Sociodemographic and clinical indicators associated with quality of life among parents of autistic children

Abstract

Background

Evidence suggests that parents/carers of autistic children have lower subjective physical and mental health which in turn can affect their overall quality of life (QoL). The aim of this study was to determine the relationship between the behavioural and emotional profile of autistic preschool children, parental stress, and an Autism (ASD) specific measure of the parental QoL using a sociodemographic lens.

Methods

A secondary analysis of the data collected from parents of autistic children from six Autism Specific Early Learning and Care Centres (ASELCCs) across six states in Australia. The standardised Quality of Life in Autism scale (QoLA) scale was used as the primary outcome to ascertain the QoL of parents/carers. Primary exposure included child’s autistic traits as well as cognitive, adaptive, and behavioural profile; parental stress; and key sociodemographic factors. Multivariable linear regression analyses were used to determine whether the sociodemographic factors and child’s autistic traits were significantly associated with parental QoL whilst adjusting for key sociodemographic factors (for the latter).

Results

Among a sample of 518 participants, findings of the regression analyses showed that sibling’s ASD diagnosis and carer’s disability status were negatively associated with parental QoL (Part A) whereas only sibling’s ASD diagnosis was negatively associated with parental QoL (Part B). Additionally, higher parental stress levels, child’s internalising, externalising, repetitive behaviours, and communication difficulties were negatively associated with both parental QoL subscales whereas greater adaptive functioning among autistic children was positively linked to better parental QoL.

Conclusion

Findings indicate that a child’s autism specific traits as reported by parents have significant impacts on their QoL. Hence, targeted supports in these areas for families could be expected to have benefits not only for the child’s outcomes but also for parental QoL.

Introduction

Parenting is a rewarding, albeit challenging role as caregivers navigate how to best meet the needs of children daily. Parenting autistic children presents unique challenges, with significant variation in the presentation of autism spectrum disorder (ASD), hitherto referred to as autism. Autism is a complex lifelong neurodevelopmental condition that emerges in childhood and is characterised by varied levels of social and communication difficulties, and the presence of restricted, repetitive behaviours, interests and/or activities [1]. While some parents face greater challenges due to difficulties with their children’s limited verbal communication skills, others may face greater challenges associated with the heightened externalising behaviours and social difficulties [2]. In the past two decades, the prevalence of childhood autistm has increased, with recent estimates indicating the prevalence to be 1 in 36 children [3]. Understanding the impact that raising an autistic child has on parenting, particularly in the early years and the impact on parental QoL may help tailor appropriate early supports for both children and parents.

There are many factors that underlie the challenges faced by parents of autistic children [4], and previous studies have examined how both intrinsic and extrinsic factors impact these parents. Studies have shown that parents’ quality of life (QoL) is impacted by level of social support [5], perception of daily hassles [6, 7], parental mental health [6, 8], and parenting stress [9,10,11,12,13]. More importantly, a number of clinical characteristics have been associated with poorer parental QoL in parents raising autistic children, compared to parents of children without autism [13, 14]. Parents of autistic children report higher stress [7,8,9, 13, 14], poorer sleep quality and increased fatigue [15], heightened anxiety [10, 14, 16] and increased depressive symptoms [6, 8, 10, 14, 17].

Given the complexities associated with raising an autistic child, much previous research has focused on identifying the extent to which factors negatively impact on parental QoL or wellbeing. Subjective measures of self-esteem, social support and life satisfaction reported by a sample of Chinese parents raising 5 to 12 year old autistic children indicated poorer self-esteem, social support and life satisfaction in comparison to the same measures completed by demographic-matched parents raising children without autism [18]. Cross-sectional assessment of maternal stress and family QoL in Australian parents has further replicated international findings that mothers of autistic children report higher levels of stress and lower family QoL across childhood [12]. More recently, cross sectional evaluation of clinical indicators in a large sample of children diagnosed with autism and attending an outpatient clinic found that mothers of children with more autism traits have significantly higher levels of stress, anxiety and depression than fathers [10]. Notably, higher education level (college or above) provided a protective factor against maternal stress whereas having an only child provided a protective factor against paternal stress [10]. While the majority of studies continue to demonstrate that perceived stress contributes to poorer QoL in parents of autistic children [e.g., 7–13, 14, 23], most focus on school aged children and are limited in the extent to which other sociodemographic factors were assessed. Furthermore, all of these studies have reported on parental QoL using generic measures which do not accurately capture the unique challenges that parents of autistic children encounter– a key objective of this study.

