Parents’ Perceptions: Environments and the Contextual Strategies of Parents to Support the Participation of Children and Adolescents with Autism Spectrum Disorder—A Descriptive Population-Based Study from Switzerland

A population-based cross-sectional descriptive study was selected, describing features at a given point in time. ‘Environment,’ defined according to the definition of the World Health Organization (see Table 1), can be qualified as barriers or as supports to participation and are addessed with research question 1. Supporting environments are those addressing the human desires to explore, understand, enhance competence, be part of the solution, and participate with others towards meaningful goals (Kaplan & Kaplan, 2008, 2009). Parental contextual strategies are addressed with research question 2. They are enacted in “experienced and situated activity settings” (King et al., 2018; p. 1835) and are described with five inherent elements of contexts: people, place, activity, objects, and time (King et al., 2018).

Participants were German-speaking parents who care for one or more children between 5 and 17 years with a recognized medical diagnosis of ASD according to ICD-10 (actual diagnostic classification in Switzerland). Excluded were parents who do not live together with their children with ASD. For the self-selected sample, parents were recruited through invitation letters sent to multiple different pathways in the German-speaking part of Switzerland, such as client and professional organizations, social media and webpages, and specialized medical services and schools for children with ASD. The invitation letter included a link to a webpage with further information and an online survey. Participation was voluntary and the given information was anonymous in accordance with the data safety laws of the European Community. Parents consented online with informed consent. This survey received a jurisdictional declaration of non-objection by the cantonal ethical committee of Zurich (BASEC Request 2018-00238).

The online questionnaire consisted of the German version of the participation and environment measure-child and youth (PEM-CY(G) (Coster et al., 2012; Krieger et al., 2020b), demographic questions, and questions about the actual manifestations of ASD symptoms.

The demographic questions included the family constellation, living in urban, suburban or rural communities, education of parents and the actual percentage of paid work. This indicates the socioeconomic status of the family. The research team refrained to ask for the annual income and marital status, as this declaration is not common to Switzerland. Cultural diversity was reflected on the reported languages spoken in each family. It was estimated that clinical diagnostic features were either unknown to parents or not currently up to date. In Switzerland, until now the ICD-10 criteria are mostly used for ASD diagnostic. Therefore, we developed an actual ASD-manifestations questionnaire according to ICD-10 diagnostic criteria (e.g. communication, restricted and repetitive behavior) and reported challenges from parents of children with ASD (e.g. ability to express themselves, difficulties handling change, sleeping situation, self-injuring behavior, age-appropriate independence) (Galpin et al., 2018). The eleven manifestations were rated for the last 4 months on a Likert-type scale between 1 and 6. The scales were formulated with qualitative anchors (e.g., 6 = “Our child can express himself or herself age-appropriately or better” versus 1 = “The expressive language of our child is very low”). The questionnaire was scrutinized by three clinical ASD experts for completeness and readability; consequentially, minor amendments were made.

The PEM-CY is a standardized parent-reported assessment of the extent and patterns of participation of children and youth between 5 and 17 years and parents’ desires for change in three different settings (home, school, and community). For each of these three settings, 10 to 17 different environmental aspects, divided into “helpfulness” (e.g., sensory features, demands, socializing persons, attitudes) and “resources” (e.g., services, information, time, or money) are assessed. Parents are asked whether these aspects are helpful or available, which are operationalized in PEM-CY as “supports,” or not helpful, which are operationalized in PEM-CY as “barriers.” These were used to answer the first research question. To answer the second research question, parents are further asked in the PEM-CY (G) with open-ended questions to describe three contextual strategies that they apply to support the participation of their children in each setting. PEM-CY has demonstrated adequate internal consistency and test–retest reliability (Coster et al., 2011). It has been applied to young people with ASD (Devenish et al., 2020; Egilson et al., 2017, 2018; Lamash et al., 2019; Simpson et al., 2018; Tint et al., 2017; Weiss & Burnham Riosa, 2015). For this study, a German-translated and cross-cultural adapted version of the PEM-CY(G) (Krieger et al., 2020b) was used.

The questionnaire was set up on an online platform (SocSicSurvey.com). For piloting, five parents of children with ASD provided feedback on comprehensiveness and user-friendliness and it was changed accordingly. As data collection took place between the 6 weeks of Covid-19 lockdown and restrictions, parents were asked to refer to participation experiences in the time before the lockdown for their answers.

The data were analyzed using IBM SPSS Statistics (Version 27.1) analytical software.

