questionnaire provides a simple descriptiveprofile and a single index value for health statusInitial psychometric properties of the 5-level EQ-5Dshow that results for test-retest reliability for averageICCs is 069 and VAS 051 indicating moderatereliability [41] The responses from this instrumentwill be used to create utility values which will beused to create QALYs for the caregivers Thisoutcome will be included as part of the costconsequence analysis

Cost-consequences A number of primary andsecondary clinical outcomes quality of life effects forthe child and quality of life effects for the caregiverswill be used as outcomes for the cost-consequencesanalysis

Use of the health care resources Resources whichare used by the children will be measured using abespoke questionnaire Parentscaregivers will beasked to report the amount of times that their childwith ASD accesses certain services (eg GP or

outpatient appointments) This information will beused to calculate the average cost of services foreach intervention

Family outcome measures

The Autism Family Experience Questionnaire(AFEQ) [7] This questionnaire measures broaderimpact of an intervention on the family Parent(Cronbachrsquos alpha 085) Family (083) ChildDevelopment (081) Child symptoms (079) AFEQtotal (092) [7]

Secondary parent outcome measures

Parent self-efficacy [42] This 15-item self-reportquestionnaire that rates parental confidence inmanaging behaviours on a 6-point scale rangingfrom 0 (no confidence) to 5 (complete confidence)

Autism Parenting Stress Index (APSI) [43] This is ameasure of parenting stress specific to core and co-occurring features of autism The overall APSI scalescore demonstrates acceptable internal consistency(Cronbachrsquos alpha 083) and test-retest stability(Cronbachrsquos alpha 088) for parents of children withautism [43] Psychometric properties are good (egCronbachrsquos alpha 083)

Warwick-Edinburgh Mental Wellbeing Scale(WEMWBS) [44] is a 14-item general populationscreen of wellbeing and is psychometrically robustwith good internal consistency (Cronbachrsquos alpha089) and test-retest reliability (ICC 083)

Participant timelineProcedures

Data collection Baseline assessment and follow-up mea-sures will be collected by RAs blinded to the outcome ofrandomisation RAs are trained to high levels of reliabil-ity in all baseline characterisation and outcome measures(see Table 1) All families will be allocated a uniquenumber that will be used to identify them on all paperassessment forms throughout the trial All data collectedon paper will be inputted into a data management sys-tem for statistical analysis and all identifying data will bestored securely separately The Clinical Data Manage-ment System (Elsevierrsquos MACRO) used for this trial isfully compliant with all regulatory frameworks forresearch of this nature Patients cannot be identifiedfrom eCRFs The CI or delegated person will monitorcompleteness and quality of data recording in eCRFsand will correspond regularly with site PIs (or theirdelegated team member) with the aim of capturing anymissing data where possible and ensuring continuous

Grahame et al Trials (2021) 22240 Page 6 of 13

high quality of data All study data will be treated in ac-cordance with the latest Directive on Good ClinicalPractice (200528EC)Randomisation will be done through the Sealed Enve-

lope (wwwsealedenvelopecom) web-based randomisa-tion service Allocation will be by a minimisation schemeinstead of stratified randomisation to minimise samplefragmentation because of the number of strata and toavoid accidental imbalance between the MRB group andthe LAA group Child level randomisation was preferredcontrolling for child level characteristics that could affectthe primary outcome treatment centre age strata (under5th birthday 5th birthday and older) ethnicity and gen-der Each case will be assigned a participant ID numberand treatment allocation emailed separately to the clinicalleads at each site The clinical leads will inform thefamilies of the outcome of randomisation and this will alsobe recorded in the participantrsquos NHS patient recordWe will collect information about severe adverse

events as well as recording severe adverse events in astandard format we will include events of special

interest particularly relevant to this trial such as signifi-cant changes in family or school situation

Data managementA Data Monitoring Committee (DMC) will meet once ayear to receive reports on recruitment and severe ad-verse events The DMC will evaluate the findings of theinternal pilot and submit report to NIHR independentlychaired and with an independent statistician Severeadverse events and actions taken will be logged by thesenior trial manager (CTU) and a report presented toDMC The RA will discuss challenges with datacollection with research leads at each site and the coreresearch team The RAs will be responsible for ongoingreview of data completeness and any concerns will bediscussed within the core research team CTU and theSponsor as appropriatePrimary secondary and relevant exploratory and

sensitivity analyses of the data will take place by the trialstatistician and health economist in collaborationwith the Chief Investigator The senior statistician

Table 1 Time points at which measures and data are collected

Procedure Screening Baseline Treatment phase Follow-up

Weeks 1ndash8 Week 10 Week 24 Week 52

Informed consent X

Child and parent demographics X X X

Eligibility X

ADOS-2 X

SRS-2 X

CGI-I X

RBQ-2 X X X X

Teacher RBQ-2 X X X X

Measurement of the target behaviour vignette X X X X

VABS 3 X X

Parent self-efficacy questionnaire X X X X

Autism Parenting Stress Index X X X X

WEMWBS X X X

Autism Family Experience Questionnaire X X X

Proxy completed CHU9D X X X

EQ-5D-5L X X X

Resource use questionnaire X X X

Time and travel questionnaire X

Randomisation X

Weekly intervention (MRB or Learning About Autism) X

Child demographics to include child age gender type of nurseryschool diagnosis current medications additional diagnoses ethnicity and previousintervention exposureParent demographics to include level of education employment status family structure attendance at previous courses or interventions relating to children witha diagnosis of ASDRandomisation to take place following completion of baseline assessment

Grahame et al Trials (2021) 22240 Page 7 of 13

strategies to address challenging RRBs lsquoAt homersquo activ-ities will be set each week for parents and children tocomplete between sessions The opportunity to identifyand practice new ideas outside of the clinic setting isimportant for investigating how best to generalise theapproach into different settings for each child andfamilyParents have the opportunities for individual support

and group learning In weeks 2 and 7 they meet thegroup leaders individually to select a target challengingRRB to focus on during the group and to practise thenew skills they are learning thus ensuring that strategiesare individually tailored for each child and are develop-mentally appropriate

Learning About Autism (LAA) parent groupLAA is an 8-week manualised parent psychoeducationgroup that will act as an attentional control LAA isdesigned for parents of young children with autism andfocuses on understanding what autism is and providessupport for parents of young children with autism It isdesigned to be run in the community with professionalswho have experience of working with young childrenwith autism and their families and have been trained bya UK-based autism charity the National Autistic Societyto deliver the intervention Each weekly session lasts forapproximately 2 h (total duration of group interventionis ~ 16 h)The groups will offer parents and carers psychoeduca-

tion on understanding autism and what that means fortheir individual child such as helping them understandtheir childrsquos social communication difficulties and behav-iour in different environments It will also provide adviceand guidance on strategies and approaches for dealingwith behaviour problems in young children with ASDThere will be opportunities for mutual support and shar-ing of ideas with other parents Thus increasing parentsrsquoand carersrsquo understanding confidence and responsive-ness to their childrsquos patterns of communication andinteraction This is equivalent to current best practice Itwill not include any specific information about the roleand functions of RRB functional analysis and tailoredstrategies to manage challenging RRB

Recruitment and eligibility criteriaFamilies will be identified via local community NationalHealth Service (NHS) clinical services Clinicians areasked to introduce the study to families fitting the inclu-sion criteria discuss the study with them and if inter-ested give them study packs For some clinical serviceswhere children may not be seen as frequently postalstudy packs will be sent to identified families Interestedparents are asked to complete and return to the researchteam an expression of interest form The research team

will then contact parents they are interested in hearingmore about the study to arrange a face-to-face meetingto discuss the study what participation involves answerany questions they may have and obtain written in-formed consentParents are eligible for study entry if their child meets

the following criteria (1) aged 3ndash9 years 11 months atthe time of consent with a clinical diagnosis of autism orautism spectrum disorder (2) experience challengingRRB Parents themselves will have (1) sufficient spokenand written English to provide written informed consentcomplete assessments and participate in the interven-tion (2) are willing to be randomised and attend allgroup sessions for the allocated arm of the study and(3) agree to maintain their childrsquos current medication re-gime until the primary 24-week end point (clinicianchanges to medication are permitted) Parents are noteligible for study entry if their child meets the followingcriteria (1) no clinical diagnosis of autism or ASD (2)no challenging RRB (3) is currently taking part in an-other parent group based intervention trial or (4) has asibling already taking part in this study Furthermorechildren of parents who have significant mental healthdifficulties will not be eligible to participate There willbe no special criteria for discontinuing or modifying al-located interventions

Training and fidelityAll parent group sessions in both arms of the trial willbe recorded both to facilitate opportunities for supervi-sion of the group leaders and to provide access to arandom sample of recorded sessions across all sites forthe independent evaluation of (i) the fidelity to thedelivery of the intervention and (ii) the therapeuticcompetence of the individual group leadersrsquo delivery ofthis manualised intervention Fidelity will be formallyassessed by independent raters watching 10 ofrandomly selected tapes and utilising the MRB FidelityRating Scale as developed during the pilot feasibility andacceptability study with adaptations for the content ofthe LAA parent groupsAt the start of the research the group leaders and site

clinical leads will attend a 1 day MRB training courselearning how to use the training manual and the mate-rials for each 2 h weekly session and planning for home-work tasks Each group leader will learn about groupprocesses and gain skills to deliver the manualised inter-vention in a participatory style in combination withstrategies to personalise the programme for each parentThe senior trainer (LD) alongside the chief investigator(VG) will deliver the introductory training course LDwill also visit each site as each new group is started Fur-thermore to ensure the intervention is being deliveredto the families as intended with high fidelity to the

Grahame et al Trials (2021) 22240 Page 4 of 13

manual the senior trainer will alongside the local siteclinical lead will undertake weekly supervision of groupleaders at each site Group Leaders delivering the LAAparent group will attend a 1-day training course specific-ally designed for this study to ensure they are trained inthe skills needed to deliver the programme This coursehas been designed by the National Autistic Society andwill be delivered by DGParent attendance at group sessions will be recorded

and monitored Parents who miss a parent group sessionwill be contacted by a group leader to provide a briefcatch up of session and handouts for that session will beposted out Therapy compliance and receipt of otherinterventions outside of the protocol will be monitored

Avoidance of contaminationThere will be separate clinical and research leads at eachsite and separate training and supervision structures Re-searchers will be located separately from staff involvedin delivery of the MRB and LAA intervention Researchinterviews and assessments will be conducted mostly athome visits with the Autism Diagnostic ObservationSchedule-2 (ADOS-2) being undertaken at Universitysites All parents will be reminded prior to every contactwith research staff not to talk about group allocationFurthermore we will also consider issues around poten-tial contamination of the active attentional control LAAgroup by ensuring the LAA families are not on the caseload of the MRB intervention group leaders The clinicaland research leads in the trial are experienced in studiesof this type and therefore able to take steps to discernany potential risk of contamination bias and avoid it

Baseline characterisation and outcome measuresBaseline characterisation measuresThe following baseline measures will be collected

Demographics a bespoke demographics tool hasbeen designed for the study

Autism Diagnostic Observation Schedule-2(ADOS-2 [27]) is an observational assessmentundertaken by a trained researcher It is a semi-structured set of play and social communicationactivities ADOS-2 has demonstrated good inter-rater reliability coefficients (Fleissrsquo ĸ was 38(range 19ndash55) and Cohenrsquos ĸ was 69 (range 61ndash76)for all modules) in the clinical setting [28]

Social Responsiveness Scale ndash Second Edition (SRS-2 [29]) is a 65-item questionnaire measure of socialand communication features The pre-school andschool-age form has demonstrated adequate inter-rater reliability across parent and teacher ratingsCorrelations were 77 and 61 respectively There isstrong consistency across items on the SRS-2 school

age form which yielded a total reliability coefficientof 095 [30]

Child outcome measuresPrimary child outcome measure

Clinical Global Impression-Improvement Scale(CGI-I) [31] provides a standardised framework toassess how much behaviour has improved orworsened relative to the childrsquos baseline state using a7-point scale Research demonstrates that theinterclass correlation coefficient (ICC) betweentherapist and patients rating of CGI-I was 065indicating a moderate to high agreement betweenthe therapists CGI-I rating and patientrsquos perspectiveon their condition [32]

Secondary child outcome measures

Target Behaviour Vignette [33] As part of thebaseline assessment we require parents to undertakea detailed semi-structured interview during whichthey are required to identify two challenging RRBThe protocol was developed by the Research Unitson Paediatric Psychopharmacology (RUPP) AutismNetwork [33] and has demonstrated high levels ofagreement between expert raters Arnold et al [33]reported an Intraclass Correlation Coefficient (ICC)of 0895 across a panel of five raters