Other studies have focussed on the associations that child characteristics have on perceived parenting capacity and QoL. More autism traits [5, 6, 19] and difficulties adapting in social settings [20, 21] and the attribution of parental health related problems associated with raising an autistic child [22] have all been shown to exacerbate parenting challenges with consequent negative impact on parenting experiences. Indeed, child characteristics including age, autism traits, adaptive behaviour profile and the intellectual ability of the child have been predicted to impact the level of parental stress in parents of preschool autistic children [23]. Collectively, these studies demonstrate that parenting autistic children adversely impact parental QoL. Observational and survey based studies assessing connections between parental QoL and child autism features have shown that parents of children who have better behavioural regulation and attention, communication and fewer restricted and repetitive behaviours and internalising behaviours have better QoL [19].

Using the Quality of Life in Autism scale (QoLA) [24], a recent cross-cultural comparison set out to determine whether (i) parent perception of their QoL (QoLA Part A) and (ii) parent report of their child’s autism-specific difficulties (QoLA Part B) differ across cultures [25]. Families who were attending services for intervention and/or childhood assessments completed the QoLA questionnaire as part of their routine care/assessment. De-identified data was then compiled for secondary data analysis to understand cross cultural similarities and differences in parent reports across collectivist and Westernised cultures. The findings indicated that parent perceptions of the level of difficulties their child experienced as a result of exhibiting autism-specific features were comparable regardless of cultural context. However, reports of parental QoL were influenced by the culture to which the parent belongs [25]. It was argued that parents from western cultures (Australia and the UK) with ease of access to higher quality health services and supports for young children reported higher overall parental QoL [25]. Whereas parents from non-western cultures rated lower overall QoL, and rated strongly on different domains relating to QoL such as self-determination (e.g. Hungarian parents), social inclusion and interpersonal relationships (e.g., Malaysian parents). Thus, use of a transcultural lens is important in considering the full extent to which raising an autistic child impacts a family. Importantly, there is also evidence that the sociodemographic characteristics of parents including education and professional status [26, 27] influence parental stress, coping and reported QoL.

While these studies provide consensus that parental QoL is adversely impacted in families raising an autistic child, the results are based on relatively small sample sizes, targeted contexts, differing measures of QoL, and are generally limited in examination of the combined associations between parent, child and sociodemographic factors that can impact QoL [24, 28]. Systematically understanding how sociodemographic, cultural and parent and child factors combine and contribute to parental QoL in a multicultural Australian sample, will inform the evidence base on how families can be best supported including addressing the social determinants relevant to QoL. We expect that this will support bespoke adaptation of programs to better support the day-to-day challenges faced by parents raising autistic children, particularly in the early years. To address this knowledge gap, the aim of this study was to determine the association between the most commonly self-reported parental variable impacting QoL - stress, together with sociodemographic factors including age, education, employment, income, as well as child behavioural and emotional profiles, and QoL in a national sample of young children attending Autism Specific Early Learning and Care Centres (ASELCCs). We hypothesised that higher parental stress, child autism traits (particularly as behavioural, communication, and cognitive functioning), and social determinants of health would be significantly associated with parental QoL.