Demographics of parents and youth were summarized and total numbers, percentages, and means and/or medians were calculated for the whole sample and two age groups (children: 5–11 years and adolescents: 12–17 years). This age division reflects the difference between primary and secondary school in Switzerland. Based on the data level, t-tests or Mann–Whitney U tests were performed to indicate a statistically significant difference level of p = 0.05 for demographic data. Parent-reported manifestation of ASD was calculated using the median and quartile range for the whole group and the two sub-groups. Calculations were based on Tukey’s range tests to account for the ordinal scale used.

PEM-CY measures 45 total environmental aspects. Their number varies per setting (home: 12; school: 17; community: 16). As recommended by the PEM-CY user guide, environmental aspects judged with “usually helps” and “usually yes” were viewed as “supports” and those judged with “usually makes harder” or “usually no” were viewed as “barriers.” We calculated the average amount of perceived “support” and “barriers” for all three settings, both age groups, and the whole sample. For more in-depth insight, for each of the 45 environmental aspects the percentage of parents who opted for a particular answer in each age group and in the whole sample was calculated.

For each setting (home, school, and community), parents were asked in an open-ended question (part of the PEM-CY questionnaire) to list three strategies that they use to support their children’s participation. Due to received rich data, the research team decided on a secondary analysis by posing another research question beyond those originally intended with the primary data (Ruggiano & Perry, 2019). The same two researchers who performed a first analysis (BK, TM) applied a summative content analysis, typically used to explore word usage or contents in texts (Hsieh & Shannon, 2005). It is seen as a deductive approach (Armat et al., 2018). The two native German-speaking researchers used the five aspects of context (people, place, activity, objects, and time) as defined by King et al. (2018) as pre-determined codes. After agreeing on definitions in a codebook, both coded the written comments of parents, which were separated for the three settings (home, school, and community) and were divided into two age groups (children and adolescents). Next, they compared, discussed, and reflected on their results. After saturation was reached, which means that no new codes or themes emerged, they presented the qualitative results to a third native-speaking researcher (AM) to refine it further. After translating the results and exemplary codes into English, they presented the results to the whole research team until saturation on positioning of all codes was reached. Finally, they agreed on calculations, contents of the table, and a narrative summary report. To ensure trustworthiness (Guba & Lincoln, 1985), all received written strategies were coded and data triangulation with quantitative and qualitative aspects ensured credibility. Researcher triangulation contributed to the confirmability of the study. Transferability was guaranteed by providing a well-documented context and research process.

The final analysis included 115 participants. The flow of eligibility is additionally given as supplementary file (see Online Resource1). The demographics of the 115 parents are listed in Table 2. Overall, 60 parents reported on children (5–11 years of age) while 55 reported on adolescents (12–17 years of age). Statistically, the two groups did not differ significantly (p = 0.05) with regards to responding parents, community type of residence, and education level of parents. However, the family constellation differed statistically in adolescents, who lived more often in separated households. Further, the paid working hours of mothers of adolescents with ASD were higher than those of mothers of children with ASD.

The demographics of the children and adolescents with ASD are listed in Table 3. The two age groups differed statistically in the average age of diagnosis, number diagnosed with Asperger’s syndrome, and the number of friends and peers (meetings via social media) they meet with per week, whereas 15–20% of parental answers for informal friendships and social media contacts were missing.

The distribution of autistic characteristics in the children and adolescents with ASD was determined from the statements of the parents (see supplementary file Online Resource 2) “Expressive language ability,” “use of communication aids,” “intellectual abilities,” and “no self-harming behavior” were rated highly in both groups, indicating a less severely challenged autistic sample. “Reaction to change,” “repetitive behavior,” “restricted behavior,” and “selective eating pattern,” were rated in the middle range for both sub-groups. In contrast, “sleeping situation,” “interaction with other children,” and “age-appropriate independence” scored in the middle range, though there was more variation in adolescents with ASD than in children with ASD. Solely “reaction to change” presented with a high spread in both groups. Overall this is read as a sample with overall “high functioning ASD”.

Table 4 lists the percentages of parents’ perceived judgements of the 45 environmental aspects in three settings. Half of all parents indicated clear environmental “supports” and “barriers” to the participation of their children. From this half, both age groups were perceived to have more “supports” than “barriers” in all three settings. The average number of perceived supports in the three settings (home: M = 4.6, SD = 2.3; school: M = 6.2, SD = 3.4; community M = 5.4, SD = 2.7) was over 50% higher than the average number of perceived “barriers” (home: M = 1.1, SD = 1.3; school: M = 2.8, SD = 3.0; community M = 3.0, SD = 3.0). There were only minor differences between the age groups.

