Social support and links to quality of life among middle-aged and older autistic adults

There is extensive literature describing the positive impact of social support on mental and physical health and on quality of life (QoL) in the general population and those with psychiatric disorders (Brookes et al., 2012; Campos et al., 2014; Chachamovich et al., 2008). Social support is often described as a supportive social network. However, it is multidimensional and reflects access to interpersonal resources including social interactions, perceived emotional support, and practical or instrumental support (Kelly et al., 2017; Sahin et al., 2019; Tyler, 2006). High overall social support has been shown to have both direct and indirect positive effects on mental and physical health outcomes. Direct effects link increased social support with reduced depression, suicidal thoughts, and time to mortality; whereas indirect effects may act by reducing depression or stress, which in turn positively impacts QoL, mood, and suicidality (Holt-Lunstad et al., 2010; Pan et al., 2020; Sahin et al., 2019). Research has shown that as individuals reach later life, subjective social support reduces, and this reduction has a significant negative impact on mood and suicidality (Hybels et al., 2011; Pan et al., 2020). Given that difficulties in social communication and interactions are a core diagnostic feature of autism spectrum disorder (ASD), it has been questioned whether the impact of variable levels of social support is the same for autistic versus non-autistic individuals (Hedley et al., 2017). To date, few studies have examined social support among autistic adults, and to our knowledge, no study has examined social support among middle-aged or older autistic adults.

Studies examining social support among autistic adolescents and young adults have identified lower rates of overall subjective social support compared to comparison groups of both neurotypical individuals and those with specific reading disability (Bishop-Fitzpatrick et al., 2018; Humphrey & Symes, 2010). In a sample of 41 autistic young and middle-aged adults (mean age = 30 years), subjective support from friends was rated as low compared to both attention-deficit hyperactivity disorder (ADHD) and neurotypical comparison groups, but no differences were observed for ratings of support from family or a significant other (Alvarez-Fernandez et al., 2017). Given the potential buffering effects associated with good social support in the general population, low social support could be an added risk for already poor QoL outcomes for autistic adults (Bennett et al., 2005; Jennes-Coussens et al., 2006; Kamio et al., 2012; Kamp-Becker et al., 2010; van Heijst & Geurts, 2014) compared to neurotypical individuals. Subjective QoL can be defined as the individual’s perception of their satisfaction with life within the context of their own society and value system, and relating to the expectations and concerns of the individual (Harper & The WHOQOL Group, 1998). As such, QoL (rated subjectively by the individual) is often used as a relevant outcome measure across different cultures, societies, and groups, including autistic individuals (Burgess & Gutstein, 2007).

Examining the association between social support and psychosocial and health-related outcomes in young and middle-aged autistic adults (mean age = 24 years), Bishop-Fitzpatrick and colleagues (2018) found a positive association between subjective social support and overall QoL. However, the source and type of social support may also impact results. Greater perceived social support from family and friends (but not significant others) was associated with higher mental health–related QoL among autistic adults (mean age = 31 years; Khanna et al., 2014). An examination of the different components of social support suggests that perceived informal support (e.g. support from friends or relatives) and number of unmet formal support needs (number of domains where support is needed but neither informal nor formal support from professionals is available) were (respectively positively and negatively) associated with overall QoL (Renty & Roeyers, 2006). A further study found that more tangible material support was associated with lower depression among autistic adults (Hedley et al., 2017). In contrast, perceived support by having someone to talk through problems or having people available to share activities had no effect on depression in this sample. It remains unclear to what extent the type of social support impacts different aspects of mental health or QoL among autistic adults.

Poor QoL among autistic people has been reported in specific domains such as physical health (Kamp-Becker et al., 2010; Khanna et al., 2014) or psychological and social aspects (Kamio et al., 2012), as well as in overall QoL (Bennett et al., 2005; Bishop-Fitzpatrick et al., 2018; van Heijst & Geurts, 2014). However, it is worth noting that not all autistic people report poor QoL (Oakley et al., 2021). QoL has been reported to vary within the autistic population and across domains of QoL; therefore, caution may be needed when considering group-level comparisons (Oakley et al., 2021). Better subjective QoL among younger autistic adults is associated with factors such as having good social and practical (professional or family) support (Kamio et al., 2012; Leader et al., 2021; Mason et al., 2018; Renty & Roeyers, 2006), being employed (Mason et al., 2018), having better sleep quality (Leader et al., 2021), and better daily living skills (Kamp-Becker et al., 2010). However intelligence quotient (IQ), age, or severity of autistic traits have been found to not be associated with QoL (Kamp-Becker et al., 2010; Leader et al., 2021; van Heijst & Geurts, 2014). Therefore, these findings suggest that both greater interpersonal social support and practical (instrumental, tangible, and material) support can have independent positive impacts on QoL for autistic young adults. Whether different aspects of social support similarly impacts QoL among middle-aged and older autistic adults has not yet been explored.

