Quality-of-Life Discrepancies Among Autistic Adolescents and Adults: A Rapid Review

Abstract

Importance: Quality of life (QoL) is a core outcome of occupational therapy, but it is decreased among autistic adolescents and adults. This is the first review of QoL from an occupational therapy standpoint.

Objective: To identify self-reported QoL differences between autistic and nonautistic samples; investigate sex differences in QoL among autistic people; examine consistency in QoL among autistic people across age, intellectual disability (ID), and self- versus proxy-report method; and appraise occupational therapy–related interventions addressing QoL among autistic people.

Data Sources: Articles published in peer-reviewed journals between 2010 and 2020, located through Academic Search Ultimate, PubMed, and OTseeker, along with the American Journal of Occupational Therapy, British Journal of Occupational Therapy, Canadian Journal of Occupational Therapy, and Australian Occupational Therapy Journal.

Study Selection and Data Collection: Article samples consisted of at least 20% autistic females and used self-reported QoL measures. Qualitative research was excluded, as well as studies with participants younger than age 13 yr. Twenty-seven articles qualified (3 Level 1B, 13 Level 3B, and 11 Level 4).

Findings: Autistic adolescents and adults demonstrated decreased self-reported QoL compared with nonautistic peers across age and ID presence. One article compared sex differences in QoL among autistic people. Interventions improved QoL among autistic people.

Conclusions and Relevance: Autistic adolescents and adults demonstrate decreased QoL in comparison with nonautistic peers. Questions remain related to sex differences in QoL among autistic people, an area for future research. Multiple evidence-based approaches to improve QoL in the autistic population are relevant to occupational therapy practice.

What This Article Adds: Self-reported QoL among autistic adolescents and adults is clearly lower than among their nonautistic counterparts. This article provides information on multiple interventions related to occupational therapy to improve QoL among autistic people.

Quality of life (QoL) is at the core of occupational therapy (American Occupational Therapy Association [AOTA], 2020b). Occupational therapy providers are well suited to address the potential QoL issues experienced by autistic people,¹ particularly adolescents and adults. However, the majority of studies of QoL among people with autism have focused on parent or proxy reporting and have not always emphasized the inclusion of autistic women and girls. These are key gaps, because self-reported QoL should be the basis on which occupational therapy practitioners approach clients and design interventions, and women and girls likely represent a larger proportion (1:3–4) of people on the autism spectrum than previously suggested (Loomes et al., 2017; Maenner et al., 2020). With occupational therapy’s principles of beneficence, autonomy, and justice (AOTA, 2015), an emphasis on autistic women and girls encourages equitable opportunities for both genders, and an emphasis on self-reports allows the autistic person’s voice to be heard rather than that of a reporter. The purpose of this rapid review is to provide a better understanding of the QoL and needs of autistic adolescents and adults and of the malleability of QoL in response to intervention, particularly with respect to autistic women and girls, allowing occupational therapy providers to more effectively implement therapeutic interventions for autistic people across the lifespan.

QoL is a multidimensional construct that includes the domains of emotional well-being, interpersonal relationships, material well-being, personal development, physical well-being, self-determination, social inclusion, and rights (Schalock, 2000). Gauging a client’s current and ideal QoL is an important part of the occupational profile (AOTA, 2020b). Not only do occupations, client factors, performance skills, performance patterns, and context and environment influence overall QoL, but QoL is a designated outcome of occupational therapy.

Occupational therapy practitioners should also consider the QoL of autistic people in the development of assessment and intervention tools. Occupational therapy–relevant interventions that address QoL have recently emerged in the literature (Curtin et al., 2016; García-Villamisar & Dattilo, 2010; Hesselmark et al., 2014; Holmefur et al., 2019; Jamison & Schuttler, 2017; Nadig et al., 2018; Siew et al., 2017; Wentz et al., 2012), but converging evidence across studies is needed to determine the degree to which QoL can be improved through intervention with autistic adolescents and adults. Occupational therapy practitioners have the potential to fulfill societal needs through a focus on education and supports that might improve QoL among autistic people with services provided across settings and ages. Specific education and program development is necessary for autistic people across the lifespan. Research on autistic adolescents and adults is clearly lacking, despite the fact that autism is a lifelong condition (Roux et al., 2015).

Many girls do not receive an autism diagnosis as a result of diagnostic methodology that uses a male sample for standardization and male bias in research samples (Frazier et al., 2014; Halladay et al., 2015; Hiller et al., 2014; Ratto et al., 2018). There are sex differences² in self-reported QoL for the nonautistic population, with women and girls typically demonstrating decreased QoL in comparison with male counterparts (da Rocha et al., 2014; Lee et al., 2020). However, less is known about discrepancies in QoL between genders in the autistic population, although research suggests that autistic people demonstrate decreased QoL in comparison with nonautistic people (Ayres et al., 2018).

When research does include autistic women, it does not always include the self-reports of autistic adolescents and adults. Emphasizing the need for self-reports, whether in isolation or in conjunction with proxy reports, prioritizes client autonomy and personal perception (AOTA, 2015; Erez & Gal, 2020; Fujiura & RRTC Expert Panel on Health Measurement, 2012). This rapid review contributes to the profession’s understanding of the needs of both male and female autistic adolescents and adults and how to design interventions that promote QoL for all autistic people.

Previous systematic reviews of QoL among autistic adolescents and adults (Ayres et al., 2018; Tobin et al., 2014; van Heijst & Geurts, 2015) have concluded that they demonstrate lower QoL in comparison with their nonautistic counterparts. Another review (Kim & Bottema-Beutel, 2019) found that social functioning was most correlated with QoL, whereas Tobin et al. (2014) focused on the interaction between social functioning and QoL among autistic adolescents and adults. However, these reviews did not always include self-report measures or a specific percentage of female participants, or they did not make conclusions related to occupational therapy practice.

To fill this key gap, our review emphasizes the use of self-reports and the inclusion of autistic females in the context of occupational therapy practice. Specifically prioritizing the inclusion of autistic females emphasizes the value of their voice in autism research, which has not always been intentionally amplified. As such, their inclusion allows for a better understanding of well-being on the autism spectrum across the sexes. Self-reported and perceived QoL is important in gauging someone’s viewpoint and can be obtained through subjective QoL measures, rather than objective QoL measures that focus more on external factors such as physical health, neighborhood quality, or other predictors (Bishop-Fitzpatrick et al., 2016; Fakhoury & Priebe, 2002; Hong et al., 2016). However, self-reported subjective QoL is an identified issue in autism research (Hong et al., 2016), with some studies indicating that autistic people experience difficulties that make it challenging to obtain accurate information related to symptoms and experiences (Mazefsky et al., 2011) and other findings indicating that autistic people respond accurately on self-report measures (Ozsivadjian et al., 2014).

Current Centers for Disease Control and Prevention (CDC) data (Maenner et al., 2020) indicate that 33% of children with autism also have an intellectual disability (ID). When research involves people with ID, self-reports are often avoided to accommodate difficulties with reliability and validity, particularly considering the high amount of meta-cognition needed for self-report (Fujiura & RRTC Expert Panel on Health Measurement, 2012). However, these challenges can be accommodated, including modification of question structure, length, and response type and ensuring comprehension, although these modifications lead to the conclusion that a general, unmodified “one-size-fits-all” measure is inadequate and may not apply to everyone in the world (Fujiura & RRTC Expert Panel on Health Measurement, 2012). Therefore, this review includes studies with participants with both autism and ID.