Repetitive Behaviour Questionnaire - 2 (RBQ-2) [34]The RBQ-2 is a 20-item questionnaire that measuresthe frequency and intensity of RRB RBQ-2demonstrates good internal consistency for RRBs forchildren with autism from 2 to 17 years Theinternal consistency was high for the Total RBQ-2scale (Cronbachrsquos alpha = 86 for 20 items and 86 for19 items) [35]

Teacher Repetitive Behaviour Questionnaire 2(Teacher RBQ-2) [36] This is the correspondingversion of the parent RBQ-2 for completion byteachersteaching staff

Vineland Adaptive Behaviour Scales3 (VABS 3) [37]The VABS 3 measures aspects of the childrsquos level ofadaptive functioning VABS 3 has demonstratedexcellent internal consistency (Cronbachrsquos alpha 090to 098) and content construct and concurrentvalidity [37] It has been used in numerous autismstudies

Economic outcome measures

Primary economic outcome measure Incremental costs to achieve target difference in the

CGI-I at 24 weeks The improvement scores from

Grahame et al Trials (2021) 22240 Page 5 of 13

the CGI-I will be taken from each randomised armof the trial to inform the efficiency of theintervention A cost per incremental improvementof CGI-I scale will be calculated in each pathway

Secondary economic outcome measures Costs to the family Cost to the family related to

MRB will be estimated Resources questionnairesand time and travel cost questionnaires are used toaid the estimation of these costs Total cost to thefamily will be reported at 24 and 52 weeks

Incremental cost per QALY gained for the childThe CHU9D [38] will be measured in both arms ofthe trial to measure quality of life in relation to thechild This will be measured at baseline 24 and 52weeks The CHU9D proxy version will be used [38]The CHU9D is a paediatric generic preference-basedmeasure of health-related quality of life that issuitable for use in this particular patient group Thismeasure has an acceptable level of internalconsistency (Cronbach alpha 0781) [39] Followingrecommended practice parentscaregivers will beasked to complete the CHU9D with the child at thethree time points The responses from thisinstrument will be used to create utility valueswhich will be incorporated in QALY outcomes Thiswill be expressed an in an average incremental costsper QALY ratio for the children in each arm

QALYs for the caregiver The EQ-5D-5L [40] will becompleted at baseline 24 and 52 weeks by thecaregiver for the child The EQ-5D-5L is astandardised instrument for use as a measure ofhealth outcome It is applicable to a wide range ofhealth conditions and treatments The EQ-5D-5Lhealth questionnaire provides a simple descriptiveprofile and a single index value for health statusInitial psychometric properties of the 5-level EQ-5Dshow that results for test-retest reliability for averageICCs is 069 and VAS 051 indicating moderatereliability [41] The responses from this instrumentwill be used to create utility values which will beused to create QALYs for the caregivers Thisoutcome will be included as part of the costconsequence analysis

Cost-consequences A number of primary andsecondary clinical outcomes quality of life effects forthe child and quality of life effects for the caregiverswill be used as outcomes for the cost-consequencesanalysis

Use of the health care resources Resources whichare used by the children will be measured using abespoke questionnaire Parentscaregivers will beasked to report the amount of times that their childwith ASD accesses certain services (eg GP or

outpatient appointments) This information will beused to calculate the average cost of services foreach intervention

Family outcome measures

The Autism Family Experience Questionnaire(AFEQ) [7] This questionnaire measures broaderimpact of an intervention on the family Parent(Cronbachrsquos alpha 085) Family (083) ChildDevelopment (081) Child symptoms (079) AFEQtotal (092) [7]

Secondary parent outcome measures

Parent self-efficacy [42] This 15-item self-reportquestionnaire that rates parental confidence inmanaging behaviours on a 6-point scale rangingfrom 0 (no confidence) to 5 (complete confidence)

Autism Parenting Stress Index (APSI) [43] This is ameasure of parenting stress specific to core and co-occurring features of autism The overall APSI scalescore demonstrates acceptable internal consistency(Cronbachrsquos alpha 083) and test-retest stability(Cronbachrsquos alpha 088) for parents of children withautism [43] Psychometric properties are good (egCronbachrsquos alpha 083)

Warwick-Edinburgh Mental Wellbeing Scale(WEMWBS) [44] is a 14-item general populationscreen of wellbeing and is psychometrically robustwith good internal consistency (Cronbachrsquos alpha089) and test-retest reliability (ICC 083)

Participant timelineProcedures

Data collection Baseline assessment and follow-up mea-sures will be collected by RAs blinded to the outcome ofrandomisation RAs are trained to high levels of reliabil-ity in all baseline characterisation and outcome measures(see Table 1) All families will be allocated a uniquenumber that will be used to identify them on all paperassessment forms throughout the trial All data collectedon paper will be inputted into a data management sys-tem for statistical analysis and all identifying data will bestored securely separately The Clinical Data Manage-ment System (Elsevierrsquos MACRO) used for this trial isfully compliant with all regulatory frameworks forresearch of this nature Patients cannot be identifiedfrom eCRFs The CI or delegated person will monitorcompleteness and quality of data recording in eCRFsand will correspond regularly with site PIs (or theirdelegated team member) with the aim of capturing anymissing data where possible and ensuring continuous

Grahame et al Trials (2021) 22240 Page 6 of 13

high quality of data All study data will be treated in ac-cordance with the latest Directive on Good ClinicalPractice (200528EC)Randomisation will be done through the Sealed Enve-

lope (wwwsealedenvelopecom) web-based randomisa-tion service Allocation will be by a minimisation schemeinstead of stratified randomisation to minimise samplefragmentation because of the number of strata and toavoid accidental imbalance between the MRB group andthe LAA group Child level randomisation was preferredcontrolling for child level characteristics that could affectthe primary outcome treatment centre age strata (under5th birthday 5th birthday and older) ethnicity and gen-der Each case will be assigned a participant ID numberand treatment allocation emailed separately to the clinicalleads at each site The clinical leads will inform thefamilies of the outcome of randomisation and this will alsobe recorded in the participantrsquos NHS patient recordWe will collect information about severe adverse

events as well as recording severe adverse events in astandard format we will include events of special

interest particularly relevant to this trial such as signifi-cant changes in family or school situation

Data managementA Data Monitoring Committee (DMC) will meet once ayear to receive reports on recruitment and severe ad-verse events The DMC will evaluate the findings of theinternal pilot and submit report to NIHR independentlychaired and with an independent statistician Severeadverse events and actions taken will be logged by thesenior trial manager (CTU) and a report presented toDMC The RA will discuss challenges with datacollection with research leads at each site and the coreresearch team The RAs will be responsible for ongoingreview of data completeness and any concerns will bediscussed within the core research team CTU and theSponsor as appropriatePrimary secondary and relevant exploratory and

sensitivity analyses of the data will take place by the trialstatistician and health economist in collaborationwith the Chief Investigator The senior statistician

Table 1 Time points at which measures and data are collected

Procedure Screening Baseline Treatment phase Follow-up

Weeks 1ndash8 Week 10 Week 24 Week 52

Informed consent X

Child and parent demographics X X X

Eligibility X

ADOS-2 X

SRS-2 X

CGI-I X

RBQ-2 X X X X

Teacher RBQ-2 X X X X

Measurement of the target behaviour vignette X X X X

VABS 3 X X

Parent self-efficacy questionnaire X X X X

Autism Parenting Stress Index X X X X

WEMWBS X X X

Autism Family Experience Questionnaire X X X

Proxy completed CHU9D X X X

EQ-5D-5L X X X

Resource use questionnaire X X X

Time and travel questionnaire X

Randomisation X

Weekly intervention (MRB or Learning About Autism) X

Child demographics to include child age gender type of nurseryschool diagnosis current medications additional diagnoses ethnicity and previousintervention exposureParent demographics to include level of education employment status family structure attendance at previous courses or interventions relating to children witha diagnosis of ASDRandomisation to take place following completion of baseline assessment

Grahame et al Trials (2021) 22240 Page 7 of 13

manual the senior trainer will alongside the local siteclinical lead will undertake weekly supervision of groupleaders at each site Group Leaders delivering the LAAparent group will attend a 1-day training course specific-ally designed for this study to ensure they are trained inthe skills needed to deliver the programme This coursehas been designed by the National Autistic Society andwill be delivered by DGParent attendance at group sessions will be recorded

and monitored Parents who miss a parent group sessionwill be contacted by a group leader to provide a briefcatch up of session and handouts for that session will beposted out Therapy compliance and receipt of otherinterventions outside of the protocol will be monitored

Avoidance of contaminationThere will be separate clinical and research leads at eachsite and separate training and supervision structures Re-searchers will be located separately from staff involvedin delivery of the MRB and LAA intervention Researchinterviews and assessments will be conducted mostly athome visits with the Autism Diagnostic ObservationSchedule-2 (ADOS-2) being undertaken at Universitysites All parents will be reminded prior to every contactwith research staff not to talk about group allocationFurthermore we will also consider issues around poten-tial contamination of the active attentional control LAAgroup by ensuring the LAA families are not on the caseload of the MRB intervention group leaders The clinicaland research leads in the trial are experienced in studiesof this type and therefore able to take steps to discernany potential risk of contamination bias and avoid it

Baseline characterisation and outcome measuresBaseline characterisation measuresThe following baseline measures will be collected

Demographics a bespoke demographics tool hasbeen designed for the study

Autism Diagnostic Observation Schedule-2(ADOS-2 [27]) is an observational assessmentundertaken by a trained researcher It is a semi-structured set of play and social communicationactivities ADOS-2 has demonstrated good inter-rater reliability coefficients (Fleissrsquo ĸ was 38(range 19ndash55) and Cohenrsquos ĸ was 69 (range 61ndash76)for all modules) in the clinical setting [28]

Social Responsiveness Scale ndash Second Edition (SRS-2 [29]) is a 65-item questionnaire measure of socialand communication features The pre-school andschool-age form has demonstrated adequate inter-rater reliability across parent and teacher ratingsCorrelations were 77 and 61 respectively There isstrong consistency across items on the SRS-2 school

age form which yielded a total reliability coefficientof 095 [30]

Child outcome measuresPrimary child outcome measure

Clinical Global Impression-Improvement Scale(CGI-I) [31] provides a standardised framework toassess how much behaviour has improved orworsened relative to the childrsquos baseline state using a7-point scale Research demonstrates that theinterclass correlation coefficient (ICC) betweentherapist and patients rating of CGI-I was 065indicating a moderate to high agreement betweenthe therapists CGI-I rating and patientrsquos perspectiveon their condition [32]

Secondary child outcome measures

Target Behaviour Vignette [33] As part of thebaseline assessment we require parents to undertakea detailed semi-structured interview during whichthey are required to identify two challenging RRBThe protocol was developed by the Research Unitson Paediatric Psychopharmacology (RUPP) AutismNetwork [33] and has demonstrated high levels ofagreement between expert raters Arnold et al [33]reported an Intraclass Correlation Coefficient (ICC)of 0895 across a panel of five raters

Repetitive Behaviour Questionnaire - 2 (RBQ-2) [34]The RBQ-2 is a 20-item questionnaire that measuresthe frequency and intensity of RRB RBQ-2demonstrates good internal consistency for RRBs forchildren with autism from 2 to 17 years Theinternal consistency was high for the Total RBQ-2scale (Cronbachrsquos alpha = 86 for 20 items and 86 for19 items) [35]

Teacher Repetitive Behaviour Questionnaire 2(Teacher RBQ-2) [36] This is the correspondingversion of the parent RBQ-2 for completion byteachersteaching staff

Vineland Adaptive Behaviour Scales3 (VABS 3) [37]The VABS 3 measures aspects of the childrsquos level ofadaptive functioning VABS 3 has demonstratedexcellent internal consistency (Cronbachrsquos alpha 090to 098) and content construct and concurrentvalidity [37] It has been used in numerous autismstudies

Economic outcome measures

Primary economic outcome measure Incremental costs to achieve target difference in the

CGI-I at 24 weeks The improvement scores from

Grahame et al Trials (2021) 22240 Page 5 of 13

the CGI-I will be taken from each randomised armof the trial to inform the efficiency of theintervention A cost per incremental improvementof CGI-I scale will be calculated in each pathway

Secondary economic outcome measures Costs to the family Cost to the family related to

MRB will be estimated Resources questionnairesand time and travel cost questionnaires are used toaid the estimation of these costs Total cost to thefamily will be reported at 24 and 52 weeks

Incremental cost per QALY gained for the childThe CHU9D [38] will be measured in both arms ofthe trial to measure quality of life in relation to thechild This will be measured at baseline 24 and 52weeks The CHU9D proxy version will be used [38]The CHU9D is a paediatric generic preference-basedmeasure of health-related quality of life that issuitable for use in this particular patient group Thismeasure has an acceptable level of internalconsistency (Cronbach alpha 0781) [39] Followingrecommended practice parentscaregivers will beasked to complete the CHU9D with the child at thethree time points The responses from thisinstrument will be used to create utility valueswhich will be incorporated in QALY outcomes Thiswill be expressed an in an average incremental costsper QALY ratio for the children in each arm