This study sought to examine social support and its associations with QoL among middle-aged and older autistic adults (40+ years of age) for the first time. We hypothesized that social support would be associated with self-reported QoL, even after accounting for variables, such as demographic, physical health, anxiety, and depression. We explored whether the pattern of associations was the same or different for aspects of QoL. Owing to the limited previous literature examining QoL and social support in middle-aged and older autistic adults, we characterized the data by decades (reporting data for those aged 40–49 years; 50–59 years, etc) and explored the data for possible age-effects.

This study supported the hypotheses and revealed significant associations between different aspects of QoL and social support in middle-aged and older autistic adults for the first time. In keeping with the literature from both non-autistic older adults and autistic young adults, better QoL was associated with greater social support. One of the unique features of this study was that it examined different aspects of QoL (physical, psychological, social and environmental, and autism-specific) as well as different aspects of social support (subjective, social interactions, and instrumental). Thus, this study allowed us to examine the unique patterns of associations between different aspects of QoL and different facets of social support and to examine associations in a group of middle-aged and older autistic adults.

In keeping with the hypotheses, social support was significantly associated with each aspect of QoL, even after accounting for demographic, health and mental health factors. The association between social support and each aspect of QoL varied by domain and magnitude (standardized Beta range = .100–.385). However, regression analyses also suggested distinct patterns of importance for different aspects of social support for each QoL measure. All components of QoL (physical, psychological, social relationships, environmental, and autism-specific) were associated with subjective social support and self-reported depression symptomatology. In addition, Physical and Psychological QoL were associated with the number and QoL interactions, whereas Social and Environmental QoL were associated with instrumental support. Variance in ASQOL was additionally associated with anxiety symptomatology and instrumental support. It is worth noting that a large proportion of variance in QoL remained unaccounted for within each model. Despite demographic, mental and physical health and social support significantly contributing to the models, other, as yet unidentified, variables important for QoL are not included in the models.

These results demonstrate the importance of perceived or subjective social support for middle and older age autistic adults and are consistent with findings from previous studies of younger autistic adults. In one study, both perceived stress and perceived social support explained a significant proportion of the variance in overall QoL for young and middle-aged autistic adults (Bishop-Fitzpatrick et al., 2018). Other studies suggested that perceived support from family and friends or maternal support specifically are significantly associated with psychological, social and health-specific QoL for young autistic adults (Kamio et al., 2012; Khanna et al., 2014). However, not all studies find this pattern of associations. One study examining the impact of friendship on outcome variables found no association between number of friends and life satisfaction (although more friends was associated with lower symptoms of anxiety and depression; Mazurek, 2013). It is worth noting that satisfaction with friendships (i.e. a subjective assessment of friendship quality) may be more important than the number of friends (i.e. a simple quantification of the number of friendships). To our knowledge only one study has examined the impact of both perceived and actual support on overall QoL in autistic adults. Support characteristics overall (perceived social support, received formal, and informal support) explained a significant proportion of the variance in overall QoL among younger autistic adults (Renty & Roeyers, 2006). However, post hoc analyses demonstrated that perceived informal support (akin to subjective social support in the current study) and unmet actual support needs were significantly associated with QoL, whereas received practical support was not in that sample. In this study, instrumental support significantly contributed to explaining the variance in Social, Environmental, and ASQOL. It is worth noting that the association between perceived social support and QoL is also recognized in the neurotypical aging literature (Hajek et al., 2016; LaRocca & Scogin, 2015; Sahin et al., 2019). In one study, perceived social support alone explained 22.1% of the variance in overall QoL, although other aspects of mental and physical health were not assessed (Sahin et al., 2019), perhaps overestimating the independent contribution of social support to QoL.