The objectives of this review were to (1) identify QoL differences between autistic and nonautistic samples on the basis of self-report; (2) investigate the existing literature on whether there are sex differences in the QoL of autistic people; (3) review the currently available literature related to consistency of QoL when considering age, presence of ID, and use of self- versus self- and proxy-reported QoL; and (4) review currently available interventions in the domain of occupational therapy practice to address QoL among autistic adolescents and adults. We hypothesized that self-reported QoL scores of autistic adolescents and adults will be lower than those of nonautistic adolescents and adults; autistic females will experience lower QoL in comparison with autistic males; QoL will be consistent across age, IQ, and report method; and interventions applicable to occupational therapy practice will improve QoL among autistic adolescents and adults.

Method

Search Strategy

We performed a rapid review of the literature to identify QoL differences in the autistic population. In contrast to lengthier systematic reviews, rapid reviews focus on a timely and targeted topic, and typically fewer databases are searched to produce a review in shorter amount of time (Dobbins, 2017; Haby et al., 2016; Harker & Kleijnen, 2012). However, whenever possible, we made an effort to conform to systematic review guidelines. We included peer-reviewed articles published in English between January 2010 and April 2020. Figure 1 shows the flow of articles that were identified, screened, and deemed eligible for inclusion in the review. Under the guidance of a university-affiliated research librarian, the team conducted a search of Academic Search Ultimate, PubMed, and OTseeker, along with journal-based searches of the American Journal of Occupational Therapy, British Journal of Occupational Therapy, Canadian Journal of Occupational Therapy, and Australian Occupational Therapy Journal. Searching occupational therapy–specific journals separately ensured that any information disseminated via these resources was included in the review. These journals were selected because of their lifespan focus, in contrast with journals geared to the pediatric population. Academic Search Ultimate, OTseeker, and all journal searches used the following search terms: “autis*” (to include both autism and autistic) and “quality of life” or “autis*” and “well-being.” PubMed searches used MeSH terms: “Self Report,” “Female,” “Autism Spectrum Disorder” OR “Autistic Disorder” OR “Asperger Syndrome.” To further narrow the search, a peer-reviewed filter was used where available. Because of the interchangeable nature of quality of life and well-being in the early stages of the research, we used both terms. Emails to corresponding authors and interlibrary loan were used to provide access to articles as needed.

Figure 1.

Flow diagram of articles identified, screened, eligible for, and included in the review.

Note. Figure format from “Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement,” by D. Moher, A. Liberati, J. Tetzlaff, and D. G. Altman; PRISMA Group, 2009, PLoS Medicine, 6(7), e1000097. https://doi.org/10.1371/journal.pmed.1000097

Selection Criteria

To advance beyond the title phase of screening, an article needed to have autis* (autism or autistic) in its title. If it did not, then the abstract was reviewed for explicit mention of autism. If the title or abstract did not include autis*, the article was excluded. Similarly, for the abstract phase of screening, the article needed to include autis* and quality of life or well-being in the title or abstract. If there was no mention of quality of life or well-being in the abstract, the article was reviewed to determine whether the text included a clear mention of them. If there was no clear reference to these terms, the article was excluded. To be included in the final phase of screening (full-text articles), articles needed to have an autistic sample that was at least 20% female, included adolescents and adults (age of all participants ≥13 yr), and included at least one self-reported measure of QoL. The 20% female threshold was selected conservatively on the basis of the lowest current estimated male-to-female prevalence figures (Maenner et al., 2020), although the current estimated male-to-female prevalence of autism can vary greatly (Fombonne, 2009). Proxy reports were included if the study also included self-reports. Self-report measures could be performed in addition to parent or proxy reports.

Data Extraction

Initial article data (from the beginning of search through total articles included) were extracted by one of two authors (Emily C. Skaletski or Laura Bradley). This information included the author names, title, DOI, source (database or journal), year, whether the article included self-reported QoL measures, and whether the article’s autistic sample consisted of 20% or more women or girls. When unsure whether an article should be included, these authors reached consensus, with Brittany G. Travers serving as a final reviewer for articles as needed. Appendix Tables A.1, A.2, and A.3 include information extracted by the authors. All information from the included articles was coded by at least two authors.

Risk-of-Bias Assessment

Risk of bias was assessed at the study level. The Oxford Levels of Evidence (Oxford Centre for Evidence-Based Medicine Levels of Evidence Working Group, 2011) were used to determine the level of evidence (1–5) for each article (AOTA, 2020a). Level 1 indicates the most rigorous study design and Level 5 the least rigorous. In the process of considering study design, a key step before assessing risk of bias, 3 articles were determined to be Level 1B (well-designed randomized controlled trials [RCTs]); the remaining articles were Level 3B or 4 (case–control, cohort, and cross-sectional studies).

Articles were assessed with one of two risk-of-bias criteria sets: (1) RCT or non-RCT or (2) pretest–posttest uncontrolled, which was also used for studies that were descriptive in nature. The use of pretest–posttest uncontrolled criteria for descriptive studies poses a potential issue with risk-of-bias assessment, but this category was decidedly the best fit for these types of studies, which are already higher in bias and do not fit many of the categories of the RCT and non-RCT criteria. Risk of bias is an important consideration in determining the generalizability, reliability, and validity of studies because it determines whether a study was well constructed and whether the results are truly a result of the intervention or experience of those with a diagnosis. Further discussion of risk of bias can be found in the Results section.

Results

From the original search, we identified 1,070 studies. After removing duplicates, 803 articles remained. Of these studies, 220 were excluded because they did not have autis* (autism or autistic) in the title or abstract. Next, 57 were excluded because they did not have quality of life or well-being in the abstract or full article. The full text of the 526 remaining articles was reviewed, and 499 were excluded (Figure 1). A total of 27 articles were included in the final review (Table A.1). Sixteen did not mention IQ or include participants with co-occurring ID. Seven articles excluded participants with a co-occurring ID or an IQ < 70. Only 4 studies included participants with co-occurring ID, so generalizing to the autistic population with a co-occurring ID should be done with caution.

Risk of Bias

Tables A.2 and A.3 indicate the risk of bias for each of the studies included in the rapid review. Included articles were primarily low in risk of bias, with some moderate in risk. Most studies were case–control, case series, or cohort studies, which are lower in overall quality and have different standards for risk of bias. As such, these types of studies with low risk of bias should not be considered to have the same level of generalizability, reliability, and validity as an RCT systematic review with low risk of bias. The average sample size for included articles was 336 (range = 9–2,341); 7 studies had a sample sizes of <50, and 13 had a sample size <100. Convenience sampling was predominantly used as a recruitment method, which further increases the risk of bias.

Measurement

A wide range of QoL instruments were used in the included studies, but three instruments were used in 4 or more studies: the Quality of Life Questionnaire (QOL; Schalock & Keith, 1993), Satisfaction With Life Scale (SWLS; Diener et al., 1985), and the World Health Organization Quality of Life–Brief Version (WHOQOL–BREF; WHOQOL Group, 1998). The SWLS has not been validated with the autistic population. The QOL was validated with the intellectual and developmental disabilities population, although whether participants had an autism diagnosis was not discussed and thus cannot be assumed (Caballo et al., 2005). The WHOQOL–BREF has been validated with the autistic population (α = .93; McConachie et al., 2018).