QALYs for the caregiver The EQ-5D-5L [40] will becompleted at baseline 24 and 52 weeks by thecaregiver for the child The EQ-5D-5L is astandardised instrument for use as a measure ofhealth outcome It is applicable to a wide range ofhealth conditions and treatments The EQ-5D-5Lhealth questionnaire provides a simple descriptiveprofile and a single index value for health statusInitial psychometric properties of the 5-level EQ-5Dshow that results for test-retest reliability for averageICCs is 069 and VAS 051 indicating moderatereliability [41] The responses from this instrumentwill be used to create utility values which will beused to create QALYs for the caregivers Thisoutcome will be included as part of the costconsequence analysis

Cost-consequences A number of primary andsecondary clinical outcomes quality of life effects forthe child and quality of life effects for the caregiverswill be used as outcomes for the cost-consequencesanalysis

Use of the health care resources Resources whichare used by the children will be measured using abespoke questionnaire Parentscaregivers will beasked to report the amount of times that their childwith ASD accesses certain services (eg GP or

outpatient appointments) This information will beused to calculate the average cost of services foreach intervention

Family outcome measures

The Autism Family Experience Questionnaire(AFEQ) [7] This questionnaire measures broaderimpact of an intervention on the family Parent(Cronbachrsquos alpha 085) Family (083) ChildDevelopment (081) Child symptoms (079) AFEQtotal (092) [7]

Secondary parent outcome measures

Parent self-efficacy [42] This 15-item self-reportquestionnaire that rates parental confidence inmanaging behaviours on a 6-point scale rangingfrom 0 (no confidence) to 5 (complete confidence)

Autism Parenting Stress Index (APSI) [43] This is ameasure of parenting stress specific to core and co-occurring features of autism The overall APSI scalescore demonstrates acceptable internal consistency(Cronbachrsquos alpha 083) and test-retest stability(Cronbachrsquos alpha 088) for parents of children withautism [43] Psychometric properties are good (egCronbachrsquos alpha 083)

Warwick-Edinburgh Mental Wellbeing Scale(WEMWBS) [44] is a 14-item general populationscreen of wellbeing and is psychometrically robustwith good internal consistency (Cronbachrsquos alpha089) and test-retest reliability (ICC 083)

Participant timelineProcedures

Data collection Baseline assessment and follow-up mea-sures will be collected by RAs blinded to the outcome ofrandomisation RAs are trained to high levels of reliabil-ity in all baseline characterisation and outcome measures(see Table 1) All families will be allocated a uniquenumber that will be used to identify them on all paperassessment forms throughout the trial All data collectedon paper will be inputted into a data management sys-tem for statistical analysis and all identifying data will bestored securely separately The Clinical Data Manage-ment System (Elsevierrsquos MACRO) used for this trial isfully compliant with all regulatory frameworks forresearch of this nature Patients cannot be identifiedfrom eCRFs The CI or delegated person will monitorcompleteness and quality of data recording in eCRFsand will correspond regularly with site PIs (or theirdelegated team member) with the aim of capturing anymissing data where possible and ensuring continuous

Grahame et al Trials (2021) 22240 Page 6 of 13

high quality of data All study data will be treated in ac-cordance with the latest Directive on Good ClinicalPractice (200528EC)Randomisation will be done through the Sealed Enve-

lope (wwwsealedenvelopecom) web-based randomisa-tion service Allocation will be by a minimisation schemeinstead of stratified randomisation to minimise samplefragmentation because of the number of strata and toavoid accidental imbalance between the MRB group andthe LAA group Child level randomisation was preferredcontrolling for child level characteristics that could affectthe primary outcome treatment centre age strata (under5th birthday 5th birthday and older) ethnicity and gen-der Each case will be assigned a participant ID numberand treatment allocation emailed separately to the clinicalleads at each site The clinical leads will inform thefamilies of the outcome of randomisation and this will alsobe recorded in the participantrsquos NHS patient recordWe will collect information about severe adverse

events as well as recording severe adverse events in astandard format we will include events of special

interest particularly relevant to this trial such as signifi-cant changes in family or school situation

Data managementA Data Monitoring Committee (DMC) will meet once ayear to receive reports on recruitment and severe ad-verse events The DMC will evaluate the findings of theinternal pilot and submit report to NIHR independentlychaired and with an independent statistician Severeadverse events and actions taken will be logged by thesenior trial manager (CTU) and a report presented toDMC The RA will discuss challenges with datacollection with research leads at each site and the coreresearch team The RAs will be responsible for ongoingreview of data completeness and any concerns will bediscussed within the core research team CTU and theSponsor as appropriatePrimary secondary and relevant exploratory and

sensitivity analyses of the data will take place by the trialstatistician and health economist in collaborationwith the Chief Investigator The senior statistician

Table 1 Time points at which measures and data are collected

Procedure Screening Baseline Treatment phase Follow-up

Weeks 1ndash8 Week 10 Week 24 Week 52

Informed consent X

Child and parent demographics X X X

Eligibility X

ADOS-2 X

SRS-2 X

CGI-I X

RBQ-2 X X X X

Teacher RBQ-2 X X X X

Measurement of the target behaviour vignette X X X X

VABS 3 X X

Parent self-efficacy questionnaire X X X X

Autism Parenting Stress Index X X X X

WEMWBS X X X

Autism Family Experience Questionnaire X X X

Proxy completed CHU9D X X X

EQ-5D-5L X X X

Resource use questionnaire X X X

Time and travel questionnaire X

Randomisation X

Weekly intervention (MRB or Learning About Autism) X

Child demographics to include child age gender type of nurseryschool diagnosis current medications additional diagnoses ethnicity and previousintervention exposureParent demographics to include level of education employment status family structure attendance at previous courses or interventions relating to children witha diagnosis of ASDRandomisation to take place following completion of baseline assessment

Grahame et al Trials (2021) 22240 Page 7 of 13

the CGI-I will be taken from each randomised armof the trial to inform the efficiency of theintervention A cost per incremental improvementof CGI-I scale will be calculated in each pathway

Secondary economic outcome measures Costs to the family Cost to the family related to

MRB will be estimated Resources questionnairesand time and travel cost questionnaires are used toaid the estimation of these costs Total cost to thefamily will be reported at 24 and 52 weeks

Incremental cost per QALY gained for the childThe CHU9D [38] will be measured in both arms ofthe trial to measure quality of life in relation to thechild This will be measured at baseline 24 and 52weeks The CHU9D proxy version will be used [38]The CHU9D is a paediatric generic preference-basedmeasure of health-related quality of life that issuitable for use in this particular patient group Thismeasure has an acceptable level of internalconsistency (Cronbach alpha 0781) [39] Followingrecommended practice parentscaregivers will beasked to complete the CHU9D with the child at thethree time points The responses from thisinstrument will be used to create utility valueswhich will be incorporated in QALY outcomes Thiswill be expressed an in an average incremental costsper QALY ratio for the children in each arm

QALYs for the caregiver The EQ-5D-5L [40] will becompleted at baseline 24 and 52 weeks by thecaregiver for the child The EQ-5D-5L is astandardised instrument for use as a measure ofhealth outcome It is applicable to a wide range ofhealth conditions and treatments The EQ-5D-5Lhealth questionnaire provides a simple descriptiveprofile and a single index value for health statusInitial psychometric properties of the 5-level EQ-5Dshow that results for test-retest reliability for averageICCs is 069 and VAS 051 indicating moderatereliability [41] The responses from this instrumentwill be used to create utility values which will beused to create QALYs for the caregivers Thisoutcome will be included as part of the costconsequence analysis

Cost-consequences A number of primary andsecondary clinical outcomes quality of life effects forthe child and quality of life effects for the caregiverswill be used as outcomes for the cost-consequencesanalysis

Use of the health care resources Resources whichare used by the children will be measured using abespoke questionnaire Parentscaregivers will beasked to report the amount of times that their childwith ASD accesses certain services (eg GP or

outpatient appointments) This information will beused to calculate the average cost of services foreach intervention

Family outcome measures

The Autism Family Experience Questionnaire(AFEQ) [7] This questionnaire measures broaderimpact of an intervention on the family Parent(Cronbachrsquos alpha 085) Family (083) ChildDevelopment (081) Child symptoms (079) AFEQtotal (092) [7]

Secondary parent outcome measures

Parent self-efficacy [42] This 15-item self-reportquestionnaire that rates parental confidence inmanaging behaviours on a 6-point scale rangingfrom 0 (no confidence) to 5 (complete confidence)

Autism Parenting Stress Index (APSI) [43] This is ameasure of parenting stress specific to core and co-occurring features of autism The overall APSI scalescore demonstrates acceptable internal consistency(Cronbachrsquos alpha 083) and test-retest stability(Cronbachrsquos alpha 088) for parents of children withautism [43] Psychometric properties are good (egCronbachrsquos alpha 083)

Warwick-Edinburgh Mental Wellbeing Scale(WEMWBS) [44] is a 14-item general populationscreen of wellbeing and is psychometrically robustwith good internal consistency (Cronbachrsquos alpha089) and test-retest reliability (ICC 083)

Participant timelineProcedures

Data collection Baseline assessment and follow-up mea-sures will be collected by RAs blinded to the outcome ofrandomisation RAs are trained to high levels of reliabil-ity in all baseline characterisation and outcome measures(see Table 1) All families will be allocated a uniquenumber that will be used to identify them on all paperassessment forms throughout the trial All data collectedon paper will be inputted into a data management sys-tem for statistical analysis and all identifying data will bestored securely separately The Clinical Data Manage-ment System (Elsevierrsquos MACRO) used for this trial isfully compliant with all regulatory frameworks forresearch of this nature Patients cannot be identifiedfrom eCRFs The CI or delegated person will monitorcompleteness and quality of data recording in eCRFsand will correspond regularly with site PIs (or theirdelegated team member) with the aim of capturing anymissing data where possible and ensuring continuous

Grahame et al Trials (2021) 22240 Page 6 of 13

high quality of data All study data will be treated in ac-cordance with the latest Directive on Good ClinicalPractice (200528EC)Randomisation will be done through the Sealed Enve-

lope (wwwsealedenvelopecom) web-based randomisa-tion service Allocation will be by a minimisation schemeinstead of stratified randomisation to minimise samplefragmentation because of the number of strata and toavoid accidental imbalance between the MRB group andthe LAA group Child level randomisation was preferredcontrolling for child level characteristics that could affectthe primary outcome treatment centre age strata (under5th birthday 5th birthday and older) ethnicity and gen-der Each case will be assigned a participant ID numberand treatment allocation emailed separately to the clinicalleads at each site The clinical leads will inform thefamilies of the outcome of randomisation and this will alsobe recorded in the participantrsquos NHS patient recordWe will collect information about severe adverse

events as well as recording severe adverse events in astandard format we will include events of special

interest particularly relevant to this trial such as signifi-cant changes in family or school situation

Data managementA Data Monitoring Committee (DMC) will meet once ayear to receive reports on recruitment and severe ad-verse events The DMC will evaluate the findings of theinternal pilot and submit report to NIHR independentlychaired and with an independent statistician Severeadverse events and actions taken will be logged by thesenior trial manager (CTU) and a report presented toDMC The RA will discuss challenges with datacollection with research leads at each site and the coreresearch team The RAs will be responsible for ongoingreview of data completeness and any concerns will bediscussed within the core research team CTU and theSponsor as appropriatePrimary secondary and relevant exploratory and

sensitivity analyses of the data will take place by the trialstatistician and health economist in collaborationwith the Chief Investigator The senior statistician

Table 1 Time points at which measures and data are collected

Procedure Screening Baseline Treatment phase Follow-up

Weeks 1ndash8 Week 10 Week 24 Week 52

Informed consent X

Child and parent demographics X X X

Eligibility X

ADOS-2 X

SRS-2 X

CGI-I X

RBQ-2 X X X X

Teacher RBQ-2 X X X X

Measurement of the target behaviour vignette X X X X

VABS 3 X X

Parent self-efficacy questionnaire X X X X

Autism Parenting Stress Index X X X X

WEMWBS X X X

Autism Family Experience Questionnaire X X X

Proxy completed CHU9D X X X

EQ-5D-5L X X X

Resource use questionnaire X X X

Time and travel questionnaire X

Randomisation X

Weekly intervention (MRB or Learning About Autism) X

Child demographics to include child age gender type of nurseryschool diagnosis current medications additional diagnoses ethnicity and previousintervention exposureParent demographics to include level of education employment status family structure attendance at previous courses or interventions relating to children witha diagnosis of ASDRandomisation to take place following completion of baseline assessment

