Protocol for a Systematic Review: Interventions for Anxiety in School‐Aged Children with Autism Spectrum Disorder (ASD): A Mixed‐Methods Systematic Review

.


Description of interventions
There are numerous interventions currently available for the treatment of anxiety in children and young people. The focus of this review will be on interventions designed to help a child's functioning in real-world settings such as school and the home. Thus, studies assessing only the impact of pharmacological interventions will be excluded, while a study investigating the impact of cognitive behaviour therapy (CBT) on academic performance would be included. We note that research indicates CBT is useful for treating anxiety disorders, but less is known about its efficacy in treating anxiety with ASD populations (Nadeau et al., 2011).

Importance of this review
Since children spend a significant portion of their day at school, teachers and clinicians working in the education sector have significant responsibility for recognising signs of ASD and anxiety, and in implementing interventions and supports that are evidence-based and tailored to the needs of the child. Further, decision-making regarding treatment should be informed by the latest evidence available. However, the sheer volume of published research, and the different aims, foci and methodology of those studies, makes evidence-based practice extremely difficult for professionals, including for those working in the education sector. The current review will provide a much needed source of valuable information for various professionals.
Finally, a number of narrative and systematic reviews on various aspects of anxiety in ASD have been published in the last ten years. These reviews have covered phenomenology and prevalence of anxiety (White et al., 2009;MacNail, Lopez, & Minnes, 2009;van Steensel et al., 2010), assessment (Wigham, & McConachie, 2014;Lecavalier et al., 2014) and treatment (Johnco & Starch, 2o15;Kreslins, Robertson, & Melvile, 2015;Ung et al., 2015;Vasa et al., 2014;Sukhodolsky et al., 2013). However, none of the reviews published thus far have: (i) focused specifically on school-aged children with ASD; (ii) covered the range of available treatments, but instead focussed only on specific treatments, such as, for example, Cognitive Behaviour Therapy or psychosocial treatments; (iii) explored mediators and moderators of treatment outcomes; (iv) provided practical guidance for education professionals and parents to enable increased use of evidence-based treatments in their everyday practice.

OBJECTIVES
This review aims to synthesise evidence about interventions to reduce anxiety symptoms in school-aged children with ASD. While clinical studies will not be excluded per se, this review seeks to move beyond interventions that are relevant only for clinical practice and care in clinical settings and prioritise studies that draw out implications for school-aged children that will help their functioning in real-world settings such as school and the home. To achieve this aim, the review will employ a mixed-methods systematic review (JBI, 2014b) which can accommodate the anticipated diverse types of available studies. These studies are likely to use quantitative methods such as quasi-experimental, mixed-methods randomised control trial approaches as well as qualitative methods such as action-research and casestudy designs.
This systematic review aims to address the following research question.
1. What is the relative effectiveness of interventions for managing anxiety of schoolaged children with ASD that have been used in school, family, and clinical settings?
In the process, this review will also identify: • the interventions used for managing anxiety of school-aged children with ASD in school, family, and clinical settings; • the sources of variability in response to the intervention. For example, (i) what works for whom and how? (ii) under what circumstances are interventions efficacious or otherwise? (iii) are there sub-groups that the intervention impacts differentially (e.g., race, age, socio-economic status, academic achievement)? • the evidence-based practices that school staff, parents, and other professionals can employ to mitigate anxiety-related symptoms in school-aged children with ASD.
Results of the review are intended to inform professionals working in the education sector and parents, but may also inform policy makers in this sector.

METHODOLOGY
Given that the focus is on a) interventions that address anxiety and school-related functioning in school settings or family contexts and b) the intention for results of the review to be of direct relevance to policy makers and practitioners, it is anticipated that the studies relevant to this review will have employed various methodologies, not only quantitative but also qualitative methods. Therefore, a mixed-methods systematic review (JBI, 2014b) is proposed as the appropriate methodology for this work.
More specifically, the mixed-methods systematic review will follow the steps recommended by Sandelowski, Leeman, Knafl and Crandel (2012) with two separate syntheses conducted first, one resulting in a synthesis of quantitative studies and one resulting in a synthesis of qualitative studies. In the current review, the quantitative studies are proposed to be synthesised by way of meta-analysis while the qualitative studies are proposed to be synthesised by way of meta-aggregation (Lookwood, Munn & Porritt, 2015).
In the final step, the results of the two separate syntheses will be integrated by way of an aggregative mixed-method synthesis. In order to integrate the results of the two syntheses, the results of the quantitative synthesis are translated into qualitative statements through Bayesian conversion. This is considered preferable to the translation of qualitative into quantitative statements as problems arise when attempting to convert verbal counts (e.g. "many", "few") into quantities (JBI, 2014b). The mixed-methods approach to this systematic review is illustrated in Figure 1.