Findings

Nonintervention Studies

Nineteen articles were exploratory and descriptive in nature. Of these 19 studies, 3 had samples whose mean age was <20 yr. Jamison and Schuttler (2015) used self-reports of QoL, and Franke et al. (2019) and Mahfouda et al. (2019) used both self- and proxy reports of QoL. Autistic adolescents experience reduced QoL and life satisfaction compared with their peers, which was found with a male and female sample, a female-only sample, and a gender-diverse sample (Jamison & Schuttler, 2015; Franke et al., 2019; Mahfouda et al., 2019, respectively).

Seven nonintervention studies included participants whose mean age was in the 20s. Four studies used self-report QoL measures (Lin, 2014; Lin & Huang, 2017; Rodgers et al., 2018; White et al., 2018). The remaining 3 articles used self- and proxy reports (Kamio et al., 2013; Knüppel et al., 2018; Pearlman-Avnion et al., 2017). In Lin’s (2014) study, autistic adults demonstrated lower QoL than a nonautistic control group and a Taiwanese health population reference group. Knüppel et al. (2018) specifically reported that self-reported QoL tended to be slightly higher than proxy-reported QoL.

Seven nonintervention studies included participants whose mean age was in the 30s. Four of these articles solely used self-reported measures of QoL (Hull et al., 2019; Leader et al., 2018; Mazurek, 2014; McDonald, 2017). The 3 remaining articles used proxy- and self-reported measures of QoL (Deserno et al., 2017; Hong et al., 2016; McConachie et al., 2018). Autistic adults demonstrated decreased QoL compared with the general population (McConachie et al., 2018) and reliably rated their QoL (Hong et al., 2016).

Two nonintervention articles included participants whose average age was in the 40s. Grove et al. (2018) examined QoL using self-reports exclusively, whereas Mason et al. (2018) used both self- and proxy reports of QoL. Autistic adults experienced lower QoL than nonautistic peers, with female sex being a predictor of reduced QoL (Mason et al., 2018).

Intervention Studies

Eight studies examined the effectiveness of an intervention, typically in an effort to improve QoL. Nearly all of these studies used different QoL outcome measures, which poses some difficulty in considering the results collectively. Of the 8 studies, 3 were appraised as Level 1B (RCTs; García-Villamisar & Dattilo, 2010; Hesselmark et al., 2014; Nadig et al., 2018). However, Nadig et al. (2018) elected not to complete statistical analysis because of their small sample size (n = 26). Three studies included participants with a mean age <20 yr (Curtin et al., 2016; Jamison & Schuttler, 2017; Siew et al., 2017). Curtin et al. (2016) had a small sample size (n = 9), so interpretation of inference tests of QoL changes was not possible. In addition, this study used both self- and proxy reports of QoL, the only intervention study to use both reporting methods. The remaining intervention studies solely used self-reports of QoL.

Three studies used informal or formal mentoring (Curtin et al., 2016; Jamison & Schuttler, 2017; Nadig et al., 2018). QoL was significantly improved postintervention (social and self-care skills group, Girls Night Out; Jamison & Schuttler, 2017), as was perceived social support (Nadig et al., 2018). The remaining 5 studies included participants whose mean ages were in the 20s and 30s; each study used a specific tool or program: customized leisure programming (García-Villamisar & Dattilo, 2010), an organization and executive functioning program (Let’s Get Organized; Holmefur et al., 2019), cognitive–behavioral therapy and recreational activity (Hesselmark et al., 2014), a college mentoring program (Siew et al., 2017), and an online support and coaching program (Wentz et al., 2012). Collectively, these studies suggest that QoL can be improved in the autistic population across a variety of programs and interventions and for a wide range of ages.

Discussion

The objectives of this rapid review were to identify differences in QoL between autistic and nonautistic people, determine any discrepancies in QoL between the sexes, evaluate whether these results are consistent across age and IQ and between self- and proxy-reported QoL, and examine interventions applicable to occupational therapy to address QoL among autistic people. Similar to previous research (Ayres et al., 2018; Kim & Bottema-Beutel, 2019; Tobin et al., 2014; van Heijst & Geurts, 2015) and supporting our first hypothesis, we found that QoL was lower among autistic compared with nonautistic adolescents and adults (Franke et al., 2019; Jamison & Schuttler, 2015; Lin, 2014; Lin & Huang, 2019; Mahfouda et al., 2019; Mason et al., 2018; McConachie et al., 2018). However, our review expands on this previous research, finding that decreased QoL was consistent across self-reported or a combination of self- and proxy-reported QoL, with samples including autistic females and participants with ID. As mentioned previously, although these results indicate that autistic participants with ID reported decreased QoL similar to that of autistic participants without ID, only 4 articles explicitly included participants with ID, which limits the generalizability of the results to participants with ID.

In an effort to get a representative sample of autistic women and girls, all studies included in the review had to have an autistic sample consisting of 20% or more female participants. However, we could not specifically determine whether QoL is decreased among autistic women and girls, because only 1 study directly compared sex differences in QoL among autistic people. To be specific, 1 Level 4 study determined that being female was a predictor of lower QoL among autistic adults (Mason et al., 2018), which is consistent with findings with nonautistic samples (da Rocha et al., 2014; Lee et al., 2020). However, a finding from a single study that used a convenience sample is insufficient to make conclusions about the impact of sex on self-reported QoL, particularly given this study’s high risk of bias.

Notably, of all the included studies, just 3 had only female participants, which did not allow for analysis of the differences in QoL between the sexes (Jamison & Schuttler, 2015, 2017). As such, there remains a key gap in the literature regarding whether autistic females experience reduced QoL in comparison with autistic males. A critical area of future research will be to examine QoL among autistic females, with an emphasis on large sample sizes and random sampling from the population of autistic people. Direct comparison of QoL between autistic females and males is essential because of the documented decreased QoL among nonautistic women (da Rocha et al., 2014; Lee et al., 2020) and the corresponding health-related consequences (CDC, 2018).

Our findings indicate that QoL is malleable, with numerous previously studied interventions applicable to occupational therapy that improved QoL among autistic adolescents and adults. These interventions include in-person and online mentoring and support, leisure programming, cognitive–behavioral therapy, cognitive strategy education (Let’s Get Organized), social and self-care skills development (Girls Night Out), and recreational activity (Curtin et al., 2016; García-Villamisar & Dattilo, 2010; Hesselmark et al., 2014; Holmefur et al., 2019; Jamison & Schuttler, 2017; Nadig et al., 2018; Siew et al., 2017; Wentz et al., 2012). Mentoring programs integrated the use of nonautistic peers as mentors, which may be a method to improve social participation while influencing QoL outcomes. Two intervention studies had authors who were either occupational therapists or affiliated with an occupational therapy program: a mentoring program (Siew et al., 2017) and a cognitive strategies program, Let’s Get Organized (Holmefur et al., 2019), that emphasizes the value of QoL intervention in occupational therapy practice.

Ultimately, QoL for autistic adolescents and adults is poorer than for their nonautistic counterparts, which is consistent across age, presence of ID, and self-report. QoL is a significant predictor of current and future health (CDC, 2018), and its consideration for the autistic population is critical, given known health disparities among autistic people (Bishop-Fitzpatrick & Kind, 2017). This rapid review indicates that multiple occupational therapy–related interventions can improve QoL among autistic people (Curtin et al., 2016; García-Villamisar & Dattilo, 2010; Hesselmark et al., 2014; Holmefur et al., 2019; Jamison & Schuttler, 2017; Nadig et al., 2018; Siew et al., 2017; Wentz et al., 2012).

