1 Preventative co-ordinated low-level support for adults with autism: systematic review and service mapping Project protocol 25 November 2015 Introduction Policy background Autism is a lifelong developmental condition that affects how a person communicates with others. It is a spectrum disorder, which means people are affected in different ways. The support needs of people with autism spectrum conditions vary. People with autism and their families often face difficulties when trying to access services, can be overlooked or ignored by professionals and frequently fail to be recognised as requiring personalised support. Historically this has particularly been true of people with high-functioning autism spectrum conditions, i.e. those without learning disabilities, who may need considerable support but have often been deemed ineligible for services. Local Authorities and health services are required to provide preventive services for adults with autism spectrum conditions and support for people in their communities as part of their duty under the Care Act 2014. In particular, under the Care Act, [a]ssessment of eligibility for care services cannot be denied on the grounds of the person s IQ 1 that is, eligilibity for services now explicitly includes people with high-functioning autism. In addition, Fulfilling and Rewarding Lives, supported by the Autism Act 2009, recommends that policy bodies take a more preventative, supportive approach, 2 in order to avoid mental health crises and the potential consequences in terms of service use. 3 Both Fulfilling and Rewarding Lives and the Care Act 2014 emphasise the need to focus on early support, which may include for example befriending, mentoring and advocacy services, as opposed to later crisis management (although such services may not be funded by DH). From a financial point of view, many local authorities and the NHS are not putting resource into these preventive services and incurring increased costs and demand for statutory services at a later point in the care process. 4 1 Department of Health (2015) Statutory guidance for Local Authorities and NHS organisations to support implementation of the Adult Autism Strategy, p. 19. 2 Department of Health (2010) Fulfilling and Rewarding Lives: The strategy for adults with autism in England, p. 61. 3 Department of Health (2015) Statutory guidance for Local Authorities and NHS organisations to support implementation of the Adult Autism Strategy, p. 36. 4 National Audit Office (2009) Supporting people with autism through adulthood.
2 Why a review is needed The total cost of autism is estimated to be 27.5 billion each year. 5 This figure assumes that the lifetime cost for someone with autism is 3.1 million and 4.6 million for high and low functioning autism respectively. Services (including social care, hospital services, residential care, day services and education) account for 59% of these costs, lost employment and the corresponding out of work benefits for 36%, and the remainder by family expenses. It is possible that a non-negligible part of these costs in the case of people with high-functioning autism could be saved by a range of targeted, preventative and supportive services that aim to help people, at any stage in their adult life, to build and maintain their own resilience so as to fulfil their potential and lead relatively independent lives. In 2014 the Department of Health issued funding for Autism innovation projects (see https://lynromeo.blog.gov.uk/wpcontent/uploads/sites/70/2014/11/ listofsuccessfulprojects-pdf.pdf). These initiatives include mentoring, advocacy, befriending, help into employment, advice and information about services, and access to social groups. However, evaluations of these and other such initiatives appear limited and evidence is currently lacking on the most effective and appropriate preventative support mechanisms for adults with autism in England. We propose to review the existing international literature and assess services being delivered by English Local Authorities, their NHS partners and by third sector organisations. The review also responds to research priorities identified by the autism community. Recent work with people with autism, parents and practitioners 6 identifies the following areas as under-served by current funding priorities: research on adults; research on real world services which can inform evidence-based service provision (as opposed to basic biomedical research); and research on services supporting people in their day-to-day life and employment. In addition, initial discussions with stakeholders suggest that unmet need for people with highfunctioning autism spectrum conditions may be a particular concern. Overview and structure The focus of the project as a whole will be on low-level support services for adults with autism spectrum conditions (including Asperger s) without learning disability ( high-functioning ). Low-level support services are defined as non-intensive services aiming to provide general 5 Knapp M, Romeo R, Beecham J (2009). Economic cost of autism in the UK. Autism 13(3):317-336. 6 Pellicano E, Dinsmore A, Charman T (2014). What should autism research focus upon? Community views and priorities from the United Kingdom. Autism 18(7): 756-770.