Methods

Study design and participant profile

Secondary analysis was undertaken of the data collected from the Autism Subtyping Project, a prospective, multisite, cohort of children less than 6 years of age with a diagnosis of ASD or probable ASD-like symptoms, receiving early intensive intervention (EII) at six ASELCCs across the six states in Australia. Of the 518 participants included in this study, the mean age (SD) of the children were 3.4 (0.9) years. A substantial proportion of children were male (80%) with only 20% female children. More than half of the children (56%) had a co-occurring condition and about one in four children (24%) were reported to have a sibling with autism. With regards to parental characteristics, about 44% reported being from a culturally and linguistically diverse (CALD) background. Additionally, more than half of the primary carers reported having a tertiary qualification with about 30% having a profession or paraprofessional occupation.

Written informed consent was obtained from parents prior to enrolling into the study. Following consent, all children completed a battery of clinical assessments to assess cognition and adaptive behaviours at entry into each service. Additionally, parents completed several questionnaires related to sociodemographic information, child behaviours, parental stress, and quality of life. Further details about the programs, assessments, and follow-up are detailed elsewhere [29].

Study measures

For the purpose of this study, we used the parental-reported quality of life in autism (QoLA) questionnaire as the outcome measure and limited the sample to those who answered the QoLA. The primary exposures included the following child measures: autistic traits, cognition and behaviour measured using the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2), Child Behaviour Checklist (CBCL), Social Communication Questionnaire (SCQ), Repetitive Behaviour Scale–Revised, Vineland Adaptive Behaviour Scales, Second Edition (VABS-2), and Mullen Scales of Early Learning (MSEL); along with parental stress reported on the Parental Stress Index (Parent Stress Index-4 Short Form). Additionally, key sociodemographic variables were assessed and adjusted for in the regression models.

Outcome measure

The Quality of Life in Autism Questionnaire [30] is a 48-item questionnaire that measures the parental quality of life of autistic children. This self-reported questionnaire has two subscales where Part A subscale is used to assess the general parental (or carers) quality of life and Part B is used to ascertain parents’ perception of the extent to which their child’s challenges related to autism impact them (the parents/carers). Scores of Part A subscale range from 28 to 140 and Part B from 20 to 100 with higher scores indicating better quality of life and fewer problems regarding their child’s autistic behaviours, respectively. In this sample, there was excellent internal consistency with Cronbach’s alpha ranging from 0.93 (Part B) to 0.94 (Part A).

Child measures

The Autism Diagnostic Observation Schedule, Second Edition (ADOS-2) [31] is a standardised assessment tool widely used to evaluate individuals suspected of being autistic, providing a comprehensive and reliable observational measure of social communication skills, repetitive behaviours, and play-based interactions. This clinician-reported assessment is particularly valuable in clinical and research settings, aiding professionals in making accurate and consistent diagnoses of autism. The ADOS-2 consists of various modules tailored to the individual’s age and language proficiency, ensuring flexibility and applicability across diverse developmental stages (Lord et al., 2012). Given the age range of 1–5 years in this sample, the toddler module and modules 1–3 were used depending on the level of verbal comprehension. The total scores were converted into calibrated severity scores which allows comparability across modules, with higher scores indicating higher severity of autistic traits. In this sample, there was good internal consistency with Cronbach’s alpha of 0.84.

The Child Behaviour Checklist (CBCL) [32] is a widely used tool designed to evaluate behavioural and emotional problems in children and adolescents. The CBCL has been used in diverse clinical and research settings to aid in diagnostic decision-making and treatment planning for children and adolescents. In this study, we used the parent-report measure of the CBCL and the internalising and externalising problem scores were used as exposure variables. Higher scores indicate increased internalising and externalising problems. In this sample, there was excellent internal consistency with Cronbach’s alpha of 0.96.

The Social Communication Questionnaire (SCQ) [33] is a widely used screening tool designed to assess social communication skills and identify potential autism in individuals. The parent-report questionnaire consists of structured questions covering areas such as reciprocal social interactions, communication skills, and repetitive behaviours. Higher scores on the SCQ typically indicate a greater likelihood of social communication difficulties and may suggest an increased risk of autism. In this sample, there was excellent internal consistency with Cronbach’s alpha of 0.92.