Research question
Study selection, data analysis Study selection, data analysis Quantitative synthesis (meta-analysis) Qualitative synthesis (meta-aggregation) Aggregative mixed-methods synthesis

Population
The target population for the review is school-aged children (5 to 18 years old) diagnosed with an ASD (inclusive of autism, ASD, Autistic Disorder, Asperger's Disorder, Asperger Syndrome, atypical autism, PDD-NOS) by a professional eligible to diagnose these conditions, and also experiencing anxiety symptoms or a diagnosis of an anxiety disorder provided by a professional eligible to diagnose such conditions. While a formal diagnosis of an ASD will be required for inclusion, anxiety symptoms will not have to be formally diagnosed. Therefore, we will include anxiety symptoms such as internalising behaviours and fear in the review. The studies can occur in schools or out-of-school settings (e.g. home, larger community) or clinical settings, as long as the intervention/s is designed to improve outcomes in real-world settings.
If studies include a sample of children in the target population as well as other children (e.g., the general population) and the findings are separated for the ASD sub-group, the study will be included in the review whereby the type of ASD diagnosed will not matter. In contrast, if the study findings are not reported separately -for example -for children with ASD and ADHD are combined for analysis, the study will be excluded from the review as the impact of the intervention on only the ASD sample will be impossible to derive.
No restrictions will be imposed in terms of background variables such as socio-economic status, or profiles of children with ASD -for example with respect to characteristics such as IQ or ASD severity/classification. However, where possible, the mediating or moderating influence of these variables on the treatment outcome(s) will be explored.
The focus is on school-aged children rather than earlier intervention since diagnosis of children with ASD who function at a relatively higher level often does not occur until primary school (Fombonne, 2003). To be included in the review, either all participants in a study have to be of school age or a majority of participants have to be of school age. -Borus et al. (2012) proposed that all existing interventions for anxiety incorporate one or more of the following seven elements: (i) psychoeducation, (ii) exposure, (iii) cognitive restructuring, (iv) parent training or parent psychoeducation, (v) relaxation, (vi) modelling, and (vii) self-monitoring. This review will include all treatments for anxiety for 5-18 year olds with ASD in schools, families or in clinical settings that encompass at least one of these elements. As such, studies which focus solely on pharmacological interventions (e.g., selective serotonin reuptake inhibitors) will be excluded from the review. Where studies occur in a clinical setting, we will seek to draw out implications for real world settings such as the family and school.

Rotheram
All forms of assessing children are eligible for inclusion in the review, including diagnostic interviews, behavioural and physiological monitoring, as well as ratings scales -irrespective of the informant (e.g., student, parent, teacher

Outcomes
The aim is to review interventions that address anxiety symptoms in school-aged children with a formal diagnosis of ASD in school, home and/or clinical settings. Measurement of anxiety (and related terms) should be undertaken using valid and reliable approaches such as screening instruments, observational ratings, and behavioural checklists. This review will include studies which refer to the variety of school-functioning related outcomes that are often associated with anxiety, including social or emotional skills in the school or family setting, school engagement, learning outcomes and/or academic achievement.

Types of Study Designs
Study designs to be included can be quantitative (e.g. descriptive, correlational, quasiexperimental and experimental) as well as qualitative (e.g. ethnographic, narrative, phenomenological, grounded theory and case study) in kind.
As studies employing a qualitative design are less likely to employ some form of valid and reliable measurement of anxiety than quantitative studies, more of the latter than the former study designs are expected to be included in the review. It should also be noted that while The following study could be included although only if it were possible to separate the data for school-aged children as some of the subjects are in pre-school. However, it illustrates the type of school-family collaborative treatments that will be considered for inclusion in the review.  It should be noted that established criteria will be used to assess the quality of the study design and resulting evidence which are appropriate to the design of the original primary study (e.g., Joanna Briggs Institute, 2014). Rather than omitting studies from the body of materials to be reviewed, the quality evaluation will be used to weight the findings from different studies.