Grahame et al Trials (2021) 22240 Page 7 of 13

high quality of data All study data will be treated in ac-cordance with the latest Directive on Good ClinicalPractice (200528EC)Randomisation will be done through the Sealed Enve-

lope (wwwsealedenvelopecom) web-based randomisa-tion service Allocation will be by a minimisation schemeinstead of stratified randomisation to minimise samplefragmentation because of the number of strata and toavoid accidental imbalance between the MRB group andthe LAA group Child level randomisation was preferredcontrolling for child level characteristics that could affectthe primary outcome treatment centre age strata (under5th birthday 5th birthday and older) ethnicity and gen-der Each case will be assigned a participant ID numberand treatment allocation emailed separately to the clinicalleads at each site The clinical leads will inform thefamilies of the outcome of randomisation and this will alsobe recorded in the participantrsquos NHS patient recordWe will collect information about severe adverse

events as well as recording severe adverse events in astandard format we will include events of special

interest particularly relevant to this trial such as signifi-cant changes in family or school situation

Data managementA Data Monitoring Committee (DMC) will meet once ayear to receive reports on recruitment and severe ad-verse events The DMC will evaluate the findings of theinternal pilot and submit report to NIHR independentlychaired and with an independent statistician Severeadverse events and actions taken will be logged by thesenior trial manager (CTU) and a report presented toDMC The RA will discuss challenges with datacollection with research leads at each site and the coreresearch team The RAs will be responsible for ongoingreview of data completeness and any concerns will bediscussed within the core research team CTU and theSponsor as appropriatePrimary secondary and relevant exploratory and

sensitivity analyses of the data will take place by the trialstatistician and health economist in collaborationwith the Chief Investigator The senior statistician

Table 1 Time points at which measures and data are collected

Procedure Screening Baseline Treatment phase Follow-up

Weeks 1ndash8 Week 10 Week 24 Week 52

Informed consent X

Child and parent demographics X X X

Eligibility X

ADOS-2 X

SRS-2 X

CGI-I X

RBQ-2 X X X X

Teacher RBQ-2 X X X X

Measurement of the target behaviour vignette X X X X

VABS 3 X X

Parent self-efficacy questionnaire X X X X

Autism Parenting Stress Index X X X X

WEMWBS X X X

Autism Family Experience Questionnaire X X X

Proxy completed CHU9D X X X

EQ-5D-5L X X X

Resource use questionnaire X X X

Time and travel questionnaire X

Randomisation X

Weekly intervention (MRB or Learning About Autism) X

Child demographics to include child age gender type of nurseryschool diagnosis current medications additional diagnoses ethnicity and previousintervention exposureParent demographics to include level of education employment status family structure attendance at previous courses or interventions relating to children witha diagnosis of ASDRandomisation to take place following completion of baseline assessment

Grahame et al Trials (2021) 22240 Page 7 of 13

will independently reproduce the primary analyseswhilst still remaining blinded to the interventiongroups The senior statistician will have an overviewof the entire analyses and will explicitly check thestatistical codes Other members of the team (egthe trial health economist) will also have access todata and will undertake analysis as appropriate andnecessary We have carefully considered the ethicalimplications in relation to this type of parent groupintervention trial and there are no anticipated detri-mental issues to participants There are therefore noplanned interim analyses Any arrangements forother researchers in the general field to have accessto the primary data will be negotiated separately andCOREC informed

Statistical analysisSample sizeWe plan to approach approximately 325 families and ex-pect to randomise 250 families (125 randomised to eacharm) Assuming 5 type I error 90 power 10 intra-group correlation and equal allocation ratio 224 families(an average of 8 families per parent group) are requiredto detect 20 improvement rate between the MRB inter-vention and Learning About Autism group at 24 weeksAllowing for an attrition rate of 12 250 families will berandomised The 10 intra-cluster correlation was basedon review of group interventions in education trials [45]Sample size was calculated in R using n4pros in CRTSizepackage [46]

Analysis planA statistical analysis plan will be written and agreed bythe Trial Steering Committee and Data MonitoringCommittee before any analysis is undertaken All statisticalanalyses will be carried out using the latest version of Rsoftware [46] All analyses will be done in accordance withintention-to-treat principle where all children and parentoutcomes are analysed as randomised In accordance withCONSORT statement for non-pharmacological interven-tions we will report all participant flow Data will be sum-marised by trial arm N Mean plusmn SD (or median plusmn IQR ifdata are skewed) Minimum and Maximum will summarisecontinuous variables whereas number and percentages willbe used to summarise categorical variablesThe analysis of the primary outcome at 24 weeks will

use generalised estimating equations with binomialdistribution and logit link Exchangeable workingcorrelation will be used to account for the clustering ofchildren by parent groups The continuous secondaryoutcomes will first be analysed at 24 weeks using adifference-in-difference model based on linear mixed ef-fect model accounting for paired data (at baseline and at24 weeks) per child and clustering of children by parent

groups The same model will be applied to the data atweek 52 which will be analysed as longitudinal data in-corporating data at baseline and 24 weeks All binary orcategorical secondary outcomes will be analysed usinggeneralised estimating equations We will also performsubgroup analysis and safety analysis sensitivity analysisfor missing data and assess the impact of the COVID-19pandemic on primary and secondary outcomes

Health economicsEconomic evaluationThe economic evaluation will be carried out from theperspective of the NHS and personal social services Acost-effectiveness analysis within trial which will com-pare the costs to achieve the target mean difference inthe CGI-I in both the MRB and Learning About Autismgroups at 24 weeks A cost-utility approach will also beundertaken using the data from the CHU9D questionnaireto synthesise QALYs for the children and compare the in-terventions using an incremental costs per QALY approachTo measure the benefits which would not be captured inthe metric of a QALY Finally a cost-consequence will beused to compare costs and benefits from a wider perspec-tive (for example the broader costs to families)

Measurement of effectsFor the cost-effectiveness analysis the effectivenessmeasure will be based on the results from the primarytrial outcome the target mean difference in the CGI-IThe costs utility analysis will use the responses from

the CHU9D based on the proxy responses from childrsquoscaregiver The CHU9D will be administered at baseline24 weeks and 52 weeks This will measure the quality oflife of the child which will be converted into QALYs foreach child using the under the curve approach and anaverage incremental cost per QALY in each randomisedarm will be calculatedThe caregiver will complete ED-5D-5L at baseline 24

and 52 weeks These responses will measure quality oflife in relation to the caregiver and will be scored usingthe values sets for England This data will also be con-verted into QALYs using the under the curve approachThe QALYs which are calculated for the carers will beincluded as part of the cost-consequence analysis Fur-ther consequences will be examined as part of the costconsequences analysis including primary and secondaryclinical outcomes particularly the health-related qualityof life of the child and their care-giver These will in-clude benefits which cannot be included with the scopeof the QALY outcome

AnalysisFor the cost-effectiveness analysis an incremental costper unit change in the CGI-I scale will be calculated

Grahame et al Trials (2021) 22240 Page 8 of 13

with the aim of calculating the cost for achieving a min-imally important difference in the CGI-I Point estimatesof costs and effects cost effect plots and acceptabilitycurves will be produced Statistical imprecision and un-certainty will be examined using a stochastic sensitivityanalysis The cost-utility analysis will be analysed in asimilar way to the cost-effectiveness analysis A formaldecision analytic model is not currently planned butmay be used if the cost of the intervention is not offsetby a reduction in resource use or gain in QALYs for thechild If the results are conclusive (ie the interventionmore effective and less costly or less effective and morecostly) then a model will not be requiredThe cost-consequence analysis will present the costs

and consequences as a difference between randomisedgroups with appropriate measures of variance

MonitoringData monitoring and ethics committeeThe project has a Trial Steering Committee (TSC) inde-pendently chaired by Professor Patricia Howlin (EmiritusProfessor of Clinical Child Psychology) and is comprisedof a panel of independent members including a Statisti-cian Health Economist and two Parent Representativesas well as Non-Independent members that form theTrial Management Group (TMG) The Data MonitoringCommittee (DMC) is independently chaired by ProfessorJohn Jerrim (Professor of Education and Social Statistics)and comprised of a panel of Independent membersincluding a Principal Clinical Psychologist and Post-Doctoral Research Associate as well as Non-Independentmembers that are part of the Trial Management GroupSerious adverse events and actions taken will be loggedby the senior trial manager (CTU) or trial manager (CTU)and a report presented to DMC The DMEC is independ-ent of sponsor and funder and declares no competing in-terests Further details of the DMEC charter are availablefrom the trial manager

Recording and reporting AEs and SAEs For thepurposes of this trial only serious adverse events(adverse events which meet the criteria for seriousness)will be captured for the parentcarer and child partici-pants Serious adverse events will be captured from thestart date of intervention until the follow-up assessmentat week 24

Events of special interest As well as collecting and en-suring SAEs are reported events of special interest willalso be collected An event of special interest is anyevent relating to child wellbeing and familylife difficul-ties which is not expected and not anticipated in lsquonormalday-to-day lifersquo but is not a physical medical eventEvents of special interest will be recorded for both the

parent and child participants from the start date of theintervention until the follow-up assessment at week 24

DisseminationThe dissemination strategy for this research will includeseveral complementary strands of activity We will workclosely with parents and the autism community toensure that the results are interpreted and reported in ameaningful way The results of the study will be sharedas follows

1 Local dissemination at each site including parentnewsletters

2 Wider national and international disseminationincluding conferences and publications

3 NHS Clinicians and Commissioners events todiscuss the implications of the research

DiscussionThe design will take a rigorous scientific approach byutilising a three-site two-group randomised controlleddesign comparing two active parent group interventionsThis study will evaluate the clinical and cost-effectiveness of the Managing Repetitive Behaviours(MRB) parent group intervention compared with theLearning About Autism psychoeducation parent groupThe Learning About Autism (LAA) parent group deliv-ered by staff trained by The National Autistic Societyand equivalent to best current practice allows us to con-trol for the non-specific social group benefit of mutualsharing of experience and support between parentsParent group based interventions provide opportun-

ities for mutual learning and sharing of ideas allowingparents to discuss how best to support their childrsquosdevelopment This fosters opportunities for parents tolearn from and support each other in turn building par-entsrsquo knowledge and confidence to support their childrsquosneeds However in our MRB parent group interventionwe have included targeted support on challenging re-stricted and repetitive behaviours that are interfering ina deleterious way for the child There is some informa-tion on restricted and repetitive behaviours in the LAAparent group but MRB focuses more on assisting parentsto understand the role that RRB may have in their autis-tic childrsquos life In this way we anticipate that MRB willhave a greater beneficial effect on reducing identifiedchallenging RRB and improving child overall wellbeingMoreover we will evaluate whether providing parents

with the skills to effectively understand their childrsquos chal-lenging RRB has a greater effect on parental wellbeingsense of competence reducing stress and improvingfamily cohesion than a general autism psychoeducationparent group In this way the results from this RCT

Grahame et al Trials (2021) 22240 Page 9 of 13

have the potential to help autism researchers explore theactive ingredients of clinical trialsWe have carefully considered in collaboration with par-

ents and professionals the utility of the outcome measureschosen to assess the effects of the intervention on challen-ging RRB overall child functioning and parent and familywellbeing We have taken a rigorous approach to outcomeassessment with independent objective measures wherepossible We are aware of limitations in autism researchfrom over reliance on unblinded parental report measuresand have therefore included other outcome measures suchas teacher-rated outcomesIf found to be effective and efficient in the proposed

evaluation study this early parent group based interventiontargeting challenging RRB has the potential to fill an identi-fied but unmet need and thus improve the wellbeing ofautistic children and their families reduce parental stressgreatly enhance community participation increase learningopportunities and improve longer-term outcomes

COVID-19 addendumFollowing the announcement from the government on23rd March 2020 in response to the COVID-19 pandemica strict UK lockdown was imposed with implications forall clinical services and research trials across the UK Forour Managing Repetitive Behaviours (MRB) randomisedcontrolled trial (RCT) this has meant a great deal of in-novative creative thinking to design a revised researchprotocol for the conduct of all aspects of the research andthe delivery of the two parent group based interventionsThe aim was to ensure that recruited families and all theclinical and research staff were safe and that all our proce-dures were compliant with national government guidancedocuments and the required health and safety recommen-dations and procedures from the National Institute forHealth Research (NIHR) and the Department of HealthWith support from the research sponsor and funder anagreed plan to do assessments remotely and run parentgroups online was madeA key priority for our clinical research team was to

continue to keep the study open to maintain as best aswe could the integrity of the study and retain recruitedfamilies to the end of the trial despite the restrictionsimposed by the COVID-19 pandemic We are aware thatfamilies of autistic children are facing additional chal-lenges in light of COVID-19 with reduced access to sup-port and services This was an added reason to remainin contact with families recruited to our study We heardfrom parents that many children are finding changes totheir usual routines very challenging and that there wasa disproportionate impact on families caring for autisticchildren This is important information that may impacton the conduct and results of this RCT For all thesereasons we were very keen to keep in touch with our

recruited families for the duration of these challengingtimes However we also needed to take proactive pre-cautions to protect the health of all the families in thetrial and our staff in line with recommendations fromthe government NHS and NIHR and comply with localsite restrictions which varied between sites The trialteam have worked with all sites sponsor and funder toensure the most appropriate actions have been takenThe Trial Management Group with sponsor support on