Limitations

The main limitation of this review is the streamlined process of a rapid review in comparison with a more comprehensive systematic review. However, rapid reviews allow for a more efficient dissemination of information to stakeholders, which is critical in the current health care climate. In addition, the majority of the articles were Level 3B and 4, although this was expected given the descriptive nature of our objectives. The majority of articles either excluded people with ID or with an IQ <70 or did not consider or mention the number of those with ID or an IQ <70, which is a limitation to generalizability across the autism spectrum. There is a high risk of bias because of convenience sampling, small sample sizes, and flaws in study protocol and development. However, the lower QoL experienced by the autistic population is clear and has been demonstrated in numerous studies.

Implications for Occupational Therapy Practice

Occupational therapy practitioners must promote the autonomy of their clients and justice in practice. This emphasizes the need for integration of QoL into assessment and intervention. It is clear that autistic people have reduced QoL in comparison with nonautistic population, which places them at risk for additional health concerns. Recommendations for occupational therapy practice include the following:

• Emphasize QoL when gathering an occupational profile, as recommended in the Occupational Therapy Practice Framework: Domain and Process—Fourth Edition (AOTA, 2020b), and whenever a plan of care or service reevaluation occurs.

• Use outcome measures that are current, evidence based, reliable, and valid and, whenever possible, provide an opportunity for self-reporting.

• Proactively consider QoL impacts when working with autistic individuals, groups, and populations, including consideration of client age and lifespan priorities.

• Implement evidence-based treatment approaches for improvement in QoL, such as mentoring, leisure programming, cognitive–behavioral therapy, and recreational activity; programs for education on the use of cognitive strategies (Let’s Get Organized), social skills and self-care programming (Girls Night Out), and online support and coaching programming may be suitable for clinical occupational therapy practice.

Conclusion

Future research must address the knowledge gap related to self-reported QoL of autistic girls and women to promote autonomy and beneficence. These findings are valuable for a variety of stakeholders, including health care professionals, policy makers, and funders. Health care providers (including occupational therapy practitioners) should place QoL at the forefront when caring for autistic clients. This QoL assessment should, whenever possible, be done with specific questions and use of self-report measures. Policy makers and funders need to consider the expansion of autistic supports throughout the lifespan, particularly for those people who may not qualify for established federal and state support systems.

Appendix

Table A.1.

Evidence Table for the Rapid Review of Quality-of-Life Discrepancies Among Autistic Adolescents and Adults