Search Strategy for Identification of Relevant Studies
Our search strategy is designed to identify published as well as unpublished literature via bibliographic databases, grey literature sources, selected websites, web searching, research registers and by manually searching targeted journals and reference lists. We will also attempt to contact key researchers in the field in order to identify any additional ongoing or unpublished studies.

Publication Date Range
Unless otherwise specified, our searches will be limited to a publication date range of 1996 to 2017. We selected 1996 as the earliest publication date in order to narrow the scope of interventions to current approaches used in the last 20 years. Given the development of understanding in this field, we believe that interventions before this date would be less progressive in their approach.

Other Criteria
The searches in our selected sources will not be restricted by geography, language, publication type or by publication status. However the selected sources are focused on the English language in keeping with our database subscriptions and the primary language of the authors.

Search terms
The following PICO (i.e. population, intervention, comparison and outcomes) concepts will form the basis of the search strategy.

AND
Outcomes: Anxiety (e.g., anxiety, anxious, internalise, internalisation, fear) AND Intervention: (e.g., intervention, treatment, therapy, psychotherapy, evaluation, outcome, program, trials, experimental, control group, random, best practice, evidence based) It should be noted that we purposely have not specified a comparison as the review question is broad in nature. In such instances, the JBI (2013) suggest that a comparison statement may be inappropriate or undesirable, especially where multiple interventions may exist. In general, though, treatment as specified in the interventions will be compared to no treatment (e.g. waiting list).
Note that in the Intervention concept, we have included both intervention and study design terms. We were concerned that the use of the intervention concept may limit the search too narrowly particularly within the social, behavioural and educational databases. By including the alternative study design terms, we expect to maximise our chance of capturing the full range of various interventions that have been studied. However where this broad concept search results in an impractically large number of results and/or a significant reduction in the precision of the results, the search strategy will be limited to include the core, intervention terms: intervention, treatment, therapy, psychotherapy.
Our general search statement, set out below, will be customised to the available search strategies within the bibliographic databases and other sources that we search. Consideration will be given to available fields, limiters and subject indices as well as to other search features.
With the exception of Scopus database which does not have a controlled vocabulary, the subject indices for each of our bibliographic databases will be consulted in order to identify any additional, relevant search terms, beyond those already identified in our general search strategy. Various combinations of free text and/or controlled vocabulary terms will be used for each database. While trying to keep the search strategies as broad as possible to ensure relevant studies are not missed, the search strategy may need to be limited, to ensure a more manageable and precise set of results.
In searching the our sources beyond the bibliographic databases, the complexity of search features found in the databases may not always be available in the other sources and so simpler search strategies will be applied as needed.
This is our general search strategy that will guide our customised searches. All terms may not be included in every search statement: In this general search statement, the * symbol is used to indicate where our search will be designed to cover variations in the root of the word. This general statement assumes plural variations, requiring just the letter 's', will be automatically searched.

Bibliographic Databases for Searching
The following bibliographic databases will be electronically searched for studies that match our inclusion criteria:  Table 3 for customised search statements that will be applied to the bibliographic databases listed above.

Google
Google will be used to identify grey literature from websites in both government and organisation domains. Our general search statement will be customised to the Google Advanced Search. The searches will be limited to 'Show the most relevant results' and limited to the relevant date range. The search statements will be:

OpenGrey (European)
OpenGrey will be used to identify relevant European grey literature. The search statement will be: (ASD OR Asperger* OR autis* OR Pervasive Developmental Disorder* OR PDD NOS OR PDD unspecified) AND (Anxiety OR anxious OR internali* OR fear) AND (Student OR child* OR adolescen* OR preadolescen* OR pre adolescen* OR youth OR teen* OR teen age* OR young people OR young person OR boy OR girl)

Institutional Repositories
We will search the 'Contents' of the Directory of Open Access Repositories (OpenDOAR) to identify research papers from institutional repositories. Our general search statement will be customised to fit with the available search fields in the OpenDOAR Google Custom Search. Our search statement will be:

Theses
• Networked Digital Library of Theses and Dissertations -a customised search strategy will be used and the search will be limited by available population tags:

(Asperger OR autism) AND Anxiety AND intervention
• WorldCat -a customised search strategy will be used and the search will be limited by Thesis/dissertation:

Asperger OR autism anxiety intervention
• American Doctoral Dissertations (EBSCO) -our full search statement will be used.