26th March 2020 agreed all study activities would beperformed remotely by sites wherever possible Followingthis we implemented non-substantial amendment 08(NS08) on 3rd April 2020 as sponsor category C to reflectremote working changes with protocol version 06

COVID-19 remote working measures

1 Study Research Associates (RA) are now workingremotely at home in line with local policiesfollowing government advice and restrictionsInvestigative site files have been re-located toclinical sites wherever possible however access tothese and any paper source documents is noweither limited or not possible at this timea RAs working remotely at home must have

access to secure laptops which connect to theUniversityNHS server as appropriate Phonecalls to parents can be made at home but nopatient identifiable data can be stored in homes

b All new paper source documents including anyRA notes will either be recorded electronicallyfor this period of time so that no patient detailsare stored outside of the local site server orwritten on paper copies with no participantIdentifiable data (PID) where applicable

c Remote supervision of RAs to be arranged ateach site with local principle investigator (PI)oversight and documented

2 Ongoing review of the situation and ability toremain open to any study activitiesa Sites to inform Chief Investigator (CI) and

Newcastle Clinical Trial Unit (NCTU) of anychanges in availability of staffing at each site

b Study recruitment should be halted at individualsite level if sites are no longer able to deliver thestudy An amendment will only be submitted ifthe study is halted at sponsor level as discussedwith the CI and NCTU

c To consider availability of the CI to providecentral study advice

d The NCTU will inform sites of any overall studyhalt It is then the responsibility of the localclinical team to risk assess participants that arealready in the study

Grahame et al Trials (2021) 22240 Page 10 of 13

A suite of documents including a contingency plan andworking instruction have been created to aid all sites inthe day to day workings of the study and remote workingprocedures A sponsor specific contingency document hasalso been created to ensure sponsor is kept up to date withall the changes put in place Furthermore any futurechanges to the study relating to COVID-19 will bereviewed in line with the government guidance updates

Intervention parent groupsWe have explored digital options with our NHS clinicalcolleagues at each of the three sites to identify a secureand reliable platform that allows us to deliver our parentgroups online It was important the technology was easyto use for parents and group leaders but still allowedgroup leaders to present information and participate ingroup chat with familiesFor some families attending online parent groups has

been an easier option for them to consider as it has re-moved some of the barriers such as geographical trans-port work or other commitments that had previouslyprevented them from attending a parent group We aremindful of issues of digital poverty but initial feedbackfrom families has been positive and most families havebeen able to join a parent group using a range of devicessuch as smart phones tablets or computer

Trial statusProtocol Version 60 approval date 02042020The trial is recruiting Recruitment began on the

following dates for each siteCNTW 02102018 Lothian 20112018 TEWV 22

112018Recruitment is due to complete on the 31082020

AbbreviationsAE Adverse event AR Adverse reaction ASD Autism spectrum disorderCHU9D Child Health Utility 9D CI Chief Investigator eCRF Electronic CaseReport Form EoI Expression of interest DMC Data Monitoring CommitteeGCP Good Clinical Practice GP General practice MRB Managing RepetitiveBehaviours NCTU Newcastle Clinical Trials Unit NIHR-HTA National Institutefor Health Research ndash Health Technology Assessment NHS National HealthService PI Principal Investigator RampD Research amp Development RA ResearchAssociate RCT Randomised control trial REC Research Ethics CommitteeRfPB Research for Patient Benefit RRB Restrictive and repetitive behavioursQALY Quality adjusted life year SAE Serious adverse event SAR Seriousadverse reaction SOP Standard operating procedure USAR Unexpectedserious adverse reaction TSC Trial Steering Committee

Supplementary InformationThe online version contains supplementary material available at httpsdoiorg101186s13063-021-05175-y

Additional file 1 SPIRIT Checklist

Additional file 2 Ethical approval documentation HRA ethical approval

Additional file 3 Copy of original funding documentation NIHRfunding documentation

AcknowledgementsWe are extremely grateful to all the parents of autistic children who haveover the past 10 years actively collaborated with us to improve theintervention and research design We also gratefully acknowledge the valuedcontribution of all the participating families in the study and referringclinicians who have helped us with recruitment It really has been a teameffort and without everyonersquos support we would not have been able toundertake this large fully powered multi-site research trialWe also thank all the MRB group leaders Teresa Ingram2 Alan Galvin1 KirstinFarquhar1 Laura Tavernor2 Aimee Corner2 Rosalind Oliphant2 JakeHutchinson3 Emily Thompson2 Hannah Tait3 Courtnie Beasley-Gibson3 LukeRodgers3 Charlotte Randall3 Katie McIntyre3 Claire Cadger3 Jasmine Miller9Calum Parker7 Sinead OprimeCarroll6 Fiona McBryde10 Eleni Alexandropoulou6Hilary Cowie5 Callum MacKinnon8We would like to acknowledge the valued contribution of The NationalAutistic Society Training and Consultancy Team for supporting andorganising the training of the Learning About Autism parent group leadersIn particular David Scowcroft Family Seminar and Support Manager fordeveloping the bespoke training materials for the studyLAA group leaders David Sanderson2 Lisa Slater2 Rachel Lisle2 Maeve Ryan6Marisa McKinlay5 Sheila Kernohan5 Francesca Smart8We thank the members of the Trial Steering Committee Professor PatriciaHowlin Emiritus Professor of Clinical Child Psychology (Chair) ProfessorRichard Emsley Professor of Medical Statistics and Trials Professor HelenMason Professor of Health Economics and our parent representatives MsNicola King and Ms Helen BlackWe also thank the members of the Data Management and EthicsCommittee Professor John Jerrim Professor of Education and Social Statistics(Chair) Dr Kathy Leadbitter Senior Research Associate and Dr AnnOzsivadjian Principal Clinical PsychologistWe acknowledge the contribution to data collection and entry of DeborahJones1 and Ewan Lomax11 Newcastle University 2 Cumbria Tyne and Wear NHS Foundation Trust 3University of Durham 4 Tees Esk and Wear Valleys NHS Foundation Trust 5NHS Lothian 6 Edinburgh University 7 Taylor Ed-Foundation 8 Bouncet-Innovative Occupational Therapy 9 Jasmine Miller Coaching 10 City of Edin-burgh Council

Authorsrsquo contributionsVG JR LD ALC designed the study VG led the trial JR DR SFW researchleads and EH EIW LM ZG clinical leads at Newcastle Teesside Edinburghsites respectively LD lead trainer ASK led the statistical design LV amp AK ledthe economic evaluation MG JG PR LR ST HP SH EW RI researchassociates FW AM trial mangers CW RW database managers VG led thewriting of the paper All authors read contributed to and approved the finalmanuscript

FundingNIHRHTA ProgrammeUniversity of SouthamptonAlpha HouseEnterprise RoadSO16 7NSTelephone 023 8059 5586NIHR Research funding - pound137718055The funding body had no role in the design of the study data collectionanalysis interpretation or in the writing of this manuscript

Availability of data and materialsAll data in the trial will be anonymised A central master file will be held bythe trial manager at the Newcastle Clinical Trials Unit This will contain thekey linking anonymised trial name to personal details This eCRF pack will bebacked up securely within the web-based data entry service of NewcastleUniversity CTU All data will be entered into the Newcastle CTU web-basedsecure MACRO database which has a full audit trail and appropriate qualitycontrol will be carried out during the trial and before the database lockPrimary analysis of the data will take place in Durham University by the trialstatistician The datasets generated andor analysed during the current studyare available from the corresponding author on reasonable request andCOREC informed

Grahame et al Trials (2021) 22240 Page 11 of 13

with the aim of calculating the cost for achieving a min-imally important difference in the CGI-I Point estimatesof costs and effects cost effect plots and acceptabilitycurves will be produced Statistical imprecision and un-certainty will be examined using a stochastic sensitivityanalysis The cost-utility analysis will be analysed in asimilar way to the cost-effectiveness analysis A formaldecision analytic model is not currently planned butmay be used if the cost of the intervention is not offsetby a reduction in resource use or gain in QALYs for thechild If the results are conclusive (ie the interventionmore effective and less costly or less effective and morecostly) then a model will not be requiredThe cost-consequence analysis will present the costs

and consequences as a difference between randomisedgroups with appropriate measures of variance

MonitoringData monitoring and ethics committeeThe project has a Trial Steering Committee (TSC) inde-pendently chaired by Professor Patricia Howlin (EmiritusProfessor of Clinical Child Psychology) and is comprisedof a panel of independent members including a Statisti-cian Health Economist and two Parent Representativesas well as Non-Independent members that form theTrial Management Group (TMG) The Data MonitoringCommittee (DMC) is independently chaired by ProfessorJohn Jerrim (Professor of Education and Social Statistics)and comprised of a panel of Independent membersincluding a Principal Clinical Psychologist and Post-Doctoral Research Associate as well as Non-Independentmembers that are part of the Trial Management GroupSerious adverse events and actions taken will be loggedby the senior trial manager (CTU) or trial manager (CTU)and a report presented to DMC The DMEC is independ-ent of sponsor and funder and declares no competing in-terests Further details of the DMEC charter are availablefrom the trial manager

Recording and reporting AEs and SAEs For thepurposes of this trial only serious adverse events(adverse events which meet the criteria for seriousness)will be captured for the parentcarer and child partici-pants Serious adverse events will be captured from thestart date of intervention until the follow-up assessmentat week 24

Events of special interest As well as collecting and en-suring SAEs are reported events of special interest willalso be collected An event of special interest is anyevent relating to child wellbeing and familylife difficul-ties which is not expected and not anticipated in lsquonormalday-to-day lifersquo but is not a physical medical eventEvents of special interest will be recorded for both the

parent and child participants from the start date of theintervention until the follow-up assessment at week 24

DisseminationThe dissemination strategy for this research will includeseveral complementary strands of activity We will workclosely with parents and the autism community toensure that the results are interpreted and reported in ameaningful way The results of the study will be sharedas follows

1 Local dissemination at each site including parentnewsletters

2 Wider national and international disseminationincluding conferences and publications

3 NHS Clinicians and Commissioners events todiscuss the implications of the research

DiscussionThe design will take a rigorous scientific approach byutilising a three-site two-group randomised controlleddesign comparing two active parent group interventionsThis study will evaluate the clinical and cost-effectiveness of the Managing Repetitive Behaviours(MRB) parent group intervention compared with theLearning About Autism psychoeducation parent groupThe Learning About Autism (LAA) parent group deliv-ered by staff trained by The National Autistic Societyand equivalent to best current practice allows us to con-trol for the non-specific social group benefit of mutualsharing of experience and support between parentsParent group based interventions provide opportun-

ities for mutual learning and sharing of ideas allowingparents to discuss how best to support their childrsquosdevelopment This fosters opportunities for parents tolearn from and support each other in turn building par-entsrsquo knowledge and confidence to support their childrsquosneeds However in our MRB parent group interventionwe have included targeted support on challenging re-stricted and repetitive behaviours that are interfering ina deleterious way for the child There is some informa-tion on restricted and repetitive behaviours in the LAAparent group but MRB focuses more on assisting parentsto understand the role that RRB may have in their autis-tic childrsquos life In this way we anticipate that MRB willhave a greater beneficial effect on reducing identifiedchallenging RRB and improving child overall wellbeingMoreover we will evaluate whether providing parents

with the skills to effectively understand their childrsquos chal-lenging RRB has a greater effect on parental wellbeingsense of competence reducing stress and improvingfamily cohesion than a general autism psychoeducationparent group In this way the results from this RCT

Grahame et al Trials (2021) 22240 Page 9 of 13

have the potential to help autism researchers explore theactive ingredients of clinical trialsWe have carefully considered in collaboration with par-

ents and professionals the utility of the outcome measureschosen to assess the effects of the intervention on challen-ging RRB overall child functioning and parent and familywellbeing We have taken a rigorous approach to outcomeassessment with independent objective measures wherepossible We are aware of limitations in autism researchfrom over reliance on unblinded parental report measuresand have therefore included other outcome measures suchas teacher-rated outcomesIf found to be effective and efficient in the proposed

evaluation study this early parent group based interventiontargeting challenging RRB has the potential to fill an identi-fied but unmet need and thus improve the wellbeing ofautistic children and their families reduce parental stressgreatly enhance community participation increase learningopportunities and improve longer-term outcomes