Author/Year	Level of Evidence, Study Design, and Risk of Bias	Participants, Inclusion Criteria, & Study Setting	Intervention & Control Groups	Outcome Measures	Results/Significance of Findings
Curtin et al. (2016)	Level 3B Pretest–posttest Risk of Bias Moderate	Participants N = 9 (M age = 15.4 yr, 22% female). Inclusion Criteria Previous diagnosis of Asperger syndrome or high-functioning autism by qualified professional Study Setting Local community centers	Intervention Group Mentorship program for 2 hr/wk over 6 mo, focusing on self-esteem, healthy relationships, independent living, community involvement, and education or vocation Control Group No control group	Quality of Life (self and proxy) • PedsQL™ Self-Esteem • RSES Social Skills • SWQ • Satisfaction questionnaires	Significant Findings None (no analysis performed) Nonsignificant Findings Improvements were made in self-esteem, social anxiety, and QoL.
Deserno et al. (2017)	Level 4 Case series Risk of Bias Low	Participants N = 2,341 (M age = 32.02; 28% female; IQ: 85% 70–130, 14% >130). Inclusion Criteria People with ASD ages ≥16 yr Study Setting Online	N/A	Well-Being (self and proxy) • Excerpt from the NVA	Significant Findings None (no analysis performed) Nonsignificant Findings Social satisfaction, societal contribution, resource access, self-reported IQ, living situation, level of daily activity, and happiness were influential in determining well-being. Self-reports did not significantly differ from proxy reports.
Franke et al. (2019)	Level 3B Case–control Risk of Bias Low	Participants N = 113 (M age = 45.62; 100% female). Inclusion Criteria Age 13–18 yr, diagnosed with autism (for autism group) Study Setting Online or paper forms	Case Group Autistic adolescents (M age = 14.98; 21.7% female) Control Group Nonautistic adolescents (M age = 14.96 yr; 53.73% female)	Life Satisfaction (self and proxy) • BMSLSS Assets • SEH–S	Significant Findings Parents and their autistic adolescents had correlated scores on the overall BMSLSS, persistence, school support, family coherence, peer support, empathy, and optimism. Autistic adolescents demonstrated decreased family, friend, self, and life satisfaction, in addition to overall BMSLSS scores. All but 2 assets were associated with life satisfaction in consideration of adolescents’ scores. Nonsignificant Findings None
García-Villamisar & Dattilo (2010)	Level 1B RCT Risk of Bias Low	Participants N = 71 (42% female) Inclusion Criteria Diagnosed with autism, Asperger syndrome, or other ASD by qualified professional Study Setting Learning center and within the community	Intervention Group (n = 37; M age = 31.49) Leisure Lifestyle Profile developed for customized intervention. 5 days/wk, 2 hr/day, for 1 yr. Included interacting with media, exercise, games, crafts, attending events, and other recreation activities. Needed to be accessible to the participants in terms of allowing for change as needed and including an active component. Control Group (n = 34; M age = 30.06 yr) Waitlist	Quality of Life (self) • QOL Questionnaire Stress • SSS	Significant Findings The intervention group had significant improvements in stress levels, total QOL, and scores on the Satisfaction and Competence/Productivity QOL Questionnaire subscales. Nonsignificant Findings Improvements occurred in empowerment–independence and social belonging–community integration QOL for the intervention group.
Grove et al. (2018)	Level 4 Case series Risk of Bias Low	Participants N = 687 (M age = 42.4 yr; 49% female). Inclusion Criteria Adults with a diagnosis of DSM–IV or DSM–5 ASD Study Setting Online	N/A	Special Interests • Informal questionnaire • SIMS Well-Being (self) • SWLS • Subjective Happiness Scale • Cantril ladder	Significant Findings Those with a current special interest rated their leisure satisfaction higher. Subjective well-being was associated with motivation to engage in special interests among autistic adults. Satisfaction with social contact and leisure was positively correlated with special interest engagement. Engagement with special interests at a high intensity (number of days or times per day) was negatively related to well-being findings. Men were more likely to have a special interest. Nonsignificant Findings None
Hesselmark et al. (2014)	Level 1B RCT Risk of Bias Low	Participants N = 68 (M age = 31.8 yr; 49% female) Inclusion Criteria Clinical diagnosis of ASD (confirmed by medical records), ADOS, and clinical interviews. No ID diagnosis. Study Setting Outpatient tertiary psychiatric clinic	Intervention 1: Cognitive–behavioral therapy (n = 34) 36 individual sessions specifically designed for patients with ASD focusing on self-esteem, ASD awareness, social contacts and everyday life, and psychological and physical health Intervention 2: Recreational activity (n = 34) Relied on structure and group setting only, without using any specific techniques	Quality of Life (self) • QOLI Self-Esteem/Other • RSES • SOC	Significant Findings All participants reported increased QoL at posttreatment with no difference between interventions, which was sustained at follow-up. Nonsignificant Findings None
Holmefur et al. (2019)	Level 3B Pretest–posttest Risk of Bias Moderate	Participants N = 55 (M age = 34.4 yr; 69% female). Inclusion Criteria Confirmed or suspected diagnosis of a mental, neurodevelopmental, or attention deficit disorder, absence of ID, and self-reported difficulties in time management Study Setting Outpatient psychiatric and habilitation settings	Intervention Group LGO–S Part 1, structured training in cognitive assistive techniques and strategies using trial-and-error learning; 10 weekly 90-min group sessions Control Group No control group	Executive Functioning • ATMS–S • WCPA (Swedish version) Life Satisfaction (self) • SDO–13	Significant Findings Significant improvements were made in time management, organization and planning skills, emotional regulation, and satisfaction with daily occupations. Time management improvements demonstrated stability at follow-up. Nonsignificant Findings None
Hong et al. (2016)	Level 4 Cohort Risk of Bias Low	Participants N = 60 (M age = 32 yr; 23% female), 30% with an ID. Inclusion Criteria In initial study phase, families with children with ASD ages ≥10 yr. This study occurred 15 yr later. Study Setting University	N/A	Quality of Life (self and proxy) • WHOQOL–BREF (with modifications) Life Experiences • 2 questions developed by researchers on experiences of being bullied Perception of Stress • PSS	Significant Findings Adults with ASD reliably rated their own individual QoL. QoL scores from adult self-reports had greater relation to maternal proxy report than to maternal report. Participants’ perceived stress and bullying frequency were associated with QoL. Independence in daily activities was positively correlated with subjective QoL. Nonsignificant Findings QoL domains based on self-report were not predicted by level of independence with daily activities and psychopathology. Mean differences were not observed between adult self-report and maternal proxy report on domains of the WHOQOL–BREF except social relationships.
Hull et al. (2019)	Level 3B Case–control Risk of Bias Low	Participants N = 832 (M age = 36.01 yr; 52% female). Inclusion Criteria Autism group had to be diagnosed by a professional Study Setting Online via Qualtrics	Case Group Autistic participants (n = 354) Male, n = 108 Female, n = 109 Other gender, n = 17 Not reported, n = 50 Control Group Nonautistic participants (n = 478) Male, n = 192 Female, n = 255 Other gender, n = 29 Not reported, n = 2	Social Camouflaging • CAT–Q Autistic Traits • BAPQ Mental Health • LSAS • PHQ–9 (autism only) • GAD–7 (autism only) General Well-Being (self) • WEMWBS	Significant Findings Total CAT–Q scores and Assimilation subscale scores were correlated with BAPQ, LSAS, WEMWBS, PHQ–9, and GAD–7 scores for both groups. In the autistic group, the Compensation subscale was correlated with total BAPQ, BAPQ pragmatic language, BAPQ rigidity, total LSAS, PHQ–9, and GAD–7. The autistic group also demonstrated a correlation between the CAT–Q Masking subscale and total LSAS, PHQ–9, and GAD–7. The nonautistic group demonstrated significant correlation between the CAT–Q and all other outcome measures. Nonsignificant Findings None
Jamison & Schuttler (2015)	Level 3B Case–control Risk of Bias Low	Participants N = 52 (100% female), none diagnosed with an ID Inclusion Criteria Female adolescents ages 14–19 yr. Autism group needed to have a DSM–IV–TR diagnosis from a psychiatrist, psychologist, developmental pediatrician, or interdisciplinary team; a reading level of Grade 4 or higher; and the ability to speak about one to two 2- to 3-word phrases per minute. Study Setting Paper forms	Case Group Autistic participants (n = 23; M age = 16.04 yr) Control Group Nonautistic participants (n = 29; M age = 16.75 yr)	Social Skills • SSIS Rating Scales Self-Perception • SPPA Quality of Life (self) • YQOL–R	Significant Findings Autistic female adolescents rated themselves as having lower QoL, social competence, and self-worth in comparison with peers. They also demonstrated more internalizing and externalizing behaviors than their peers. There was an inverse relationship between parents’ rating of autism severity and social competence. Nonsignificant Findings None
Jamison & Schuttler (2017)	Level 3B Pretest–posttest Risk of Bias Low	Participants N = 33 (autism, n = 28; M age = 15.97 yr; 100% female); included participants with both ID and DD Inclusion Criteria Diagnosis of autism provided by appropriate professional or diagnosis of DD with history of social skills difficulties; reading at or above fourth grade level; conversational speech, operationalized as ≥5 spontaneous sentences during a 10-min timespan Study Setting Community-based, varied on the basis of session content or requirements	Intervention Group Social skills and self-care program (Girls Night Out) focused on relating to others, self-care, and self-determination using same-age peers as mentors to develop skills. Intervention is a program with 1 2-hr session/wk for 12–16 wk. Groups include 4-6 participants and the same number of peer mentors. Each weekly session focuses on a different skill area. Control Group No control group	Social Skills • SSIS Self-Perception • SPPA Quality of Life (self) • YQOL–R	Significant Findings Adolescent reports on the SSIS (total and Empathy subscale) were significantly improved, along with total YQOL–R scores. Adolescents also reported significant decreases in internalizing symptoms after the intervention period. Nonsignificant Findings Improvements were noted in all areas of the SSIS (both adolescent and parent reports), SPPA, and YQOL–R subscales.
Kamio et al. (2013)	Level 4 Case series Risk of Bias Low	Participants N = 154 (M age = 27.6 yr; 20.1% female). Inclusion Criteria Adults ages ≥18 yr, ASD diagnosis, live in the community, use ≥1 supports during the study, staff selected those who they felt would be able to respond Study Setting Outpatient clinic	N/A	Quality of Life (self) • WHOQOL–BREF Current Family Support • Question developed by researchers	Significant Findings Higher QoL was indicated by having mother’s support, along with early diagnosis (before age 4). Poorer QoL was associated with aggressive behavior and comorbid psychiatric conditions. Psychological and social QoL was significantly lower in comparison with the nonautistic standardization sample. Nonsignificant Findings None
Knüppel et al. (2018)	Level 4 Case series Risk of Bias Low	Participants N = 1,738 (self-reports: n = 875, M age = 20.46 yr, 21.4% female; parent proxy reports: n = 863, M age = 20.8 yr, 17.3% female). Included people with ID: 10.7% for both groups. Inclusion Criteria Professional diagnosis of ASD before age 14 yr Study Setting Questionnaires	N/A	Quality of Life • INICO–FEAPS	Significant Findings Self-reported QoL scores were significantly higher when compared with proxy reports although, when considered by domain, scores were significantly higher only for social inclusion and interpersonal relationships. Regardless of report type, psychiatric comorbidity, sleeping difficulty, ID, maladaptive behavior, adaptive functioning, autism symptoms, main daytime activity, and residence were significantly associated with QoL. Nonsignificant Findings None