The Repetitive Behaviour Scale-Revised (RBS-R) is a comprehensive assessment tool used in both clinical and research settings to assist clinicians and researchers in systematically examining and quantifying the frequency and severity of repetitive behaviours including stereotypies, self-injurious behaviours, compulsions, and rituals, and provides valuable insights into the individual’s behavioural profile [34]. Higher scores on the parent-reported RBS-R scale suggest an increased frequency and severity of repetitive behaviours in individuals being assessed. In this sample, there was excellent internal consistency with Cronbach’s alpha of 0.94.

The Vineland Adaptive Behaviour Scales, Second Edition (VABS-2) [35] is a standardised assessment tool that provides valuable information about an individual’s ability to function independently and adapt to the demands of everyday life. The parent-reported tool is designed to measure adaptive/functional skill level in four broad domains of daily living skills, communication, socialization, and motor skills, with higher scores indicating better adaptive behaviours. In this sample, there was good internal consistency with Cronbach’s alpha of 0.84.

The Mullen Scales of Early Learning (MSEL) is a widely used clinician completed measure of children’s cognitive abilities and developmental progress in five key areas of gross motor, fine motor, visual reception, receptive language, and expressive language [36]. In this sample, there was excellent internal consistency with Cronbach’s alpha of 0.90.

Parent measure

The Parent Stress Index-4 Short Form (PSI-4-SF) [37] is a brief self-report tool designed to evaluate the stress experienced by parents in relation to their parenting roles and responsibilities. The PSI-4-SF covers areas such as parental distress, parent-child dysfunctional interaction, and difficult child characteristics, with higher scores indicating higher level of perceived stress. In this sample, there was excellent internal consistency with Cronbach’s alpha of 0.93.

The key sociodemographic indicators used in this study comprised child-level characteristics including age in years, gender (male or female), first born child (yes or no), other medical conditions (yes or no), and whether the siblings have autism (yes or no). Parent-level characteristics included age in years, country of birth (Australia or Other), preferred language (English or non-English), culturally and linguistically diverse (CALD) status– based on ‘preferred language’ to be non-English and/or the ‘country of birth’ was a non-English-speaking country (yes or no), education level (Primary/secondary level or Tertiary/graduate level), occupation (professional/paraprofessional or other), living arrangements (both biological parents or other arrangements), carer’s disability status (yes or no), and annual income (up to $40K, $40K to 100 K, and above $100K) used as an indicator for socioeconomic status.

Data analysis

Descriptive statistics of mean with standard deviation (SD) and frequency counts with percentages were used to report the sociodemographic characteristics as well as the clinical and self-reported assessments. Shapiro-Wilks test for normality and analysis of normal quantile-quantile plots were used to assess the normality of distribution.

Univariate linear regression analysis was used to determine the factors independently associated with parental quality of life. Given the large number of sociodemographic, cultural, and socioeconomic variables, only variables with a p-value ≤ 0.20 in the univariate regression analysis were entered into the multivariable models. This is a widely accepted approach in statistical modelling which reduces potential overfitting, reduces noise from other highly non-significant variables in the univariate model and improves model interpretability. The rationale is that a less stringent cutoff (e.g., p-value ≤ 0.20 instead of the conventional 0.05) helps avoid excluding potentially important predictors that may become significant when adjusting for confounders in the multivariate analysis [38, 39]. Pearson’s correlation analysis was conducted to determine any significant correlation between the child, parental assessments, and the QoLA subscales before entering them into the multivariable models. Multivariable linear regression analyses were used to determine whether the sociodemographic factors and clinical indicators were significantly associated with parental QoL whilst adjusting for key sociodemographic factors (for the latter). A p-value < 0.05 was considered as significant.

Results

Sociodemographic and clinical characteristics

The sociodemographic and clinical characteristics of the participants are shown in Tables 1 and 2. The mean (SD) score for the QoLA Part A subscale was 95.93 (19.9) whereas the mean (SD) score of part B was 69.81 (21.1) (Table 2).