COVID-19 addendumFollowing the announcement from the government on23rd March 2020 in response to the COVID-19 pandemica strict UK lockdown was imposed with implications forall clinical services and research trials across the UK Forour Managing Repetitive Behaviours (MRB) randomisedcontrolled trial (RCT) this has meant a great deal of in-novative creative thinking to design a revised researchprotocol for the conduct of all aspects of the research andthe delivery of the two parent group based interventionsThe aim was to ensure that recruited families and all theclinical and research staff were safe and that all our proce-dures were compliant with national government guidancedocuments and the required health and safety recommen-dations and procedures from the National Institute forHealth Research (NIHR) and the Department of HealthWith support from the research sponsor and funder anagreed plan to do assessments remotely and run parentgroups online was madeA key priority for our clinical research team was to

continue to keep the study open to maintain as best aswe could the integrity of the study and retain recruitedfamilies to the end of the trial despite the restrictionsimposed by the COVID-19 pandemic We are aware thatfamilies of autistic children are facing additional chal-lenges in light of COVID-19 with reduced access to sup-port and services This was an added reason to remainin contact with families recruited to our study We heardfrom parents that many children are finding changes totheir usual routines very challenging and that there wasa disproportionate impact on families caring for autisticchildren This is important information that may impacton the conduct and results of this RCT For all thesereasons we were very keen to keep in touch with our

recruited families for the duration of these challengingtimes However we also needed to take proactive pre-cautions to protect the health of all the families in thetrial and our staff in line with recommendations fromthe government NHS and NIHR and comply with localsite restrictions which varied between sites The trialteam have worked with all sites sponsor and funder toensure the most appropriate actions have been takenThe Trial Management Group with sponsor support on

26th March 2020 agreed all study activities would beperformed remotely by sites wherever possible Followingthis we implemented non-substantial amendment 08(NS08) on 3rd April 2020 as sponsor category C to reflectremote working changes with protocol version 06

COVID-19 remote working measures

1 Study Research Associates (RA) are now workingremotely at home in line with local policiesfollowing government advice and restrictionsInvestigative site files have been re-located toclinical sites wherever possible however access tothese and any paper source documents is noweither limited or not possible at this timea RAs working remotely at home must have

access to secure laptops which connect to theUniversityNHS server as appropriate Phonecalls to parents can be made at home but nopatient identifiable data can be stored in homes

b All new paper source documents including anyRA notes will either be recorded electronicallyfor this period of time so that no patient detailsare stored outside of the local site server orwritten on paper copies with no participantIdentifiable data (PID) where applicable

c Remote supervision of RAs to be arranged ateach site with local principle investigator (PI)oversight and documented

2 Ongoing review of the situation and ability toremain open to any study activitiesa Sites to inform Chief Investigator (CI) and

Newcastle Clinical Trial Unit (NCTU) of anychanges in availability of staffing at each site

b Study recruitment should be halted at individualsite level if sites are no longer able to deliver thestudy An amendment will only be submitted ifthe study is halted at sponsor level as discussedwith the CI and NCTU

c To consider availability of the CI to providecentral study advice

d The NCTU will inform sites of any overall studyhalt It is then the responsibility of the localclinical team to risk assess participants that arealready in the study

Grahame et al Trials (2021) 22240 Page 10 of 13

A suite of documents including a contingency plan andworking instruction have been created to aid all sites inthe day to day workings of the study and remote workingprocedures A sponsor specific contingency document hasalso been created to ensure sponsor is kept up to date withall the changes put in place Furthermore any futurechanges to the study relating to COVID-19 will bereviewed in line with the government guidance updates

Intervention parent groupsWe have explored digital options with our NHS clinicalcolleagues at each of the three sites to identify a secureand reliable platform that allows us to deliver our parentgroups online It was important the technology was easyto use for parents and group leaders but still allowedgroup leaders to present information and participate ingroup chat with familiesFor some families attending online parent groups has

been an easier option for them to consider as it has re-moved some of the barriers such as geographical trans-port work or other commitments that had previouslyprevented them from attending a parent group We aremindful of issues of digital poverty but initial feedbackfrom families has been positive and most families havebeen able to join a parent group using a range of devicessuch as smart phones tablets or computer

Trial statusProtocol Version 60 approval date 02042020The trial is recruiting Recruitment began on the

following dates for each siteCNTW 02102018 Lothian 20112018 TEWV 22

112018Recruitment is due to complete on the 31082020

AbbreviationsAE Adverse event AR Adverse reaction ASD Autism spectrum disorderCHU9D Child Health Utility 9D CI Chief Investigator eCRF Electronic CaseReport Form EoI Expression of interest DMC Data Monitoring CommitteeGCP Good Clinical Practice GP General practice MRB Managing RepetitiveBehaviours NCTU Newcastle Clinical Trials Unit NIHR-HTA National Institutefor Health Research ndash Health Technology Assessment NHS National HealthService PI Principal Investigator RampD Research amp Development RA ResearchAssociate RCT Randomised control trial REC Research Ethics CommitteeRfPB Research for Patient Benefit RRB Restrictive and repetitive behavioursQALY Quality adjusted life year SAE Serious adverse event SAR Seriousadverse reaction SOP Standard operating procedure USAR Unexpectedserious adverse reaction TSC Trial Steering Committee

Supplementary InformationThe online version contains supplementary material available at httpsdoiorg101186s13063-021-05175-y

Additional file 1 SPIRIT Checklist

Additional file 2 Ethical approval documentation HRA ethical approval

Additional file 3 Copy of original funding documentation NIHRfunding documentation

AcknowledgementsWe are extremely grateful to all the parents of autistic children who haveover the past 10 years actively collaborated with us to improve theintervention and research design We also gratefully acknowledge the valuedcontribution of all the participating families in the study and referringclinicians who have helped us with recruitment It really has been a teameffort and without everyonersquos support we would not have been able toundertake this large fully powered multi-site research trialWe also thank all the MRB group leaders Teresa Ingram2 Alan Galvin1 KirstinFarquhar1 Laura Tavernor2 Aimee Corner2 Rosalind Oliphant2 JakeHutchinson3 Emily Thompson2 Hannah Tait3 Courtnie Beasley-Gibson3 LukeRodgers3 Charlotte Randall3 Katie McIntyre3 Claire Cadger3 Jasmine Miller9Calum Parker7 Sinead OprimeCarroll6 Fiona McBryde10 Eleni Alexandropoulou6Hilary Cowie5 Callum MacKinnon8We would like to acknowledge the valued contribution of The NationalAutistic Society Training and Consultancy Team for supporting andorganising the training of the Learning About Autism parent group leadersIn particular David Scowcroft Family Seminar and Support Manager fordeveloping the bespoke training materials for the studyLAA group leaders David Sanderson2 Lisa Slater2 Rachel Lisle2 Maeve Ryan6Marisa McKinlay5 Sheila Kernohan5 Francesca Smart8We thank the members of the Trial Steering Committee Professor PatriciaHowlin Emiritus Professor of Clinical Child Psychology (Chair) ProfessorRichard Emsley Professor of Medical Statistics and Trials Professor HelenMason Professor of Health Economics and our parent representatives MsNicola King and Ms Helen BlackWe also thank the members of the Data Management and EthicsCommittee Professor John Jerrim Professor of Education and Social Statistics(Chair) Dr Kathy Leadbitter Senior Research Associate and Dr AnnOzsivadjian Principal Clinical PsychologistWe acknowledge the contribution to data collection and entry of DeborahJones1 and Ewan Lomax11 Newcastle University 2 Cumbria Tyne and Wear NHS Foundation Trust 3University of Durham 4 Tees Esk and Wear Valleys NHS Foundation Trust 5NHS Lothian 6 Edinburgh University 7 Taylor Ed-Foundation 8 Bouncet-Innovative Occupational Therapy 9 Jasmine Miller Coaching 10 City of Edin-burgh Council

Authorsrsquo contributionsVG JR LD ALC designed the study VG led the trial JR DR SFW researchleads and EH EIW LM ZG clinical leads at Newcastle Teesside Edinburghsites respectively LD lead trainer ASK led the statistical design LV amp AK ledthe economic evaluation MG JG PR LR ST HP SH EW RI researchassociates FW AM trial mangers CW RW database managers VG led thewriting of the paper All authors read contributed to and approved the finalmanuscript

FundingNIHRHTA ProgrammeUniversity of SouthamptonAlpha HouseEnterprise RoadSO16 7NSTelephone 023 8059 5586NIHR Research funding - pound137718055The funding body had no role in the design of the study data collectionanalysis interpretation or in the writing of this manuscript

Availability of data and materialsAll data in the trial will be anonymised A central master file will be held bythe trial manager at the Newcastle Clinical Trials Unit This will contain thekey linking anonymised trial name to personal details This eCRF pack will bebacked up securely within the web-based data entry service of NewcastleUniversity CTU All data will be entered into the Newcastle CTU web-basedsecure MACRO database which has a full audit trail and appropriate qualitycontrol will be carried out during the trial and before the database lockPrimary analysis of the data will take place in Durham University by the trialstatistician The datasets generated andor analysed during the current studyare available from the corresponding author on reasonable request andCOREC informed

Grahame et al Trials (2021) 22240 Page 11 of 13

have the potential to help autism researchers explore theactive ingredients of clinical trialsWe have carefully considered in collaboration with par-

ents and professionals the utility of the outcome measureschosen to assess the effects of the intervention on challen-ging RRB overall child functioning and parent and familywellbeing We have taken a rigorous approach to outcomeassessment with independent objective measures wherepossible We are aware of limitations in autism researchfrom over reliance on unblinded parental report measuresand have therefore included other outcome measures suchas teacher-rated outcomesIf found to be effective and efficient in the proposed

evaluation study this early parent group based interventiontargeting challenging RRB has the potential to fill an identi-fied but unmet need and thus improve the wellbeing ofautistic children and their families reduce parental stressgreatly enhance community participation increase learningopportunities and improve longer-term outcomes

COVID-19 addendumFollowing the announcement from the government on23rd March 2020 in response to the COVID-19 pandemica strict UK lockdown was imposed with implications forall clinical services and research trials across the UK Forour Managing Repetitive Behaviours (MRB) randomisedcontrolled trial (RCT) this has meant a great deal of in-novative creative thinking to design a revised researchprotocol for the conduct of all aspects of the research andthe delivery of the two parent group based interventionsThe aim was to ensure that recruited families and all theclinical and research staff were safe and that all our proce-dures were compliant with national government guidancedocuments and the required health and safety recommen-dations and procedures from the National Institute forHealth Research (NIHR) and the Department of HealthWith support from the research sponsor and funder anagreed plan to do assessments remotely and run parentgroups online was madeA key priority for our clinical research team was to

continue to keep the study open to maintain as best aswe could the integrity of the study and retain recruitedfamilies to the end of the trial despite the restrictionsimposed by the COVID-19 pandemic We are aware thatfamilies of autistic children are facing additional chal-lenges in light of COVID-19 with reduced access to sup-port and services This was an added reason to remainin contact with families recruited to our study We heardfrom parents that many children are finding changes totheir usual routines very challenging and that there wasa disproportionate impact on families caring for autisticchildren This is important information that may impacton the conduct and results of this RCT For all thesereasons we were very keen to keep in touch with our

recruited families for the duration of these challengingtimes However we also needed to take proactive pre-cautions to protect the health of all the families in thetrial and our staff in line with recommendations fromthe government NHS and NIHR and comply with localsite restrictions which varied between sites The trialteam have worked with all sites sponsor and funder toensure the most appropriate actions have been takenThe Trial Management Group with sponsor support on

26th March 2020 agreed all study activities would beperformed remotely by sites wherever possible Followingthis we implemented non-substantial amendment 08(NS08) on 3rd April 2020 as sponsor category C to reflectremote working changes with protocol version 06

COVID-19 remote working measures

1 Study Research Associates (RA) are now workingremotely at home in line with local policiesfollowing government advice and restrictionsInvestigative site files have been re-located toclinical sites wherever possible however access tothese and any paper source documents is noweither limited or not possible at this timea RAs working remotely at home must have

access to secure laptops which connect to theUniversityNHS server as appropriate Phonecalls to parents can be made at home but nopatient identifiable data can be stored in homes

b All new paper source documents including anyRA notes will either be recorded electronicallyfor this period of time so that no patient detailsare stored outside of the local site server orwritten on paper copies with no participantIdentifiable data (PID) where applicable

c Remote supervision of RAs to be arranged ateach site with local principle investigator (PI)oversight and documented

2 Ongoing review of the situation and ability toremain open to any study activitiesa Sites to inform Chief Investigator (CI) and

Newcastle Clinical Trial Unit (NCTU) of anychanges in availability of staffing at each site

b Study recruitment should be halted at individualsite level if sites are no longer able to deliver thestudy An amendment will only be submitted ifthe study is halted at sponsor level as discussedwith the CI and NCTU

c To consider availability of the CI to providecentral study advice

d The NCTU will inform sites of any overall studyhalt It is then the responsibility of the localclinical team to risk assess participants that arealready in the study