Leader et al. (2018)	Level 3B Case–control Risk of Bias Low	Participants N = 240 Inclusion Criteria hfASD group: needed to have a formal diagnosis non-hfASD group: did not need to have an autism diagnosis Study Setting Questionnaires	Case Group Autistic participants (n = 103; M age = 37.03; 64.1% female) Control Group Nonautistic participants (n = 137; M age = 20.31; 60.6% female) The groups differed in educational status, occupational status, and some types of psychopathology	Gelotophobia • Geloph <15> Quality of Life/Life Satisfaction (self) • WHOQOL–BREF • SWLS Social Components • Retrospective Bullying Questionnaire • ISEL–12 • SFQ	Significant Findings The hfASD group demonstrated higher prevalence of gelotophobia and past experiences of bullying. They also demonstrated lower levels of QoL, life satisfaction, social functioning, and perceived social support. Nonsignificant Findings None
Lin (2014)	Level 3B Case–control Risk of Bias Low	Participants N = 82 Inclusion Criteria ASD: diagnosed by registered psychiatrist using the DSM–IV–TR criteria, no ID Control: no known DD or ID Study Setting Questionnaires	Case Group Autistic participants (n = 41; M age = 26.9 yr, 26.8% female) Control Group Nonautistic participants (n = 41; M age = 26.9 yr, 26.8% female)	Quality of Life (self) • WHOQOL–BREF (Taiwanese) • Informal health and happiness assessment	Significant Findings Adults with ASD scored significantly lower in all domains than the non-ASD control group, as well as the Taiwanese health population reference group (standardization sample). Self-related health status and perceived happiness were correlated with increased QoL in both groups. Comorbid psychiatric disorders were correlated with decreased QoL in the ASD group. Nonsignificant Findings None
Lin & Huang (2019)	Level 3B Case–control Risk of Bias Low	Participants N = 151 (M age = 27.8 yr) Inclusion Criteria Professional DSM–5 autism diagnosis, no ID Study Setting Nonautistic group completed questionnaires on their own. Autism group completed questionnaires via interview with occupational therapist.	Case Group Autistic participants (n = 66; 35% female) Control Group Nonautistic participants (n = 85; 39% female)	Quality of Life (self) • WHOQOL–BREF (Taiwanese) Anxiety • Beck Anxiety Inventory (Chinese version) Loneliness/Social Isolation • ULS–8 (Chinese version) Sensory Processing • Adolescent/Adult Sensory Profile (Chinese version)	Significant Findings Autistic participants demonstrated significantly lower QoL in all areas and higher levels of autistic traits, anxiety, loneliness, and all areas of sensory processing except for sensory seeking. Higher levels of anxiety and sensory seeking were associated with decreased physical QoL. Comorbid psychiatric diagnosis, anxiety, loneliness, and sensory sensitivity were associated with decreased psychological QoL. Last, comorbid psychiatric diagnosis, autism traits, loneliness, sensory sensitivity, and anxiety were associated with decreased social QoL. Nonsignificant Findings Sensory seeking was higher for the autistic group.
Mahfouda et al. (2019)	Level 4 Cohort Risk of Bias Low	Participants N = 104 (M age = 14.62 yr; 76% female [birth-assigned sex], 22% female [gender identity]) Inclusion Criteria People with ASD ages ≥16 yr Study Setting Specialized service center	N/A	Core ASD Symptoms • SRS™–2 Quality of Life (self and proxy) • PedsQL	Significant Findings ASD significantly predicted internalizing behaviors and overall behavioral problems. In addition, ASD significantly predicted lower scores on the PedsQL. Nonsignificant Findings None
Mason et al. (2018)	Level 4 Cross-sectional Risk of Bias Low	Participants N = 370 (M age = 42 yr; 48% female) Inclusion Criteria Diagnosed with autism Study Setting Online/paper forms	N/A	ASD Symptoms • SRS–2 Adult Self-Report Quality of Life (self and proxy) • WHOQOL–BREF	Significant Findings Being female, symptom severity (per SRS–2), and mental health diagnoses were significant predictors of lower QoL on the WHOQOL–BREF. Autistic participants experienced statistically significant lower QoL in comparison with normative data. Nonsignificant Findings None
Mazurek (2014)	Level 4 Case series Risk of Bias Low	Participants N = 108 (M age = 32.4 yr; 47.2% female) Inclusion Criteria Adults with ASD, ages ≥18 yr, can complete measures independently, and previous professional ASD diagnosis Study Setting Online	N/A	Loneliness • ULS–8 Friendship • General questionnaire developed for this study • URCS Life Satisfaction (self) • SWLS Mental Health • RSES • PHQ–9, GAD–7	Significant Findings Loneliness was significantly correlated with increased depression, increased anxiety, decreased life satisfaction, and decreased self-esteem. Decreased loneliness was associated with higher quantity and quality of friendships. Number of friendships had greater effects in predicting anxiety, depression, and self-esteem than did loneliness. Nonsignificant Findings None
McConachie et al. (2018)	Level 4 Cross-sectional Risk of Bias Low	Participants N = 309 (M age = 37.35; 49.5% female) Inclusion Criteria Diagnosed with autism Study Setting Online/paper forms	N/A	Quality of Life (self and proxy) • WHOQOL–BREF • WHO Disabilities Module • ComQoL–A5 • ASQoL Anxiety and Depression • HADS Barriers • CHIEF–SF Social Support • ISEL–12	Significant Findings Autistic participants demonstrated decreased QoL in comparison with the general population. The WHOQOL–BREF is correlated with HADS depression and anxiety scores, barriers from the CHIEF–SF, and social support from the ISEL–12. QoL scores from the WHOQOL–BREF were correlated with ASQoL scores. Nonsignificant Findings None
McDonald (2017)	Level 4 Cross-sectional Risk of Bias Low	Participants N = 1,139 (M age = 35.09 yr [female] and 32.01 [male]; 59.1% female) Inclusion Criteria Age ≥18 yr, own legal guardian, identify with or have an ASD diagnosis Study Setting Online forms	N/A	Autism Symptoms • ASIS • SS Self-Esteem • RSES Quality of Life (self) • AAQOL • FutQOL • General questions on loneliness and self-care	Significant Findings ASIS Positive Difference scores were significantly correlated with AAQOL Life Outlook, Psychological Health, and Relationships; all components of the SS; the RSES; the FutQOL; and loneliness questions. ASIS Context Dependent scores demonstrated a significant relationship with AAQOL Life Productivity, SS Disclosure, and general self-care questions. ASIS Spectrum Abilities scores were correlated with AAQOL Life Outlook, SS Positive Aspects, RSES, and FutQOL. ASIS Changeability was significantly correlated with all components of the AAQOL, SS Discrimination, SS Positive Aspects, RSES, FutQOL, and self-care questions. Nonsignificant Findings None
Nadig et al. (2018)	Level 1B RCT Risk of Bias Low	Participants N = 26 Inclusion Criteria Ages 18–32 yr, diagnosed with ASD (confirmed in study), no ID, able to communicate in English, and not participating in another transition support service Study Setting University	Intervention Group (n = 17; M age = 20.65 yr, 41.2% female) 5 modules of curriculum materials in domains of social communication, self-determination, and working with others. 4–6 participants and 2 facilitators (graduate students) met for 10 weekly 2-hr-long sessions Control Group (n = 9; M age = 23.3 yr; 22.2% female) Waitlist	Quality of Life (self) • QOL (abridged version) Self-Determination • SDS	Significant Findings None; conservative analysis approach used rather than determination of statistical significance. Nonsignificant Findings Positive intervention effects were observed on self-reported QoL. Positive effects were observed on the SDS Interpersonal Cognitive Problem-Solving subdomain, with the intervention group scoring higher than the control group.
Pearlman-Avnion et al. (2017)	Level 3B Case–control Risk of Bias Low	Participants N = 31 (M age = 27.79; 35.5% female) Inclusion Criteria Formal autism diagnosis Study Setting Online forms	Case Group (relationships; n = 14) Control Group (nonrelationships; n = 17)	Quality of Life (self) • QOL (Hebrew) Sexual Well-Being • Sexual Well-Being Questionnaire (Hebrew); only 23 participants answered the Sexual Well-Being Questionnaire	Significant Findings Autistic people in relationships demonstrated higher scores on social belonging and community inclusion compared with those not in relationships. Productive capacity (QOL) and sexual well-being were correlated for the relationships group. Productive capacity (QOL) had an inverse relationship with sexual preoccupation (worries). For the nonrelationships group, satisfaction and sexual well-being were significantly related. Nonsignificant Findings None
Rodgers et al. (2018)	Level 3B Case–control Risk of Bias Low	Participants N = 253 (M age = 25.17 yr; 77% female) Inclusion Criteria Age ≥18 yr, successful completion of online surveys, recruited from autism-related websites Study Setting Online forms	Case Group (high ASDC; n = 46) AQ score ≥26 or self-reported ASD diagnosis (n = 21) Control Group (low ASDC; n = 207) AQ score < 26	Well-Being (self) • Abbreviated Ryff Psychological Well-Being Scale • Life Engagement Test • SWLS Mental Health • RSES • IPIP Anxiety Facet scale • CES–D	Significant Findings Experiencing higher characteristics of autism was correlated with lower well-being, which is mediated by personality. Nonsignificant Findings None
Siew et al. (2017)	Level 3B Pretest–posttest Risk of Bias Low	Participants N = 10 (M age = 18.0 yr; 30% female) Inclusion Criteria Self-reported DSM–IV diagnosis of autistic disorder or similar, mentee participants in first-semester mentorship program at university Study Setting Face to face (location not provided)	Intervention Group Curtin Specialist Mentoring Program, a program for students with ASD, which provides 1:1 peer mentoring designed around the participant’s goals. Peer matches meet weekly for approximately 1 hr. Control Group No control group	Well-Being and Communication (self) • AMAS–C • SPS • SCAM • SPCC • PRCA–24 • General social validity of program	Significant Findings There was a significant improvement in perceived social support and a decrease in general communication apprehension. Nonsignificant Findings None
Wentz et al. (2012)	Level 3B Pretest–posttest Risk of Bias Low	Participants N = 12 (including 2 dropouts; M age = 20.1 yr; 41% female, 3 participants with borderline IQ) Inclusion Criteria Ages 15–26 yr, Internet access at home, diagnosis of ADHD or ASD Study Setting Not reported	Intervention Group Internet-based support and coaching program for 8 wk after initial in-person meeting with a coach. Participants offered 14 chat sessions and 2 visits over this 8-wk period. Control Group No control group	Quality of Life (self) • MANSA Mental Health • HADS • SOC • RSES • Interviews	Significant Findings After the program, participants reported improvements in self-esteem, sense of coherence, and QoL. Nonsignificant Findings Depressive symptoms decreased.
White et al. (2018)	Level 4 Cross-sectional Risk of Bias Low	Participants N = 30 (M age = 21.27 yr; 33% female; all with average IQ) Inclusion Criteria Diagnosis of ASD based on ADOS–2 or SCQ Lifetime form, no ID, ages 18–30 yr, English speaking Study Setting University	N/A	Quality of Life (self) • QOL Self-Determination • SDS • AIR–S	Significant Findings SDS and AIR–S scores were positively correlated. Both tests’ total scores were correlated with reported QOL scores. The SDS Autonomy, Psychological Empowerment, Self-Realization subscales and AIR–S Capacity component were correlated with QOL scores. Self-determination was thus found to be a critical predictor of QOL scores. Nonsignificant Findings None