support to people in their everyday lives, and which is not directed at treating a clinical problem or deficit. This could include, for example,practical support, social and emotional support, befriending and opportunities for social interaction, peer-led interventions, education or training, and brokering or advocacy around employment or services, delivered in a range of settings, including health services, social care or community settings, as well as phone- or internet-based services. In particular, low-level support may involve multi-disciplinary teams and/or liaison between multiple service providers, and signposting individuals to services based on individual needs. This project will not look at clinical services or psychotherapy as stand-alone interventions. We have developed a draft logic model, shown in Appendix 1, on the basis of information gained from current policy documents, information received from Advisory Group members, and our initial scoping searches of the research literature. The logic model sets out a schematic overview of the kinds of components that might be included, the intermediate outcomes and mechanisms through which they could work, and the outcomes they could impact on. This logic model will be revised in light of feedback from the Project Advisory Group and emerging evidence from the project. The project consists of two phases which will be pursued in parallel: 1) A systematic review of international research evidence on the effectiveness and costeffectiveness of low-level support services for people with autism; 2) A descriptive map of existing service provision in England. Detailed methods for these two phases are set out below. In addition, we will: 3) Compare the findings from (1) and (2) to determine which services and interventions are supported by evidence of effectiveness and cost-effectiveness; 4) Map the data from (1) and (2) onto the logic model, in order to identify gaps in the evidence and inform future evaluation work; 5) If the evidence gathered during phase 1 permits, attempt to model the costs and benefits of the services identified in phase (2). We expect to complete the draft report in March/April 2016. The report will be sent to the Advisory Group before completing a final report for peer review by the Department of Health Policy team. However, this timeline is provisional and may change depending upon the volume of evidence located, further methodological decisions (particularly regarding the cost-benefit analysis (5)) and any revisions to the scope of the project. 3
4 Methods: Advisory Group Throughout the project we will seek advice from the Project Advisory Group including policy stakeholders, service user representatives and topic experts. After an initial teleconference in September 2015 with a subset of Advisory Group members, we asked the Group by email for advice on the planning stages of the project, which informed the development of this protocol. We aim to convene a face-to-face Advisory Group meeting in early 2016, where we will present an interim overview of the evidence from phases (1) and (2) and seek advice from Advisory Group members. Areas for discussion could include: the important dimensions that make up the full range of preventative support services; types of intervention, settings, outcomes, or people with autism that should be a particular focus for the review s analyses; possible sources for finding evaluation work; and sources of information about recent service delivery in the UK. Methods: evidence review The review process will be systematic and fully documented, although given the relatively brief timescale, searching will be focused rather than fully exhaustive. The review will be registered on PROSPERO. The review methods will follow CRD guidance 7 and the review will be reported according to the PRISMA statement (www.prisma-statement.org). Review question What is known about the effectiveness, cost-effectiveness, and barriers and facilitators of low-level support services for adults with high-functioning autism? Searching The search terms will take the form: (autism or Asperger s) AND (adults) AND (low-level support) The last cluster will contain a wide range of terms to capture the various types of intervention which might form part of a low-level support service. The draft full strategy as developed for the MEDLINE database is shown in Appendix 2. The following database sources will be searched from inception to the most current records: ASSIA EMBASE ERIC MEDLINE PsycInfo Social Care Online (simplified strategy) 7 Centre for Reviews and Dissemination (2008), Systematic Reviews: CRD s guidance for undertaking reviews in health care, 3rd ed.
5 Social Policy and Practice The following journals will be handsearched for the last five years: Autism Journal of Autism and Developmental Disorders Google searches will be conducted using specialised strategies to locate reports conducted by relevant bodies. In addition, any unpublished reports located in the service mapping phase of the project will also be manually screened for inclusion in the review. Screening The inclusion criteria will be: 1. Does the study present primary empirical data of relevance to an intervention? Include trials, one-group studies, retrospective studies, process evaluations, economic analyses (cost-effectivenesss or cost-benefit analyses), and qualitative research which reports substantive data on an intervention. Exclude observational or qualitative studies which may include data on services generally, but do not relate to (a) specific intervention(s). Exclude case studies without primary qualitative data. Exclude nonsystematic reviews; retain systematic reviews of relevant data for reference checking. 2. Does the study include people with high-functioning autism (HFA), or their families or carers, or concern an intervention for people with HFA? Include any autistic spectrum condition (ASC), including Asperger Syndrome, without learning disability; include participants without a formal diagnosis if intervention is mainly aimed at ASC; include at abstract stage if population is reported as ASC not further specified (i.e. if it is unclear whether participants are high- or low-functioning). Include studies of professional training if the intervention is designed to support delivery of a specific service for people with HFA. Include studies of mixed populations including high-functioning ASC if separate analyses of this group are reported (or if they represent 50% of the sample), otherwise exclude. 3. Does the study concern a low-level support intervention? Include any service designed to support individuals in their daily lives, including: the provision of advice, information, or advocacy services; assistance in accessing services; peer support or support groups; supported employment; support with social interaction or participation. Exclude clinical interventions including psychotherapy and CBT, any intervention mainly focused on reducing specific psychological morbidity (e.g. anxiety, sensory disorders, repetitive behaviour), and facilitated communication. Exclude interventions narrowly focused on a single deficit, such as pure social skills training, but include multi-component interventions with some component of social skills.