The other half of parents did not indicate environmental “supports” or “barriers” clearly. Out of these, two groups could be distinguished: first, a third of all parents choose “sometimes yes/helpful, sometimes no/hard” (home: 31%; school: 27%; community: 30%), indicating a swing between the environment being a “support” or being a “barrier.” Second, for 20% of all parents, environmental aspects were “not an issue” (home: 21%; school: 17%; community:17%). Most of the 115 parents answered all 45 questions. “Attitudes” had the lowest answer rate with 108, as seen in Table 5.

In the following paragraph, the highest percentages of “supports” and “barriers” are reported along each of the three settings. At home, parents judged most of the environmental aspects as “supports.” Specifically, “resources” (such as “supplies,” “information,” “money”) were rated highly. “Sensory quality” and “social demands of activities” were perceived as strong “barriers” in the home setting. At school, “attitudes” of teachers and staff and most of the “resources” (such as “programs and services,” “information,” and “time”) were rated as strong “supports.” Highly endorsed “barriers” at school were the “sensory quality” and the “social demands of activities.” In the community, parents judged that only “safety” was a clear support regarding aspects of “helpfulness,” while most “resources” (such as “transportation,” “policies and procedures,” “information,” “time,” and “money”) were also rated as strong “supports.” Clear “barriers” in the community were “sensory quality” and the “social demand of activities.” The community was the setting in which most parents chose to swing between “sometimes helps and sometimes makes harder.”

A summary of “supports” and “barriers is presented for each setting graphically in Fig. 1. It is apparent (as reported earlier) that in all three settings, parents reported more “supports” than “barriers,” with a decline in support from home to school to the community. Next, the distribution regarding the 45 environmental aspects remained nearly similar for both age groups. Differences over 10% between age groups were found in “weather conditions,” which were stronger “barriers” at home in younger ages. Supportive “attitudes” at school were higher in children with ASD than in adolescents with ASD, while the reverse was true for “public transport” at school. Not all “supports” and “barriers” per setting were equally distributed. Parents in all three settings perceived more “supports” in aspects summarized as “resources” (such as “services,” “information,” “time,” “money”) than those summarized under “helpfulness” (such as “physical layout,” “sensory qualities,” “physical demands,” “social demands,” “attitudes,” or “social relations”). This is mirrored by “barriers” in all settings: parents perceived more “barriers” in aspects summarized under “helpfulness” than those summarized as “resources.”

The following paragraph contains a summary of contextual strategies using the five themes of “people,” “activity,” “time,” “objects,” and “place,” which are presented in nine clusters such as “encouraging participation” or “accompanying, sharing, or supervising activities.” Each cluster contains several single strategies. Table 5 lists five themes and nine clusters and describes 22 single strategies, for which exemplary quotes from three settings and two age groups are displayed. Most strategies were used in all three settings. In total, 624 strategies were mentioned (home: 231 strategies; school: 220 strategies; community 172 strategies). Only minor differences between the age groups were found. The most mentioned contextual strategies are presented first.

The next three parental strategies are clustered in one cluster with similar strategies.

Applying strategies connected to “time” (11% of all comments), parents opted for regularity and rituals (often also in combination with rules), provided altogether more time for participation experiences, and reduced the pace of activities. Parents mentioned they reduced the total length of activities or planned free time slots between activities. Parents described further how they were mindful of the actual state of mind of their child. Waiting for the “right” moment to introduce a new activity or being well-rested before a socially-demanding participation experience seemed central.

Parental strategies around “objects” (6% of all comments) were connected to the availability of materials to visualize concepts as this was described as easing communication. “Objects” were also mentioned in combination with rewards and incentives used to motivate the children and adolescents with ASD for externally demanded activities that were often necessary but unpleasant, such as showering or joining a social event.

Last, parental strategies around “places” (6% of all comments) described themes connected to sensory overload. Parents chose rooms or spaces to prevent sensory overload or taught their children strategies to prevent it. Next, parents applied strategies to react to possible sensory overload by looking for places to withdraw or by providing opportunities to rest and restore from sensory overload.

This study resented “supports” and “barriers” descriptively and graphically without summary scores. This in contrast to other studies, which used summary scores of supportiveness (Bakaniene & Prasauskiene, 2020; Kaelin et al., 2020) or combined the answers “not an issue” and “usually helps” (Egilson et al., 2018). Without summary scores it became obvious that in the present study “supports” were twice as numerous as “barriers,” and were at the same time different in nature. “Supports” (such as “services,” “information,” “time,” “money,” “equipment and supplies,” and “transportation”) are either automatically provided by the Swiss social system or families in Switzerland can opt for them. Families can opt, for example, to reduce parental workload and thus gain additional time to support the participation of the child or adolescent with ASD.