Note. AAQOL = Adult ADHD Quality of Life Scale; ADHD = attention deficit hyperactivity disorder; ADOS = Autism Diagnosis Observation Schedule; AIR–S = American Institutes for Research Self-Determination Scale–Student Form; AMAS–C = Adult Manifest Anxiety Scale™–College Version; AQ = Autism Quotient; ASD = autism spectrum disorder; ASIS = Autism Spectrum Identity Scale; ASQoL = Autism Spectrum Quality of Life Module; ASDC = ASD characteristics; ATMS–S = Asssessment of Time Management Skills, Swedish version; BAPQ = Broad Autism Phenotype Questionnaire; BMSLSS = Brief Multidimensional Students’ Life Satisfaction Scale; CAT–Q = Camouflaging Autistic Traits Questionnaire; CES–D = Center for Epidemiological Studies Depression Scale; CHIEF–SF = Craig Hospital Inventory of Environmental Factors–Short Form; ComQoL–A5 = Comprehensive Quality of Life Scale–Adult; DD = developmental disability; DSM–IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.); DSM–IV–TR = Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.); DSM–5 = Diagnostic and Statistical Manual of Mental Disorders (5th ed.); FutQOL = Future Quality of Life; GAD–7 = Generalized Anxiety Disorder 7; HADS = Hospital Anxiety and Depression Scale; hfASD = high-functioning ASD; ID = intellectual disability; INICO–FEAPS = Evaluación Integral de la Calidad de Vida de personas con Discapacidad Intelectual o del Desarrollo; IPIP = International Personality Item Pool; ISEL–12 = Interpersonal Support Evaluation List–12; LGO–S = Let’s Get Organized, Swedish version; LSAS = Leibowitz Social Anxiety Scale; MANSA = Manchester Short Assessment of Quality of Life; N/A = not applicable; NVA = Nederlandse Vereniging voor Autisme; PedsQL = Pediatric Quality of Life Scale; PHQ–9 = Patient Health Questionnaire; PRCA–24 = Personal Report of Communication Apprehension; PSS = Perceived Stress Scale; QoL = quality of life; QOL = Quality of Life Questionnaire; QOLI = Quality of Life Inventory; RCT = randomized controlled trial; RSES = Rosenberg Self-Esteem Scale; SCAM = Situational Communication Apprehension Measure; SCQ = Social Communication Questionnaire; SDO–13 = Satisfaction with Daily Occupations; SDS = The ARC’s Self-Determination Scale; SEH–S = Social Emotional Health Survey–Secondary; SFQ = Social Functioning Questionnaire; SIMS = Special Interest Motivation Scale; SOC = Sense of Coherence Scale; SPCC = Self-Perceived Communication Competence Scale; SPPA = Harter’s Self-Perception Profile for Adolescents; SPS = Social Provisions Scale; SRS–2 = Social Responsiveness Scale, Second Edition; SS = Stigma Scale; SSIS = Social Skills Improvement System Rating Scales; SSS = Stress Survey Schedule for Persons with Autism and Other Pervasive Developmental Disabilities; SWQ = Social SWLS = Satisfaction With Life Scale; SWQ = Social Worries Questionnaire; ULS–8 = UCLA Loneliness Scale–8; URCS = Unidimensional Relationship Closeness Scale; WCPA = Weekly Calendar Planning Activity; WEMWBS = Warwick–Edinburgh Mental Wellbeing Scale; WHO = World Health Organization; WHOQOL–BREF = WHO Quality of Life–Brief Version; YQOL–R = Youth Quality of Life Instrument–Research Version.