Conference Proceedings
In addition to conference proceedings and papers indexed in our selected databases, we will identify conference literature via a search on SCOPUS which is a multidisciplinary database. This search will be limited to conference papers in the collections other than the Social Sciences, Humanities or Neuroscience as these collections will be completely searched as described in the section headed Bibliographic Databases for Searching -see Scopus.

Research Reviews
In designing the search for each of the following sources, consideration will be given to the size of the resource and whether the full search strategy is necessary or even possible to apply: • Campbell Library • Cochrane Central Register of Controlled Trials (CENTRAL) • The JBI Database of Systematic Reviews and Implementation Reports.
• Database of Promoting Health Effectiveness Reviews (DoPHER) • Evidence for Policy and Practice Information and Co-ordinating Centre (EPPI-Centre) • Cochrane Database of Systematic Reviews • PROSPERO International prospective register of systematic reviews

Targeted Searches of Selected Websites
We will browse or search the websites of selected agencies, research centres and professional associations including the following:

Citations Searching
We will conduct forward citation searching on the studies identified for analysis, using SCOPUS database.

Manual Searching
We will search the reference lists of previously published reviews and meta-analyses that we identify as well as the reference lists of each of the studies identified for our analysis.

Current literature
After our initial search, we will set up alerts in Google Scholar and, where possible, in the bibliographic databases, in order to identify any new literature within the time of our study. We will also manually scan the table of contents for the most current journal issues of our key journal titles.

Colleagues
We will identify and attempt to contact key researchers, seeking details of any further ongoing or unpublished studies.

Ongoing Trials
We will attempt to identify current and ongoing trials via this trial registry:

International Clinical Trials Registry Platform Search Portal
We will use this search statement: Anxiety in the 'Title' AND Asperger OR Autism in the 'Condition'.

Management of References
A full set of search results will be directly imported into an Endnote Library wherever possible or, if necessary, manually entered into the Endnote Library. The Endnote Library will provide a means to manage the full set of references and to identify duplicates.

Selection of studies
The PRISMA flow chart illustrates the process for the selection of studies (see Figure 1). Figure 1. Process of study selection (Source: Moher et al., 2009) As a first step in the screening process, all reviewers will independently assess titles and abstracts of a purposely heterogeneous subset of five studies identified through the searches. The purpose of this step is twofold: first it determines their potential eligibility for inclusion in the review and second it serves to develop a common understanding and application of inclusion criteria. Once a consensus regarding the application has been reached, all abstracts will be assessed by at least two reviewers. Where two reviewers disagree regarding the inclusion of an abstract in the study resolution will be sought through discussion or the decision of a third reviewer. At the end of this step, studies that clearly do not meet the criteria as well as duplicates will have been removed.

Prisma Flow Diagram
Full-text articles will then be retrieved for the included abstracts. Data extraction from the full-text articles will be undertaken independently by two reviewers. Extracted data will be compared between reviewers for seven articles. Any discrepancies will be resolved through discussion and, where necessary, further details added to data definitions which may be unclear. Data from a further two articles will be extracted to see whether agreement between reviewers has been reached.
Again, a selection of five articles will be critically appraised by all reviewers in order to ensure consistent application of criteria. As before, once consensus has been achieved, each article will be assessed independently by two reviewers for methodological validity using standardised critical appraisal instruments from the Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI) or the Qualitative Assessment and Review Instrument (JBI-QARI). See Table 1 for a list of checklists and Appendix 1 for examples of the checklists. Any disagreements that arise between the reviewers will be resolved through discussion, or with a third reviewer. For studies that might reasonably have been expected to be included but which did not meet the inclusion criteria, the specific reasons for exclusion will be documented.