Table 1 Sociodemographic and Sociocultural characteristics of the sample

Table 2 Parent and child assessments

Findings of the regression analyses

Findings of the regression analyses exploring the influence of sociodemographic factors (Table 3) showed that, at univariate level, sibling’s ASD diagnosis and carer’s disability status were negatively associated with parental QoL (Part A) whereas higher annual family income and CALD background were positively associated with parental QoL (Part A). Consistent with the univariate level, sibling’s ASD diagnosis (β = -6.02, 95% CI: -11.87, -0.17) and carer’s disability status (β = -7.33, 95% CI: -14.97, -0.29) remained significant and were negatively associated with parental QoL (Part A). Further, having multiple children with ASD was significantly associated with a greater negative impact on parents, as indicated by the Part B subscale (β = -6.62, 95% CI: -11.94, -1.29).

Table 3 Linear regression models showing sociodemographic risk factors associated with quality of life in autism (QoLA)

In terms of clinical indicators associated with parental QoL (Table 4), findings of the multivariable linear regression analyses (consistent with the univariate regression) showed that higher parental stress levels were significant and negatively linked to both Part A (β = -0.42, 95% CI: -0.53, -0.31) and Part B (β = -0.36, 95% CI: -0.49, -0.22) QoLA subscales. In terms of child’s autistic traits, higher internalising (Part A - β = -0.66, 95% CI: -0.99, -0.33 and Part B - β = -0.86, 95% CI: -1.23, -0.50) and externalising behaviours (Part A - β = -0.54, 95% CI: -0.81, -0.26 and Part B - β = -0.62, 95% CI: -0.93, -0.31) as well as repetitive behaviours (Part A - β = -0.24, 95% CI: -0.43, -0.04 and Part B - β = -0.27, 95% CI: -0.47, -0.06) were significant and negatively associated with both parental QoL subscales. Additionally, social communication difficulties were also significantly associated with both Part A (β = -0.80, 95% CI: -1.29, -0.31) and Part B (β = -0.85, 95% CI: -1.41, -0.29) QoLA subscales. Further, greater adaptive functioning in terms of communication (only significant for Part B subscale), daily living, socialisation, and motor living skills was positively linked to better parental QoL.

Table 4 Linear regression models showing clinical risk factors associated with quality of life in autism (QoLA)

There were significant positive interactions (interaction results not shown in tables) between CALD status and motor skills (VABS motor skill and MSEL fine motor skills scores) in relation to QoLA Part A indicating that child’s motor skills functioning corresponds to greater parental QoL for families with CALD status. On the other hand, there was a significant negative interaction between parental stress and CALD status, indicating that higher parental stress levels were associated with a more substantial decrease in QoL, particularly for families from CALD backgrounds. In terms of Part B subscale, there was a significant positive interaction between repetitive behaviours and family income, indicating that higher family income may increase parental capacity to source supports and make accommodations which may buffer its impact on parents. Additionally, consistent with the regression modelling, a significant positive interaction between autism severity and sibling’s ASD diagnosis was noted which suggests that the impact of higher autism severity on the parents’ experiences may be amplified with having another autistic child.

Discussion

This study examined the association between sociodemographic factors, child autism traits, parental stress, and parentalQoL to understand how the combination of parent, child and contextual factors contribute to perceived QoL in multicultural Australian families with preschool aged autistic children. Multivariable linear regression analyses identified that parents with higher stress levels and children with higher social communication difficulties had lower parental QoL. Furthermore, parents reporting less internalising and externalising problems and fewer repetitive behaviours in autistic children had higher QoLA B scores.

Examination of parent-specific factors indicated that increased perceptions of parental stress were associated with lower QoL. This concurs with previous international research that has shown that parental stress is associated with poorer parental QoL in parents of autistic children compared to parents of non-autistic children [9,10,11,12,13]. The findings also provide further evidence on the challenges faced by parents particularly when there are associated internalising and externalising behaviours [2], and higher scores on the repetitive behaviour scale. However, higher socioeconomic status (annual family income) was linked to higher QoL. This finding is similar to previous findings that higher levels of parental education and income are associated with increased self-reported QoL [26, 40].