Copyright © 2021 by the American Occupational Therapy Association. This table may be freely reproduced for personal use in clinical or educational settings as long as the source is cited. All other uses require written permission from the American Occupational Therapy Association. To apply, visit www.copyright.com.

Suggested citation: Skaletski, E. C., Bradley, L., Taylor, D., Travers, B. G., & Bishop, L. (2021). Quality-of-life discrepancies among autistic adolescents and adults: A rapid review (Table A.1). American Journal of Occupational Therapy, 75, 7503180090. https://doi.org/10.5014.ajot.2021.046391

Table A.2.

Risk-of-Bias Table for Randomized and Nonrandomized Controlled Trials

Citation	Selection Bias (Risk of Bias Arising From Randomization Process)			Performance Bias (Effect of Assignment to Intervention)		Detection Bias		Attrition Bias: Incomplete Outcome Data (Data for All or Nearly All Participants)	Reporting Bias: Selective Reporting (Results Being Reported Selected on the Basis of the Results)	Overall Risk of Bias (Low, Moderate, High)
	Random Sequence Generation	Allocation Concealment (Until Participants Enrolled and Assigned)	Baseline Differences Between Intervention Groups	Blinding of Participants During the Trial	Blinding of Study Personnel During the Trial	Blinding of Outcome Assessment: Self-Reported Outcomes	Blinding of Outcome Assessment: Objective Outcomes (Aware of Intervention Received?)
García-Villamisar & Dattilo (2010)	?	?	+	−	−	?	+	?	?	L
Hesselmark et al. (2014)	+	?	+	+	?	+	−	+	+	L
Nadig et al. (2018)	+	+	+	−	?	+	+	+	+	L

Note. Categories for risk of bias are as follows: + = low risk of bias, ? = unclear risk of bias, − = high risk of bias; L = low risk of bias. Scoring for overall risk-of-bias assessment is as follows: 0–3 minuses, low risk of bias; 4–6 minuses, moderate risk of bias; 7–9 minuses, high risk of bias. Risk-of-bias table format adapted from Higgins, J. P. T., Sterne, J. A. C., Savovic, J., Page, M. J., Hrobjartsson, A., Boutron, I., . . . Eldridge, S. (2016). A revised tool for assessing risk of bias in randomized trials. Cochrane Database of Systematic Reviews, 2016(10, Suppl. 1), 29–31. https://doi.org/10.1002/14651858.CD201601

Copyright © 2021 by the American Occupational Therapy Association. This table may be freely reproduced for personal use in clinical or educational settings as long as the source is cited. All other uses require written permission from the American Occupational Therapy Association. To apply, visit www.copyright.com.

Suggested citation: Skaletski, E. C., Bradley, L., Taylor, D., Travers, B. G., & Bishop, L. (2021). Quality-of-life discrepancies among autistic adolescents and adults: A rapid review (Table A.2). American Journal of Occupational Therapy, 75, 7503180090. https://doi.org/10.5014.ajot.2021.046391

Table A.3.

Risk-of-Bias Table for Before–After (Pretest–Posttest) Studies With No Control Group and Other Designs

Citation	Study Question or Objective Clear	Eligibility or Selection Criteria Clearly Described	Participants Representative of Real-World Patients	All Eligible Participants Enrolled	Sample Size Appropriate for Confidence in Findings	Intervention Clearly Described and Delivered Consistently	Outcome Measures Prespecified, Defined, Valid/Reliable, and Assessed Consistently	Assessors Blinded to Participant Exposure to Interventions	Loss to Follow-Up After Baseline 20% or less	Statistical Methods Examined Changes in Outcome Measures From Before to After intervention	Outcome Measures Were Collected Multiple Times Before and After Intervention	Overall Risk-of-Bias Assessment (Low, Moderate, High Risk)
Curtin et al. (2016)	Y	Y	N	N	N	Y	Y	N	NR	Y	Y	M
Deserno et al. (2017)	Y	Y	N	N	Y	NR	Y	NR	NR	NR	NR	L
Franke et al. (2019)	Y	Y	Y	Y	N	NR	Y	NR	Y	Y	NR	L
Grove et al. (2018)	Y	N	N	N	Y	NR	Y	NR	NR	NR	NR	L
Holmefur et al. (2019)	Y	Y	N	N	N	Y	Y	N	Y	Y	Y	M
Hong et al. (2016)	Y	Y	N	N	Y	NR	Y	NR	NR	NR	NR	L
Hull et al. (2019)	Y	Y	Y	Y	Y	NR	Y	NR	Y	Y	NR	L
Jamison & Schuttler (2015)	Y	Y	Y	Y	N	NR	Y	NR	Y	Y	NR	L
Jamison & Schuttler (2017)	Y	Y	Y	NR	Y	NR	Y	NR	Y	Y	NR	L
Kamio et al. (2013)	Y	Y	N	N	Y	NR	Y	NR	NR	NR	NR	L
Knüppel et al. (2018)	Y	Y	Y	NR	Y	NR	NR	NR	NR	NR	NR	L
Leader et al. (2018)	Y	Y	NR	Y	Y	NR	Y	NR	NR	Y	NR	L
Lin (2014)	Y	Y	Y	Y	Y	NR	NR	NR	NR	NR	NR	L
Lin & Huang (2019)	Y	Y	Y	Y	Y	NR	Y	NR	NR	Y	NR	L
Mahfouda et al. (2019)	Y	Y	N	N	Y	NR	Y	NR	NR	NR	NR	L
Mason et al. (2018)	Y	Y	Y	Y	Y	NR	Y	NR	Y	Y	NR	L
Mazurek (2014)	Y	Y	N	NR	Y	NR	Y	NR	NR	NR	NR	L
McConachie et al. (2018)	Y	Y	Y	Y	Y	NR	Y	NR	Y	Y	NR	L
McDonald (2017)	Y	Y	Y	N	Y	NR	Y	NR	NR	Y	NR	L
Pearlman-Avnion et al. (2017)	Y	Y	NR	N	N	NR	Y	NR	NR	Y	NR	L
Rodgers et al. (2018)	Y	Y	N	NR	Y	NR	Y	NR	NR	Y	NR	L
Siew et al. (2017)	Y	Y	Y	Y	N	N	Y	N	Y	Y	NR	L
Wentz et al. (2012)	Y	Y	Y	Y	N	N	Y	N	Y	Y	NR	L
White et al. (2018)	Y	Y	N	Y	N	NR	Y	NR	NR	Y	NR	L

Note. N = no; NR = not reported; Y = yes. Scoring for overall risk-of-bias assessment is as follows: 0–3 N, low risk of bias (L); 4–8 N, moderate risk of bias (M); 9–11 N, high risk of bias. Risk-of bias table format adapted from National Heart Lung and Blood Institute. (2014). Quality assessment tool for before-after (prepost) studies with no control group. Retrieved from https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools

Copyright © 2021 by the American Occupational Therapy Association. This table may be freely reproduced for personal use in clinical or educational settings as long as the source is cited. All other uses require written permission from the American Occupational Therapy Association. To apply, visit www.copyright.com.

Suggested citation: Skaletski, E. C., Bradley, L., Taylor, D., Travers, B. G., & Bishop, L. (2021). Quality-of-life discrepancies among autistic adolescents and adults: A rapid review (Table A.3). American Journal of Occupational Therapy, 75, 7503180090. https://doi.org/10.5014.ajot.2021.046391