This freedom of choice contrasts with the perceived “barriers” (such as “sensory quality” “cognitive demands of activities,” “social demands of activities,” “relationships,” and “attitudes”) which are part of the built or socially-constructed environment and go beyond the scope of single families. By scoring these things as “barriers,” parents expressed a certain helplessness. Their ability as parents or families to influence things like environmental noise, social rules, or attitudes to accommodate the needs of their children with ASD is perceived as rather low (Butler & Gillis, 2011; Jones & Harwood, 2009). Instead, these areas contain potential for service providers and society in general to support parents with children with ASD effectively. Examples for this are general noice restricting constructions or regulations, a higher diversity regarding fulfilling unspoken social rules and campains to address negative stereotyping of persons with autism.

Half of the parents swung between reporting the environment as a clear “support” or “barrier,” but instead said the environment served “sometimes as a help and sometimes as a hindrance” or was “not an issue.” Eglison et al. (2018) reported a similar pattern. The authors see two possible explanations for this uncertainty: first, parents found it hard to determine an overall influence of the environment, as demanded by the construct underlying PEM-CY (Krieger et al., 2020b). A dichotomous answer option might not be adequate to capture the complexities of environmental influence. Second, environmental aspects are dependent on enacted contextual situations as proposed by King et al (2018). For example, whether “being together with other people” is supporting or hindering in a setting depends on the persons in the setting. In a school setting, a librarian might be supportive of a child with ASD, while in the same setting, a sports teacher may not be. This swing was primarily found in the community setting. The authors suspect this is because community or public environments are more diverse and less predictable.

“Supports” and “barriers” did not differ considerably across the three settings in our sample, a finding similar to that seen in children and adolescents with non-ASD disabilities (Bakaniene & Prasauskiene, 2020; Coster et al., 2011; Shabat et al., 2021). However, it is pointed to particularities at the home and the community setting:

In a home setting, parents have the operating agency to adapt the environment to the needs of their children and adolescents with ASD. Parents rated “the relationships with family members” as the strongest “support” in the home. Family relationships were found to be highly important in other research as well (Krieger et al., 2018; Orsmond et al., 2006; Taylor & Seltzer, 2011). Parents rated “sensory quality” as the highest “barrier,” indicating a need to find suitable solutions (Hazen et al., 2014; Little et al., 2014). An intervention framework to address sensory issues in the home exists to guide service providers and health workers (Ashburner et al., 2014). It proposes universal design principles and self-regulation strategies to optimize participation experiences for children and adolescents with ASD.

In community settings, the reported barriers from Switzerland (CH) exceed those reported from Australia (AU) (Devenish et al., 2020). Differences were specifically found in “relationships with peers” (CH 31%; AU 8%), “safety” (CH 22%; AU 8%), “sensory quality” (CH 45%; AU 33%), and “social demands” (CH 40%; AU 35%). The age difference between these two samples (CH: 5–17 years; AU: 5–12 years) could be one explanation for the discrepancy. However, the authors assume that attitudinal, cultural, or policy differences between the two countries contribute the most to these differences. Unfortunately, most research into attitudes and stigmatization in ASD is nationally-based (Thompson-Hodgetts et al., 2020). From an anthropological viewpoint, research suggests that cultural differences can even mediate environmental differences such as maternal education, ethnicity, and the perceived negative impact of ASD (Carr & Lord, 2012). Region-based cultural aspects need to be included in any further research focusing on transnational environments. Further, a supportive environment should be perceived as supportive from the persons with ASD themselves (Gardiner & Iarocci, 2014).

The most common contextual strategies of parents (“people” and “activities”) correspond to the two main diagnostic features of ASD (American Psychiatric Association, 2013): (1) social-communicative differences, connected to strategies that were condensed into “people” and (2) restricted and repetitive behavior, connected to strategies that were condensed into “activities”. As such, parents provide support targeted to address autistic traits that hinder the participation of their children with ASD. Further, the results generally reflect the seven strategic patterns Egilson et al. (2018) found in adolescents with and without ASD, but this present study tailored them more specifically to children and adolescents on the autistic spectrum.