Grahame et al Trials (2021) 22240 Page 10 of 13

A suite of documents including a contingency plan andworking instruction have been created to aid all sites inthe day to day workings of the study and remote workingprocedures A sponsor specific contingency document hasalso been created to ensure sponsor is kept up to date withall the changes put in place Furthermore any futurechanges to the study relating to COVID-19 will bereviewed in line with the government guidance updates

Intervention parent groupsWe have explored digital options with our NHS clinicalcolleagues at each of the three sites to identify a secureand reliable platform that allows us to deliver our parentgroups online It was important the technology was easyto use for parents and group leaders but still allowedgroup leaders to present information and participate ingroup chat with familiesFor some families attending online parent groups has

been an easier option for them to consider as it has re-moved some of the barriers such as geographical trans-port work or other commitments that had previouslyprevented them from attending a parent group We aremindful of issues of digital poverty but initial feedbackfrom families has been positive and most families havebeen able to join a parent group using a range of devicessuch as smart phones tablets or computer

Trial statusProtocol Version 60 approval date 02042020The trial is recruiting Recruitment began on the

following dates for each siteCNTW 02102018 Lothian 20112018 TEWV 22

112018Recruitment is due to complete on the 31082020

AbbreviationsAE Adverse event AR Adverse reaction ASD Autism spectrum disorderCHU9D Child Health Utility 9D CI Chief Investigator eCRF Electronic CaseReport Form EoI Expression of interest DMC Data Monitoring CommitteeGCP Good Clinical Practice GP General practice MRB Managing RepetitiveBehaviours NCTU Newcastle Clinical Trials Unit NIHR-HTA National Institutefor Health Research ndash Health Technology Assessment NHS National HealthService PI Principal Investigator RampD Research amp Development RA ResearchAssociate RCT Randomised control trial REC Research Ethics CommitteeRfPB Research for Patient Benefit RRB Restrictive and repetitive behavioursQALY Quality adjusted life year SAE Serious adverse event SAR Seriousadverse reaction SOP Standard operating procedure USAR Unexpectedserious adverse reaction TSC Trial Steering Committee

Supplementary InformationThe online version contains supplementary material available at httpsdoiorg101186s13063-021-05175-y

Additional file 1 SPIRIT Checklist

Additional file 2 Ethical approval documentation HRA ethical approval

Additional file 3 Copy of original funding documentation NIHRfunding documentation

AcknowledgementsWe are extremely grateful to all the parents of autistic children who haveover the past 10 years actively collaborated with us to improve theintervention and research design We also gratefully acknowledge the valuedcontribution of all the participating families in the study and referringclinicians who have helped us with recruitment It really has been a teameffort and without everyonersquos support we would not have been able toundertake this large fully powered multi-site research trialWe also thank all the MRB group leaders Teresa Ingram2 Alan Galvin1 KirstinFarquhar1 Laura Tavernor2 Aimee Corner2 Rosalind Oliphant2 JakeHutchinson3 Emily Thompson2 Hannah Tait3 Courtnie Beasley-Gibson3 LukeRodgers3 Charlotte Randall3 Katie McIntyre3 Claire Cadger3 Jasmine Miller9Calum Parker7 Sinead OprimeCarroll6 Fiona McBryde10 Eleni Alexandropoulou6Hilary Cowie5 Callum MacKinnon8We would like to acknowledge the valued contribution of The NationalAutistic Society Training and Consultancy Team for supporting andorganising the training of the Learning About Autism parent group leadersIn particular David Scowcroft Family Seminar and Support Manager fordeveloping the bespoke training materials for the studyLAA group leaders David Sanderson2 Lisa Slater2 Rachel Lisle2 Maeve Ryan6Marisa McKinlay5 Sheila Kernohan5 Francesca Smart8We thank the members of the Trial Steering Committee Professor PatriciaHowlin Emiritus Professor of Clinical Child Psychology (Chair) ProfessorRichard Emsley Professor of Medical Statistics and Trials Professor HelenMason Professor of Health Economics and our parent representatives MsNicola King and Ms Helen BlackWe also thank the members of the Data Management and EthicsCommittee Professor John Jerrim Professor of Education and Social Statistics(Chair) Dr Kathy Leadbitter Senior Research Associate and Dr AnnOzsivadjian Principal Clinical PsychologistWe acknowledge the contribution to data collection and entry of DeborahJones1 and Ewan Lomax11 Newcastle University 2 Cumbria Tyne and Wear NHS Foundation Trust 3University of Durham 4 Tees Esk and Wear Valleys NHS Foundation Trust 5NHS Lothian 6 Edinburgh University 7 Taylor Ed-Foundation 8 Bouncet-Innovative Occupational Therapy 9 Jasmine Miller Coaching 10 City of Edin-burgh Council

Authorsrsquo contributionsVG JR LD ALC designed the study VG led the trial JR DR SFW researchleads and EH EIW LM ZG clinical leads at Newcastle Teesside Edinburghsites respectively LD lead trainer ASK led the statistical design LV amp AK ledthe economic evaluation MG JG PR LR ST HP SH EW RI researchassociates FW AM trial mangers CW RW database managers VG led thewriting of the paper All authors read contributed to and approved the finalmanuscript

FundingNIHRHTA ProgrammeUniversity of SouthamptonAlpha HouseEnterprise RoadSO16 7NSTelephone 023 8059 5586NIHR Research funding - pound137718055The funding body had no role in the design of the study data collectionanalysis interpretation or in the writing of this manuscript

Availability of data and materialsAll data in the trial will be anonymised A central master file will be held bythe trial manager at the Newcastle Clinical Trials Unit This will contain thekey linking anonymised trial name to personal details This eCRF pack will bebacked up securely within the web-based data entry service of NewcastleUniversity CTU All data will be entered into the Newcastle CTU web-basedsecure MACRO database which has a full audit trail and appropriate qualitycontrol will be carried out during the trial and before the database lockPrimary analysis of the data will take place in Durham University by the trialstatistician The datasets generated andor analysed during the current studyare available from the corresponding author on reasonable request andCOREC informed

Grahame et al Trials (2021) 22240 Page 11 of 13

A suite of documents including a contingency plan andworking instruction have been created to aid all sites inthe day to day workings of the study and remote workingprocedures A sponsor specific contingency document hasalso been created to ensure sponsor is kept up to date withall the changes put in place Furthermore any futurechanges to the study relating to COVID-19 will bereviewed in line with the government guidance updates

Intervention parent groupsWe have explored digital options with our NHS clinicalcolleagues at each of the three sites to identify a secureand reliable platform that allows us to deliver our parentgroups online It was important the technology was easyto use for parents and group leaders but still allowedgroup leaders to present information and participate ingroup chat with familiesFor some families attending online parent groups has

been an easier option for them to consider as it has re-moved some of the barriers such as geographical trans-port work or other commitments that had previouslyprevented them from attending a parent group We aremindful of issues of digital poverty but initial feedbackfrom families has been positive and most families havebeen able to join a parent group using a range of devicessuch as smart phones tablets or computer

Trial statusProtocol Version 60 approval date 02042020The trial is recruiting Recruitment began on the

following dates for each siteCNTW 02102018 Lothian 20112018 TEWV 22

112018Recruitment is due to complete on the 31082020

AbbreviationsAE Adverse event AR Adverse reaction ASD Autism spectrum disorderCHU9D Child Health Utility 9D CI Chief Investigator eCRF Electronic CaseReport Form EoI Expression of interest DMC Data Monitoring CommitteeGCP Good Clinical Practice GP General practice MRB Managing RepetitiveBehaviours NCTU Newcastle Clinical Trials Unit NIHR-HTA National Institutefor Health Research ndash Health Technology Assessment NHS National HealthService PI Principal Investigator RampD Research amp Development RA ResearchAssociate RCT Randomised control trial REC Research Ethics CommitteeRfPB Research for Patient Benefit RRB Restrictive and repetitive behavioursQALY Quality adjusted life year SAE Serious adverse event SAR Seriousadverse reaction SOP Standard operating procedure USAR Unexpectedserious adverse reaction TSC Trial Steering Committee

Supplementary InformationThe online version contains supplementary material available at httpsdoiorg101186s13063-021-05175-y

Additional file 1 SPIRIT Checklist

Additional file 2 Ethical approval documentation HRA ethical approval

Additional file 3 Copy of original funding documentation NIHRfunding documentation

AcknowledgementsWe are extremely grateful to all the parents of autistic children who haveover the past 10 years actively collaborated with us to improve theintervention and research design We also gratefully acknowledge the valuedcontribution of all the participating families in the study and referringclinicians who have helped us with recruitment It really has been a teameffort and without everyonersquos support we would not have been able toundertake this large fully powered multi-site research trialWe also thank all the MRB group leaders Teresa Ingram2 Alan Galvin1 KirstinFarquhar1 Laura Tavernor2 Aimee Corner2 Rosalind Oliphant2 JakeHutchinson3 Emily Thompson2 Hannah Tait3 Courtnie Beasley-Gibson3 LukeRodgers3 Charlotte Randall3 Katie McIntyre3 Claire Cadger3 Jasmine Miller9Calum Parker7 Sinead OprimeCarroll6 Fiona McBryde10 Eleni Alexandropoulou6Hilary Cowie5 Callum MacKinnon8We would like to acknowledge the valued contribution of The NationalAutistic Society Training and Consultancy Team for supporting andorganising the training of the Learning About Autism parent group leadersIn particular David Scowcroft Family Seminar and Support Manager fordeveloping the bespoke training materials for the studyLAA group leaders David Sanderson2 Lisa Slater2 Rachel Lisle2 Maeve Ryan6Marisa McKinlay5 Sheila Kernohan5 Francesca Smart8We thank the members of the Trial Steering Committee Professor PatriciaHowlin Emiritus Professor of Clinical Child Psychology (Chair) ProfessorRichard Emsley Professor of Medical Statistics and Trials Professor HelenMason Professor of Health Economics and our parent representatives MsNicola King and Ms Helen BlackWe also thank the members of the Data Management and EthicsCommittee Professor John Jerrim Professor of Education and Social Statistics(Chair) Dr Kathy Leadbitter Senior Research Associate and Dr AnnOzsivadjian Principal Clinical PsychologistWe acknowledge the contribution to data collection and entry of DeborahJones1 and Ewan Lomax11 Newcastle University 2 Cumbria Tyne and Wear NHS Foundation Trust 3University of Durham 4 Tees Esk and Wear Valleys NHS Foundation Trust 5NHS Lothian 6 Edinburgh University 7 Taylor Ed-Foundation 8 Bouncet-Innovative Occupational Therapy 9 Jasmine Miller Coaching 10 City of Edin-burgh Council

Authorsrsquo contributionsVG JR LD ALC designed the study VG led the trial JR DR SFW researchleads and EH EIW LM ZG clinical leads at Newcastle Teesside Edinburghsites respectively LD lead trainer ASK led the statistical design LV amp AK ledthe economic evaluation MG JG PR LR ST HP SH EW RI researchassociates FW AM trial mangers CW RW database managers VG led thewriting of the paper All authors read contributed to and approved the finalmanuscript

FundingNIHRHTA ProgrammeUniversity of SouthamptonAlpha HouseEnterprise RoadSO16 7NSTelephone 023 8059 5586NIHR Research funding - pound137718055The funding body had no role in the design of the study data collectionanalysis interpretation or in the writing of this manuscript

Availability of data and materialsAll data in the trial will be anonymised A central master file will be held bythe trial manager at the Newcastle Clinical Trials Unit This will contain thekey linking anonymised trial name to personal details This eCRF pack will bebacked up securely within the web-based data entry service of NewcastleUniversity CTU All data will be entered into the Newcastle CTU web-basedsecure MACRO database which has a full audit trail and appropriate qualitycontrol will be carried out during the trial and before the database lockPrimary analysis of the data will take place in Durham University by the trialstatistician The datasets generated andor analysed during the current studyare available from the corresponding author on reasonable request andCOREC informed

Grahame et al Trials (2021) 22240 Page 11 of 13

Declarations

Ethics approval and consent to participateThis study received ethical approval from the South West - Cornwall ampPlymouth Research Ethics Committee on the 200818 (ref 18SW0173)Informed consent will be obtained from all study participants

Consent for publicationNot applicable

Competing interestsProfessor Luke Vale was a panel member of the NIHR HTA Clinical Trials andEvaluation Panel until March 2018 There are no other competing interests