Data extraction and management
Once papers are selected for inclusion in the review, data will be extracted using the standardised data extraction tools from JBI-MAStARI and (JBI-QARI). See Appendix 2 for the quantitative and qualitative data extraction forms. The data extracted from quantitative studies will include specific details about the interventions, populations, study methods and outcomes of significance to the review question and specific objectives. The data extracted from qualitative studies will include specific details about the phenomena of interest, populations, study methods and outcomes of significance to the review question and specific objectives.
While the forms provided in Appendix 2 are currently generic, information specific to this review -for example the ASD diagnosis or which components of Rotheram-Borus et al.'s (2012) framework for interventions were present -will be added after a discussion by the reviewer team prior to the start of the data extraction.
Where data are missing, authors will be contacted to obtain the relevant information. In some instances, it is possible to recreate missing data from available information. Where this is done, the assumptions and calculations will be detailed in the review report.
Only one outcome per study will be used in the separate quantitative and qualitative syntheses to avoid double counting. The primary outcome of a study will be chosen for data extraction. Where the primary outcome cannot be ascertained an outcome will be randomly chosen.

Data Synthesis and Analysis
Quantitative data will, where possible, be pooled by way of statistical meta-analysis using JBI-MAStARI. All results will be subject to double data entry. Effect sizes expressed as odds ratio (for categorical data) and weighted mean differences (for continuous data) and their 95% confidence intervals will be calculated for analysis. Heterogeneity will be assessed statistically using the standard Chi-square and also explored using subgroup analyses based on the different study designs included in this review. Where statistical pooling is not possible the findings will be presented in narrative form including tables and figures to aid in data presentation where appropriate.
Qualitative research findings will, where possible be pooled using JBI-QARI. This will involve the aggregation or synthesis of findings to generate a set of statements that represent that aggregation, through assembling the findings rated according to their quality, and categorising these findings on the basis of similarity in meaning. These categories are then subjected to a meta-synthesis in order to produce a single comprehensive set of synthesised findings that can be used as a basis for evidence-based practice. Where textual pooling is not possible the findings will be presented in narrative form.
In the final step, the results of the quantitative review and the qualitative review will be integrated using the JBI Mixed Methods Aggregation Instrument (MMARI). To achieve integration, quantitative review findings will be translated into qualitative results through Bayesian conversion to produce synthesised findings.

Subgroup and heterogeneity analysis
In order to understand what interventions work for whom and how, we will investigate, where available, demographic factors such as gender, age, and setting. This may also inform the heterogeneity between interventions.
Heterogeneity analysis is used in meta-analysis to estimate the degree to which the heterogeneity of effect sizes across studies is due to true heterogeneity or within-study error (Borenstein et al., 2009). A general procedure in estimating heterogeneity is to compute the study-to-study observed variation, estimate the expected variation amount if the true effect were the same across the studies, and then calculate the remaining variation. We would assume this variation to represent true differences, or heterogeneity, in the effect size distribution. Rather than assessing whether heterogeneity is present, we will use the I 2 statistic, reported as part of the Forest Plot available in JBI-MAStARI software, to help assess the proportion of variability associated with between-study heterogeneity. The I 2 statistic represents the percentage variability in effect size estimates that is due to heterogeneity rather than sampling variability (Higgins & Green, 2011). The general guide of 50% is interpreted to represent moderate heterogeneity. In addition to reporting I 2 as a relative measure of heterogeneity, the heterogeneity parameter estimate (̂2) as an absolute measure of heterogeneity will also be reported.

Testing for publication bias
Assessing risk of publication bias will be an important task because of its potential influence on estimates of intervention effects. This review will analyse and ameliorate possible publication bias by implementing the trim-and-fill method (Duval & Tweedie, 2000) using the Meta package in R (Schwarzer, 2007) 2 . This allows an initial assessment of whether unpublished data on Autism and anxiety interventions (likely to have null results) is an important issue in this area (Uljarević & Hamilton, 2013).

Sensitivity analysis
Given the diversity of interventions and the potentially small sample of included studies within each intervention type, it will be important to conduct a sensitivity analysis of the impact of a single study or the impact of including studies of varying levels of quality on the overall observed effect size for interventions in any meta-analysis. Two main sensitivity analyses will be conducted, one excluding studies of lower quality and one focussing on excluding single studies which may have had an unduly large effect on the results. Results will then be compared to provide an indication of the robustness of the review's findings.

TIMEFRAME
We anticipate the following updated timeline. Due to delays in the protocol review process, we have requested an extension for the submission of the draft final review to December 2017. This will delay the publication of the final review until February 2018 which we hope will be acceptable.