In the contextual strategies labeled “people,” parents use their confidential relationship with their children and adolescents with ASD to motivate, provide company, and serve as gatekeepers for new social contacts and participation. Parents and families are the most critical environment in childhood (Rosenbaum & Gorter, 2012). While less dependence on parents is assumed in adolescence (Imms et al., 2017), parents are still extensively involved in the lives of adolescents with ASD. The constant work of securing and helping to form connections (Krieger et al., 2018), which can involve the parent taking on many different roles—such as motivator, door-opener, or companionship—to create a supportive environment for children and adolescents with ASD, can be tiring over time. Other persons from the wider support circle—friends, practitioners, or service providers—can relieve some of the strains parents experience. For example, school personnel can replace parents in supervising homework, or other children can walk together and socialize with youth with ASD (Ziviani et al., 2006). Hence, children and adolescents need company and consistency, but these things do not necessarily have to be provided by parents (Adams et al., 2019; Krieger et al., 2018, 2020a). Although parents often see themselves as gatekeepers in connecting their children with others, it is noticeable how few strategies parents recalled using to connect their children with peers. Service providers may focus on educating parents about how to support peer-to-peer interaction. Shared activities, such as walking together to school or working on homework in a group, increase social participation in children and adolescents with ASD (Chen et al., 2016).

“Activities” comprised nearly a third of all parental strategies, indicating how frequently parents search for and adapt activities and motivate and inform their children about them. The poor executive functioning combined with less flexibility and poor problem-solving capacities seen in ASD might be one reason for this (Lopez et al., 2005). From all parental strategies, 15 percent comprised informing and making children and adolescents knowledgeable about a topic. Adolescents with ASD expressed a high need to get information and become knowledgeable before they could participate in a community (Krieger et al., 2018, 2020a), something that goes beyond parental strategies and can inform service providers and therapists as well (Adams et al., 2019).

Three strengths of this study are notable. First, it uniquely presents estimates of 45 activities regarding the participation of children and adolescents with ASD in three settings in Switzerland. Second, it describes for the first time parental contextual strategies tailored to children and adolescents with ASD which cover the whole phenotype of autism. Despite the wide use of the PEM-CY with various populations and research questions, parental strategies are rarely published, probably due to limited space in journals. Our paper’s last strength lies in the combined use of contextual theory (King et al, 2018) with parental strategies, which makes the transactional interconnectedness between a person and contexts, setting, and environment for the sake of an enacted participation more transparent.

In addition to a first qualitative analysis, the authors found it useful to apply the context definition of King et al. (2013, 2018) for our second analysis. While doing so, it was noticed that attitudes of others, institutional barriers, or social-economic situations were not reflected in the definition of contexts. These were only mentioned incidentally and did not interfere with the coding of parental strategies.

Several limitations of this study require consideration when interpreting the findings. Due to the population-based nature of the sample and the fact, that participants identified themselves as parents of children or adolescents with ASD, a participant bias might be possible. Next, as for all parent reports, the possibility of under- or over-rating is inherent (Allonsius et al., 2021; Rios & Scharoun Benson, 2020). Parents needed to score an environment they are fundamentally connected with. A selection bias is possible because filling in the PEM-CY is challenging and often missing data sets are described (Chien et al., 2020). After consenting, the enrollment rate was 85% and a completion rate 73%, indicating a lost a quarter of participants after reading the instructions and while completing the PEM-CY. No particular pattern of those who quit was observed, but the sample of the study represent a higher education rate as reported from official statistics (Bundesamt für Statistik, 2021). Regarding family constellation, workload and the number of spoken languages in the family, the sample of this study data fits those from official statistics (Bundesamt für Statistik, 2017). Further, another demographic aspect in combination with the environment was noteworthy: 74% of the Swiss population lives in urban or suburban communities (Bundesamt für Statistik, 2021). Instead, in the sample of the study, only 42% of participants were urban and suburban residents, while 57% of participating families were living in rural communities. It is yet unclear whether this is a possible self selection bias, or whether parents with a child with ASD chose due to aspects such as availability, financial, or social reasons rural environments. Fourth, the authors acknowledge that Covid-19 has changed the face of society, accessibility, and services due to the long ripple effects of the pandemic. The authors could not have foreseen nor controlles these effects, which might influence the interpretation of the results of this present study. Finally, as inherent in any cross-sectional study design, causality cannot be attributed and a direct connection between the two age groups cannot be supposed. However, the authors deliberately chose a purely descriptive analysis, and the results should be informative to health care providers and policy developers not acquainted with the PEM-CY construct. Due to the nature of a cross-sectional study, qualitative and quantitative methodology may not equally be presented.