Author details1Cumbria Northumberland Tyne and Wear NHS Foundation Trust ComplexNeurodevelopmental Disorder Service (CNDS) Walkergate Park BenfieldRoad Newcastle upon Tyne NE6 4QD UK 2Salvesen Mindroom ResearchCentre The University of Edinburgh Kennedy Tower Morningside TerraceEdinburgh EH10 5HF UK 3National Autistic Society North East ResourceCentre Newcastle upon Tyne NE5 2LW UK 4Population Health SciencesInstitute Newcastle University Level 3 Sir James Spence Institute RoyalVictoria Infirmary Queen Victoria Road Newcastle upon Tyne NE1 4LP UK5Salvesen Mindroom Research Centre Child Life and Health Royal Hospitalfor Sick Children Edinburgh EH9 1LF UK 6Royal Edinburgh HospitalKennedy Tower Morningside Terrace Edinburgh EH10 5HF UK 7DurhamResearch Methods Centre Durham University South Road Durham DH1 3LEUK 8Population Health Sciences Institute Newcastle UniversityBaddiley-Clark Building Richardson Road Newcastle upon Tyne NE2 4AA UK9Department of Anthropology Durham University South Rd Durham DH13LE UK 10Newcastle Clinical Trials Unit Newcastle University 1-4 ClaremontTerrace Newcastle upon Tyne NE2 4AE UK 11Psychology DepartmentDurham University Science Laboratories South Road Durham DH1 3LE UK12Derwentside CAMHS 192 Medomsley Road Consett DH8 5HT UK

Received 15 October 2020 Accepted 8 March 2021

References1 Elsabbagh M Divan G Koh YJ Kim YS Kauchali S Marciacuten C Montiel-Nava C

Patel V Paula CS Wang C Yasamy MT Fombonne E Global prevalence ofautism and other pervasive developmental disorders Autism Res 20125(3)160ndash79 httpsdoiorg101002aur239

2 Baird G Simonoff E Pickles A Chandler S Loucas T Meldrum DCharman T Prevalence of disorders of the autism spectrum in apopulation cohort of children in South Thames the Special Needs andAutism Project (SNAP) Lancet 2006368(9531)210ndash5 httpsdoiorg101016S0140-6736(06)69041-7

3 Diagnostic and Statistical Manual of Mental Disorders DSM-5 5th edArlington American Psychiatric Association 2013 DSM-V httpsdoiorg101176appi

4 Manor-Binyamini I Schreiber-Divon M Repetitive behaviors listening tothe voice of people with high-functioning autism spectrum disorderRes Autism Spectr Disord 20196423ndash30 httpsdoiorg101016jrasd201904001

5 Bakan MB The musicality of stimming promoting neurodiversity in theethnomusicology of autism MC [Internet] 2015May22 [cited 2021Mar1]41(2) Available from httpsjournalslibunbcaindexphpMCarticleview22914

6 Greenberg JS Seltzer MM Hong J Orsmond GI Bidirectional effects ofexpressed emotion and behavior problems and symptoms in adolescentsand adults with autism Am J Ment Retard 2006111(4)229ndash49 httpsdoiorg1013520895-8017(2006)111[229BEOEEA]20CO2

7 Leadbitter K Aldred C McConachie H Le Couteur A Kapadia D Charman Tet al The autism family experience questionnaire (AFEQ) an ecologically-valid parent-nominated measure of family experience quality of life andprioritised outcomes for early intervention J Autism Dev Disord 201848(4)1052ndash62 httpsdoiorg101007s10803-017-3350-7

8 South M Ozonoff S McMahon WM Repetitive behavior profiles in Aspergersyndrome and high-functioning autism J Autism Dev Disord 200535(2)145ndash58 httpsdoiorg101007s10803-004-1992-8

9 Hayes SA Watson SL The impact of parenting stress a meta-analysis ofstudies comparing the experience of parenting stress in parents of childrenwith and without autism spectrum disorder J Autism Dev Disord 201343(3)629ndash42 httpsdoiorg101007s10803-012-1604-y

10 Leekam SR Prior MR Uljarevic M Restricted and repetitive behaviors inautism spectrum disorders a review of research in the last decade PsycholBull 2011137(4)562ndash93 httpsdoiorg101037a0023341

11 Grossman RB Edelson LR Tager-Flusberg H Emotional facial and vocalexpressions during story retelling by children and adolescents with high-functioning autism J Speech Language Hearing Res 201356(3)1035ndash44httpsdoiorg1010441092-4388(201212-0067)

12 Lampi A Fitzpatrick P Romero V Amaral J Schmidt RC Understanding theinfluence of social and motor context on the co-occurring frequency ofrestricted and repetitive behaviors in autism J Autism Dev Disord 202050(5)1479ndash96 httpsdoiorg101007s10803-018-3698-3

13 Coon JC Rapp JT Brief report evaluating college studentsrsquo perceptions of achild displaying stereotypic behaviors Do changes in stereotypy levelsaffect ratings J Autism Dev Disord 202050(5)1827ndash33 httpsdoiorg101007s10803-019-03916-2

14 Griffin WB Peer perceptions of students with autism spectrum disordersFocus on Autism and Other Developmental Disabilities 201934(3)183ndash92httpsdoiorg1011771088357618800035

15 Cunningham AB Schreibman L Stereotypy in autism the importance offunction Res Autism Spectr Disord 20082(3)469ndash79 httpsdoiorg101016jrasd200709006

16 Gordon C Commentary considerations on the pharmacological treatmentof compulsions and stereotypies with serotonin reuptake inhibitors inpervasive developmental disorders J Autism Dev Disord 200030(5)437ndash8httpsdoiorg101023A1005503607728

17 Hodgson AR Grahame V Garland D Gaultier F Lecouturier J Le Couteur AParentsrsquo opinions about an intervention to manage repetitive behaviours inyoung children with autism spectrum disorder a qualitative study J ApplRes Intellect Disabil 201831(S2)165ndash78 httpsdoiorg101111jar12317

18 Health Do Education Df Skills National Service Framework for ChildrenYoung People and Maternity Services Standard 8 Disabled Children andYoung People and those with Complex Health Needs 2004 httpswwwgovukgovernmentpublicationsnational-service-framework-children-young-people-and-maternity-services Accessed 21 May 2019

19 Oono IP Honey EJ McConachie H Parent-mediated early interventionfor young children with autism spectrum disorders (ASD) Evid BasedChild Health Cochrane Rev J 20138(6)2380ndash479 httpsdoiorg101002ebch1952

20 Green J Charman T McConachie H Aldred C Slonims V Howlin P leCouteur A Leadbitter K Hudry K Byford S Barrett B Temple K MacdonaldW Pickles A PACT Consortium Parent-mediated communication-focusedtreatment in children with autism (PACT) a randomised controlled trialLancet 2010375(9732)2152ndash60 httpsdoiorg101016S0140-6736(10)60587-9

21 Harrop C Evidence-based parent-mediated interventions for youngchildren with autism spectrum disorder the case of restricted andrepetitive behaviors Autism 201519(6)662ndash72 httpsdoiorg1011771362361314545685

22 Wallace S Parr J Hardy A One in a hundred putting families at the heart ofautism research Autistica 2013 httpswwwautisticaorguk Accessed 21May 2019

23 Pellicano L Dinsmore A Charman T A future made together shapingautism research in the UK 2013 httpsdiscoveryuclacukideprint14955831A_Future_Made_Together_12_LRpdf Accessed 21 May 2019

24 Preacutefontaine I Lanovaz MJ McDuff E McHugh C Cook JL Using mobiletechnology to reduce engagement in stereotypy a validation of decision-making algorithms Behav Modif 201943(2)222ndash45 httpsdoiorg1011770145445517748560

25 Grahame V Brett D Dixon L McConachie H Lowry J Rodgers J Steen N leCouteur A Managing repetitive behaviours in young children with autismspectrum disorder (ASD) pilot randomised controlled trial of a new parentgroup intervention J Autism Dev Disord 201545(10)3168ndash82 httpsdoiorg101007s10803-015-2474-x

26 Davidson J Children and young people in mind the final report of theNational CAMHS Review London Department of Health 2008 httpswwwbaswcouksystemfilesresourcesbasw_31351-4_0pdf Accessed 21May 2019

Grahame et al Trials (2021) 22240 Page 12 of 13

27 Lord C Rutter M DiLavore P Risi S Gotham K Bishop S Autism DiagnosticObservation Schedule Second Edition (ADOS-2) Manual (Part 1) Modules1ndash4 Torrence CA Western Psychological Services 2012 httpswwwwpspublishcomados-2-autism-diagnostic-observation-schedule-second-edition Accessed 21 May 2019

28 Zander E Willfors C Berggren S Choque-Olsson N Coco C Elmund AMoretti AringH Holm A Jifaumllt I Kosieradzki R Linder J Nordin V OlafsdottirK Poltrago L Boumllte S The objectivity of the Autism DiagnosticObservation Schedule (ADOS) in naturalistic clinical settings Eur ChildAdolescent Psychiatry 201625(7)769ndash80 httpsdoiorg101007s00787-015-0793-2

29 Constantino J Gruber C Social Responsiveness Scale - SecondEdition (SRS-2) Torrence CA Western Psychological Services 2012httpswwwwpspublishcomsrs-social-responsiveness-scale Accessed21 May 2019

30 Bruni TP Test Review Social Responsiveness ScalendashSecond Edition (SRS-2) J Psychoeduc Assess 201432(4)365ndash9 httpsdoiorg1011770734282913517525

31 Guy W Editor ECDEU assessment manual for psychopharmacologyRockville MD US Department of Heath Education and Welfare PublicHealth Service Alcohol Drug Abuse and Mental HealthAdministration 1976

32 Forkmann T Scherer A Boecker M Pawelzik M Jostes R Gauggel S Theclinical global impression scale and the influence of patient or staffperspective on outcome BMC Psychiatry 201111(1)83 httpsdoiorg1011861471-244X-11-83

33 Arnold LE Vitiello B McDougle C Scahill L Shah B Gonzalez NM et alParent-defined target symptoms respond to risperidone in RUPP autismstudy customer approach to clinical trials J Am Acad Child AdolescPsychiatry 200342(12)1443ndash50 httpsdoiorg10109700004583-200312000-00011

34 Turner MA Repetitive behaviour and cognitive functioning in autism[dissertation] University of Cambridge 1996

35 Lidstone J Uljarević M Sullivan J Rodgers J McConachie H Freeston M leCouteur A Prior M Leekam S Relations among restricted and repetitivebehaviors anxiety and sensory features in children with autism spectrumdisorders Res Autism Spectr Disord 20148(2)82ndash92 httpsdoiorg101016jrasd201310001

36 Johnson N Rodgers J Grahame V Honey E Association betweenintolerance of uncertainty restricted and repetitive behaviours and anxietyin ASD [dissertation] Newcastle University 2017

37 Sparrow SS Cicchetti DV Saulnier CA Vineland Adaptive Behavior ScalesThird ed San Antonio TX Pearson 2016 httpswwwpearsonassessmentscomstoreusassessmentsenStoreProfessional-AssessmentsBehaviorAdaptiveVineland-Adaptive-Behavior-Scales-7C-Third-Editionp100001622html Accessed 21 May 2019

38 Stevens K Valuation of the child health utility 9D indexPharmacoeconomics 201230(8)729ndash47 httpsdoiorg10216511599120-000000000-00000

39 Furber G Segal L The validity of the Child Health Utility instrument(CHU9D) as a routine outcome measure for use in child and adolescentmental health services Health Qual Life Outcomes 201513(1)22 httpsdoiorg101186s12955-015-0218-4

40 Van Reenen M Janssen B Oppe M Kreimeier S Greiner W EQ-5D-Y userguide basic information on how to use the EQ-5D-Y instrument RotterdamEuroQoL Group 2014 httpseuroqolorgpublicationsuser-guidesAccessed 21 May 2019

41 Janssen MF Birnie E Haagsma JA Bonsel GJ Comparing the standard EQ-5D three-level system with a five-level version Value Health 200811(2)275ndash84 httpsdoiorg101111j1524-4733200700230x

42 Sofronoff K Farbotko M The effectiveness of parent managementtraining to increase self-efficacy in parents of children with Aspergersyndrome Autism 20026(3)271ndash86 httpsdoiorg1011771362361302006003005

43 Silva LM Schalock M Autism parenting stress index initial psychometricevidence J Autism Dev Disord 201242(4)566ndash74 httpsdoiorg101007s10803-011-1274-1

44 Tennant R Hiller L Fishwick R Platt S Joseph S Weich S Parkinson J SeckerJ Stewart-Brown S The Warwick-Edinburgh mental well-being scale(WEMWBS) development and UK validation Health Qual Life Outcomes20075(1)63 httpsdoiorg1011861477-7525-5-63

45 Xiao Z Kasim A Higgins S Same difference Understanding variation in theestimation of effect sizes from educational trials Int J Educ Res 2016771ndash14 httpsdoiorg101016jijer201602001

46 R Core Team R A language and environment for statistical computingVienna Austria R Foundation for Statistical Computing 2014 httpswwwr-projectorg Accessed 21 May 2019

Publisherrsquos NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations

Grahame et al Trials (2021) 22240 Page 13 of 13