In this study, parents’ CALD status and living arrangements did not significantly contribute to QoL scores. This is in contrast to other studies with school aged children that have shown cross cultural differences in perceived QoL using study specific [26] or validated QoL measures [25]. Nevertheless, it is important to note that protective factors were evident at the univariate level, with CALD background which may reflect better social and family supports, higher education and professional occupation which may suggest better health literacy and proactive help-seeking, and annual income that may indicate parental capacity to source supports. This also highlights the need to provide comprehensive supports for those from disadvantaged backgrounds by providing wraparound social care support alongside autism specific supports in order to enhance the overall QoL for children and their families and support networks. Given that all study participants had access to early learning supports potential disparities in CALD or more vulnerable families may have been masked. Indeed, previous studies argue that access to high quality supports improves parent reported QoL [25]. The mechanisms by which CALD status impacts QoL are beyond the scope of this study, as CALD status was inferred by reported country of birth and language spoken at home. Further, systematic evaluation of large CALD samples would improve our understanding of the different aspects of culture that provide the greatest protective factors for families with young children.

Closer inspection of the child-specific factors indicated that higher CBCL internalising and externalising scores were linked to greater difficulties supporting autistic children. Similarly, higher scores on the RBS-R were linked to higher problems in parental ability to support the child needs. Conversely, lower SCQ scores indicative of better social communication skills in children were associated with higher QoL in parents. Furthermore, parental reports of greater adaptive functioning in terms of communication (significant for QoLA Part B subscale), daily living, socialisation and motor skills was associated with better parental QoL. These findings concur with other studies that have shown that the level of autistic features coupled with increased adaptive functioning, and better behavioural regulation and attention are associated with higher parental QoL [19]. This is also consistent with previous studies indicating better communication, adaptive functioning and behavioural profiles are associated with parents reporting less difficulties and challenges in the day to day management of autistic children [5, 6, 20, 21]. However, in the multivariable analyses in the current study, neither parent report of their preschoolers’ adaptive behaviours or clinician administered developmental and behavioural assessments were found to be associated with QoL.

Strengths of the current study include a large national sample of preschool autistic children in multicultural Australia. Understanding how sociodemographic and clinical indicators contribute to QoL of families with preschool aged children shed light on factors that can be targeted by early supports to continue on their best trajectory. Additionally, data on both parent and child variables alongside sociodemographic characteristics were available that allowed examination of the factors that impact parental QoL from a sociodemographic lens while accounting for individual child and parent characteristics. Limitations include a lack of longitudinal follow up beyond preschool years, thus inferences into enduring outcomes are not possible without further longitudinal studies. Secondly, the findings demonstrate an association rather than a causative relationship between predictors and outcomes. Thirdly, given that many of the child and parent measures were self-reported by parents, there is the limitation of shared variance and single rater bias, as well as the additional possibility that the responses may be biased due to social desirability [41, 42]. Finally, the sample may lack external validity as it is not a representative sample of the general population.

Future research should further explore the reasons behind the protective factors identified in this study and how this can be leveraged in strength-based approaches to identify individual, family and wider community related factors that are associated with better parental QoL. Further such findings can be utilized to inform the development of interventions and support services that build upon existing child and family strengths to promote better health, wellbeing and QoL outcomes. Further population level studies are needed to understand the ensuing impact of early interventions and targeted support for preschoolers with a focus on parental and family QoL.

Conclusion

The results of this study concur with previous studies that have shown that parents of autistic children experience heightened levels of stress and that higher severity of autistic symptoms, as well as associated internalising, externalising and restricted and repetitive behavioural issues negatively impact parental QoL. Future research examining how these factors can be supported via early intervention programs and supports will form an important step in ensuring young children and their families have access to the most appropriate early intervention support programs.