In keeping with studies in both younger autistic adults and non-autistic adults across adulthood, mental health factors such as presence of anxiety and depression symptoms were also associated with QoL (Chachamovich et al., 2008; Kamio et al., 2012; Layte et al., 2013; Mason et al., 2018, 2019). Previous studies have found that psychiatric comorbidities were associated with different aspects of QoL for young and middle-aged autistic adults (Kamio et al., 2012; Mason et al., 2018). For example, one previous study of autistic adults found that self-reported depression and anxiety symptoms were significantly associated with Physical, Psychological, and Environmental QoL, while Social QoL was only associated with depression symptoms (Mason et al., 2019). Studies in older neurotypical adults show a similar pattern of results, with mental health factors, and depression specifically (even at low levels), significantly impacting the QoL of older neurotypical adults (Chachamovich et al., 2008; Kisvetrova et al., 2021; Layte et al., 2013).

Sex assigned at birth contributed to the model explaining Physical and Environmental QoL. This reflected the finding that males reported higher Physical and Environmental QoL compared to females. This pattern of better self-reported QoL among males compared to females is frequently observed among neurotypical older adults (Campos et al., 2014; Gallicchio et al., 2007; Rollero et al., 2014 but see Kirchengast & Haslinger, 2008 for higher QoL in females).

There is little evidence of age-differences in QoL or social support measures among autistic adults in this study. This is somewhat discrepant from findings in non-autistic older adults. Previous research has suggested that non-autistic older adults may have better QoL compared to middle-aged adults, although in later old-age QoL has been shown to be reduced, possibly related to declines in health (Diehr et al., 2013; Raggi et al., 2016). One study has suggested that age-effects on QoL may be curvilinear with a person’s QoL peaking in their late sixties and declining thereafter (Layte et al., 2013). Social support has also been shown to change with age among non-autistic adults, although the pattern of change is complex. Some studies suggest that the nature (friends vs family), but not the size, of social support networks may change with age, whereas other studies have shown increased social isolation with aging, and different trajectories at different age ranges (Czaja et al., 2018; Field & Minkler, 1988; van Tilburg, 1998). The results in this study suggest few age-effects, although it is unclear if this may reflect a different (less negative) pattern of age-effects for middle-aged and older autistic adults or some effect of sampling differences in either the recruitment of or diagnostic criteria applied to autistic adults of different ages. Ultimately, to answer these important developmental questions, longitudinal data during middle and older adulthood in ASD are needed in order to evaluate changes in QoL or social support, including evaluating the potential for both linear and non-linear trajectories of change.

This study should be considered with certain strengths and limitations in mind. Caution should be used when considering results relating to ASQOL, as this measure was not administered alongside the Disability module of the WHOQOL-BREF. Furthermore, a recent paper has suggested that response to items on the ASQOL may differ by sex (Williams & Gotham, 2021), meaning that ASQOL results here relating to sex differences should be considered with caution. However, the WHOQOL-BREF has been validated for use with autistic adults (McConachie et al., 2018) and has evidence of good factor structure, fit and test–retest validity, meaning the results are likely to be robust. The models described here do not fully explain the variance in QoL, and future studies will need to consider and investigate additional contributing variables. The study relies on self-report online data, which may increase sampling bias. However, it is critically important to acknowledge that subjective QoL is based on an individual’s perception and therefore self-report is not only appropriate, but also preferred, whenever possible. The cognitive demands of the survey and the participants’ generally high education level suggest that this sample has abilities within the normal range or higher and, therefore, does not fully reflect the experiences of all autistic adults (e.g. those with co-occurring intellectual disability). The SPARK sample also has limited ethno-racial representation, although it has been shown that self-disclosure of diagnoses in this sample is accurate (Fombonne et al., 2021). However, the sample also includes a large number of middle-aged and older autistic adults, particularly those assigned female at birth, both of which are under-studied groups. To our knowledge, this is the first study to examine the impact of social support on QoL among middle-aged and older autistic adults.

In conclusion, this study of 388 middle-aged and older autistic adults found that social support, in particular, subjective social support were associated with different aspects of QoL. Social support remained significantly associated with QoL even after accounting for the effects of demographic and health factors, and symptoms of anxiety and depression. Results suggest that different aspects of social support (instrumental, subjective, and social interactions) are important for outcomes for middle-aged and older autistic adults. Therefore interventions bolstering different forms and types of supports may cascade to benefit older autistic adults. Further longitudinal studies are required to explore the impact of common age-related changes, such as size, content, quality, and arrangement of social and other support networks on outcomes for autistic adults.

The authors are grateful to all of the families in SPARK, the SPARK clinical sites, and SPARK staff. They appreciate obtaining access to recruit participants through SPARK research match on SFARI Base.