Phase Stage Timeframe
Title

REVIEW AUTHORS
Lead review author: The lead author will develop and co-ordinates the review team, discuss and assign roles for individual members of the review team, liaise with the editorial base and take responsibility for the on-going updates of the review.

Name: Petra Lietz
Title: Dr

ROLES AND RESPONSIBILITIES
A brief description of content and methodological expertise within the review team is provided. It includes personnel on the review team who have content expertise, methodological expertise, statistical expertise, and information retrieval expertise.

Content:
Dr For further content guidance, she will work closely with the proposed advisory panel members, particularly Dr Mirko Uljarević who has agreed to contribute detailed topic relevant advice to the review free of charge in return for being a co-author on the published review.

Systematic review methods:
Dr Petra Lietz was a co-author of systematic reviews (Best et al., 2013;Lietz et al., 2016) and meta-analyses (Lietz 2006a, b) demonstrating her expertise with these methods.
Dr Katherine Dix has been formally trained and is an accredited JBI Comprehensive Systematic Review reviewer.

Information retrieval:
Ms Jenny Trevitt is an experienced librarian with more than 10 years' experience as a librarian in ACER's Cunningham Library. Ms Trevitt has also been directly involved with information retrieval for previous systematic reviews as detailed above.

Support staff:
Mrs Elizabeth O'Grady will provide research support to the review team. Elizabeth has extensive experience in education research methods as well as statistical analysis as demonstrated by the following publications: ; Lietz et al. (2015);De Bortoli et al. (2014).

Authors' responsibilities
By completing this form, you accept responsibility for preparing, maintaining, and updating the review in accordance with Campbell Collaboration policy. The Coordinating Group will provide as much support as possible to assist with the preparation of the review.
A draft protocol must be submitted to the Coordinating Group within one year of title acceptance. If drafts are not submitted before the agreed deadlines, or if we are unable to contact you for an extended period, the Coordinating Group has the right to de-register the title or transfer the title to alternative authors. The Coordinating Group also has the right to de-register or transfer the title if it does not meet the standards of the Coordinating Group and/or the Campbell Collaboration.
You accept responsibility for maintaining the review in light of new evidence, comments and criticisms, and other developments, and updating the review every five years, when substantial new evidence becomes available, or, if requested, transferring responsibility for maintaining the review to others as agreed with the Coordinating Group.

Publication in the Campbell Library
The support of the Coordinating Group in preparing your review is conditional upon your agreement to publish the protocol, finished review, and subsequent updates in the Campbell Library. The Campbell Collaboration places no restrictions on publication of the findings of a Campbell systematic review in a more abbreviated form as a journal article either before or after the publication of the monograph version in Campbell Systematic Reviews. Some journals, however, have restrictions that preclude publication of findings that have been, or will be, reported elsewhere and authors considering publication in such a journal should be aware of possible conflict with publication of the monograph version in Campbell Systematic Reviews. Publication in a journal after publication or in press status in Campbell Systematic Reviews should acknowledge the Campbell version and include a citation to it. Note that systematic reviews published in Campbell Systematic Reviews and co-registered with the (student OR child* OR adolescen* OR preadolescen* OR (pre W/0 adolescen*) OR youth OR teen* OR (teen W/0 age*) OR "young people" OR "young person" OR boy OR girl) S2 TITLE-ABS-KEY (anxiety OR anxious OR internali* OR fear) S1 TITLE-ABS-KEY (asd OR asperger* OR autis* OR "Pervasive Developmental" W/0 disorder* OR "PDD NOS" OR "PDD unspecified)

Search Queries
Search Mode: Boolean/Phrase. Related Terms Not Applied.

S4
Intervention OR treatment OR therap* OR psychotherap* OR evaluation OR outcome OR program OR trial* OR experimental OR control group OR random* OR best practi* or evidence based

S3
Student OR child* OR adolescen* OR preadolescen* OR pre adolescen* OR youth OR teen* OR teen age* OR young people OR young person OR boy OR girl

S2
Anxiety OR anxious OR internali* OR fear S1 ASD OR Asperger* OR autis* OR Pervasive Developmental Disorder OR PDD NOS OR PDD unspecified