6 4. Does the study report data on a policy- or practice-relevant outcome? Exclude tests of purely cognitive or psychological deficits which do not reflect a real world impact. For example: tests of recognition of facial affect (e.g. Face Emotion Identification Test); tests of emotional cognition (e.g. Cambridge Mind Reading battery; Hinting Task); social skills questionnaires (e.g. Social Communication Skills Questionnaire); tests of cognitive skills or memory. Include all other outcomes, either self-rated or observer-rated. For example: actual social behaviour (e.g. Social Responsiveness Scale), participation in social situations, or quality of social relationships (e.g. Index of Peer Relations); job performance, employment or wages earned; mental health, wellbeing or satisfaction; independence or activities of daily living; etc. 5. Does the study population include adults aged 18 or older? Include studies with 50% of participants 18 years, or with separate analyses of adult participants. At abstract stage exclude abstracts describing population as children or schoolchildren, but include those describing them as young people or adolescents, or focusing on transitions to adulthood (also exclude non-asc parents of young children with ASC). 6. (Full-text screening only) Is the study published in English? An initial sample of 10% of titles and abstracts will be screened by two reviewers independently and differences resolved by discussion. If agreement is satisfactory at this stage, remaining abstracts will be screened by a single reviewer. Full text will be retrieved for all abstracts meeting criteria and screened by two reviewers independently. Having applied the above criteria, the evidence will be organised by intervention type (e.g. employment support; peer support). If the evidence permits, we will apply study type filters (e.g. including only prospective controlled studies); these filters may vary between categories. This is to ensure that the review findings are based on the best available evidence for each intervention type, without excluding evidence on less well-studied interventions. Quality assessment, data extraction and synthesis Quality assessment and data extraction will use structured tools to extract information on the sampling and recruitment of participants, the characteristics of the intervention and comparator, the study design, and the outcomes. The exact tools used for quality assessment will depend on the study types included (see above): we will use specialised tools for RCTs and non-randomised studies, and different tools for economic studies and for qualitative studies, if any are included.
7 Data synthesis of effectiveness and cost-effectiveness findings will be carried out using a narrative synthesis, using the logic model as an initial framework to structure the synthesis. 8 (We do not anticipate that the data will support quantitative meta-analysis, but will consider conducting meta-analysis if this turns out not to be the case.) If any qualitative data are included, these will be integrated into the narrative using a thematic framework synthesis. Methods: service mapping The service mapping phase aims to provide an overview of current provision of low-level support services for adults with high-functioning autism in England. Compared to the review phase, the methods for this phase will be less systematic and more pragmatic, although the process will be fully documented and therefore transparent. The methods will also evolve as the project progresses, in consultation with the Project Advisory Group. Data sources may include: Information from DH including documentation of the Autism Innovation Fund; Information supplied by Project Advisory Group members (including substantial documentation already supplied by Peter Hopkins); Internet searches; Information from service directories, e.g. the Autism Services Directory maintained by the National Autistic Society; and/or The Autism Self-Assessment Framework returns. The criteria for including services in the mapping will be similar to those in the review i.e. lowlevel support for adults with high-functioning ASC but may be interpreted somewhat more broadly depending on the relevance of the service to the overall project goals and/or the availability of data. We will consult with the Project Advisory Group and the Department of Health on an ongoing basis to ensure that the information collected is useful. The initial focus of the service mapping will be on providing a high-level overview of the services offered by different organisations, based on publicly available information. We will initially prioritise projects for which substantive information is available and extract information available on: the services offered; the populations served; and any data on processes or outcomes. 8 Centre for Reviews and Dissemination (2008), Systematic Reviews: CRD s guidance for undertaking reviews in health care, 3 rd ed., 1.3.5.1 (pp. 48-53); Popay J et al. (2006), Guidance on the Conduct of Narrative Synthesis in Systematic Reviews: A Product from the ESRC Methods Programme.
8 Appendix 1. Draft logic model
Appendix 2. Draft MEDLINE search strategy Database: Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations and Ovid MEDLINE(R) <1946 to Present> Search Strategy: -------------------------------------------------------------------------------- 1 autistic disorder/ (17281) 2 asperger syndrome/ (1602) 3 (autism or autistic or asperger$).ti,ab. (29387) 4 (adult$ adj2 ASD).ti,ab. (456) 5 1 or 2 or 3 or 4 (31904) 6 (low level intervention$ or low intensity intervention$).ti,ab. (133) 7 (low level support$ or low intensity support$).ti,ab. (12) 8 Social skills/ (238) 9 "Activities of Daily Living"/ (53991) 10 Occupational Therapy/ (10812) 11 (social skill$ or life skill$).ti,ab. (4416) 12 (skill$ adj2 (independen$ or employ$)).ti,ab. (709) 13 (independen$ adj2 (live or lives or living)).ti,ab. (3893) 14 ((promot$ or encourage$ or support$ or enhanc$ or increas$) adj2 (empathy or socialization or socialisation or interaction or friend$)).ti,ab. (10865) 15 ((promot$ or encourage$ or support$ or enhanc$ or increas$) adj2 (independen$ or engagement or involvement or inclusion or participation)).ti,ab. (26527) 16 ((reduc$ or decreas$ or discourag$ or prevent$ or lessen) adj2 (isolation or dependenc$ or loneliness)).ti,ab. (3681) 17 Self care/ (25987) 18 Self-Help Groups/ (8170) 19 (self help or selfhelp or self care or selfcare or support group$).ti,ab. (21770) 20 Social Participation/ or Friends/ (3983) 21 (social activit$ or social group$ or social involve$ or social inclusion or social network$ or social participat$).ti,ab. (19163) 22 (community activit$ or community group$ or community involve$ or community inclusion or community network$).ti,ab. (3459) 23 (games or leisure or sport or sports or hobby or hobbies or club$).ti,ab. (74912)
24 exp Leisure Activities/ (179407) 25 community networks/ or social support/ (61301) 26 Mentors/ (8323) 27 (support$ or coach$ or mentor$ or befriend$ or broker$ or advise$ or advisor$ or buddy or buddies or facilitat$ or outreach).ti,ab. (1467410) 28 Patient Advocacy/ (22884) 29 (advocacy or advocate$).ti,ab. (48840) 30 (access adj2 (service$ or facilities or resources or activities or advice)).ti,ab. (4712) 31 (liaison adj services).ti,ab. (198) 32 (information adj2 (service$ or benefit$ or provid$ or provision or give or gives or giving or gave)).ti,ab. (138037) 33 (advice adj2 (service$ or benefit$ or provid$ or provision or give or gives or giving or gave)).ti,ab. (4704) 34 "advice and information".ti,ab. (272) 35 exp Self Concept/ (78463) 36 (Well being or wellbeing).ti,ab. (52818) 37 ((promot$ or encourage$ or support$ or enhanc$ or increas$) adj2 (self esteem or selfesteem or self-esteem or self confiden$ or selfconfiden$ or self-confiden$ or resilient$ or resiliency or confident$ or confidence)).ti,ab. (6161) 38 person centred approach.ti,ab. (136) 39 Patient-Centered Care/ (12555) 40 Caregivers/ (24263) 41 ((Carer$1 or caregiv$ or caretaker$ or care taker$ or custodian$ or guardian$ or family or families or father$ or mother$ or parent$ or sibling$ or brother$ or sister$ or spouse$ or wife$ or husband$ or partner$ or peer or peers or friend$ or online or neighbour$) adj2 (group$ or support$ or network$)).ti,ab. (41575) 42 (peer adj2 (training or mediat$ or advoca$)).ti,ab. (528) 43 Employment/ or Employment, Supported/ (39330) 44 ((employ$ or work or workplace or job or vocational) adj2 (train$ or prepar$ or opportunit$ or skill$ or rehabilitat$ or support$ or placement$ or interview$ or search$ or competitive$)).ti,ab. (22258) 45 "individual placement and support".ti,ab. (173) 46 (vocational adj (independence or engagement)).ti,ab. (15) 47 (autism adj3 champion$).ti,ab. (0) 48 ((money or financ$) adj2 (manag$ or advice)).ti,ab. (2211)
49 Public Assistance/ (2625) 50 (benefit$ adj2 (payment$ or advice or welfare)).ti,ab. (663) 51 ((housing or tenancy or tenant$ or accommodation) adj2 (advice or support$)).ti,ab. (702) 52 ((support$ or mentor$) adj2 (universit$ or college$)).ti,ab. (374) 53 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42 or 43 or 44 or 45 or 46 or 47 or 48 or 49 or 50 or 51 or 52 (2158008) 54 5 and 53 (6985) 55 exp adult/ or middle aged/ or young adult/ (5977614) 56 (adult$ or men or women or man or woman or worker$ or employee$).ti,ab. (2240431) 57 (people or person$).ti,ab. (744455) 58 55 or 56 or 57 (7234915) 59 54 and 58 (2717) 60 exp animals/ not humans.sh. (4135218) 61 59 not 60 (2662)