Applied behavior analytic intervention services for children with autism spectrum disorders: an investigation of Chinese parents

Abstract

Introduction

Applied behavior analysis (ABA)-based interventions have been used worldwide for several decades to improve socially significant behaviors of individuals with autism. However, as the utilization of the ABA-based interventions in the Chinese mainland is relatively new, it remains unclear about some fundamental aspects of using these interventions in this region.

Method

This study investigated the use of ABA-based interventions in Beijing and surrounding areas by using a quantitative survey (N = 85) and a qualitative semi-structured interview (N = 10) to collect data from parents of children with autism.

Results

This study found that (a) the ABA-based intervention service was the most popular one chosen by parents, and the majority of these parents reported improvements in children’s behaviors and gave positive evaluations for the work of professionals; (b) this service was mainly delivered at autism organizations (i.e. autism clinics/schools) or children’s homes, but rarely used in mainstream school settings; (c) the high cost of ABA interventions was a financial burden for families; and (d) a high proportion of parents were involved in interventions as assistant therapists. Furthermore, some aspects of the use of this service varied with families’ socioeconomic levels. The qualitative interviews expounded upon motivations for parents’ choices and evaluations for this service.

Conclusion

These findings provided valuable insights for service providers, policymakers, and researchers to optimize ABA-based intervention services for Chinese children with autism.

Applied Behavior Analysis (ABA) is a scientific approach for improving socially significant behaviors through the management of environmental contingencies (Baer et al. 1968, Cooper et al. 2019). Based on ABA, researchers and practitioners have developed intervention strategies and techniques to improve social interactions and communication, cognitive skills, language abilities, life skills, and adaptive behaviors of children with autism spectrum disorders (ASD; Hume et al. 2021, Sam et al. 2020). ABA-based interventions typically include many techniques and strategies such as discrete trial training (DTT; e.g. Smith 2001), natural environmental teaching (NET; e.g. Lovaas 1987), and functional behavior assessment (FBA; e.g. LaBelle and Charlop-Christy 2002).

ABA-based interventions were originally developed in the United States in the 1960s (Baer et al. 1968) and then spread world-wide (Keenan et al. 2015). These interventions were introduced into the Chinese mainland in the early 1990s (Zhou et al. 2014). Given this relatively short history, not many empirical studies have been done to investigate the use of ABA-based intervention service within the Chinese mainland. The use of autism intervention services involves many aspects (e.g. Thomas et al. 2007, Leif et al. 2020). The present study was not exhaustive in its examination of all aspects of ABA-based interventions but focused on some fundamental ones, including the popularity of such services, the patterns of intervention delivery, cost, and clients’ evaluation of the interventions (Horiguchi 2014, Alnemary et al. 2017, Sridhar et al. 2022). To do this, we used a quantitative survey and qualitative interviews to collect data from parents of children with ASD. Furthermore, we examined whether parents’ choices and evaluations varied across family socioeconomic levels (see Figure 1 for the framework of the present study).

Figure 1.

The conceptual framework of the present study.

Note: The content with the white background was the concepts measured with a quantitative survey, and the content with the gray background was the concepts measured with qualitative interviews. The content with the dark background referred to topics presented in the quantitative survey or the qualitative interviews.

Use of ABA-based intervention service in the Chinese mainland

Popularity of ABA-based interventions

An important inquiry about the use of ABA-based intervention service is to understand the popularity of such service (Sun et al. 2013, Campbell et al. 2021). In the current study, popularity refers to the proportion of families choosing ABA-based intervention service (Campbell et al. 2021; Thomas et al. 2007). However, little data have been collected to show the popularity of this service in the Chinese mainland. A survey from Shenzhen Autism Society (2013) shows that this service was provided by most autism organizations in the Chinese mainland (88.9% of organizations) and ranked as the fourth most popular approach by parents (40.6%; other services included sensory integration (73.9%), speech and language therapy (61.7%), and play therapy (53.7%)). Over 60% of the respondents in this survey were from the south of China. However, another study (Sun et al. 2013) reported that ABA-based interventions were the most popular treatment services (used by 70% of parents) compared with others (e.g. sensory/auditory integration therapies). A majority of these parent respondents were from the north of China, and their children were receiving autism services in Beijing. The difference in the reported popularity between the two studies may be due to that samples were collected from different regions of China. Since ABA-based interventions were initially introduced in Beijing (Zhou et al. 2014) before spreading to other regions, Beijing has more resources for ABA-based interventions compared to other areas in China. Consequently, ABA-based interventions may be more popular in northern China than in other regions. The regional difference needs to be taken into account when depicting the popularity of ABA-based interventions.

In addition to popularity, we also need to further understand the reasons for parents’ choice of ABA-based interventions. Unfortunately, this has not been explored in previous studies (Sun et al. 2013, Shenzhen Autism Society 2013). In-depth interviews with parents can help reveal how they learned about ABA-based interventions, as well as their experiences, understanding, and evaluations about these interventions. This information is useful for us to understand the extent to which these services have been adopted across the Chinese mainland, as well as the factors driving parents’ preferences for them.

Patterns of ABA-based intervention delivery

How the ABA-based intervention service is delivered to children with ASD is another critical aspect of the use of this service. Here, the patterns of ABA-based intervention delivery encompass the intervention delivery format and the involvement of parents in ABA-based interventions. Understanding the delivery format helps service providers allocate their resources more efficiently (Dixon et al. 2017, Roberts et al. 2011), and facilitates parents’ choices of suitable services for their children (Sridhar et al. 2022). Furthermore, previous research has suggested that many Chinese parents are involved in interventions (Liao et al. 2020). However, the percentage of parents involved in ABA-based interventions and their reasons for involvement need to be further investigated.

Intervention delivery formats

ABA-based intervention services provide multiple delivery formats in which intervention programs are administered at autism centers (including clinics), children’s homes, and mainstream schools (Lambert-Lee et al. 2015, Denne et al. 2015). Different intervention delivery formats may have different functions (Dixon et al. 2017, Eldevik et al. 2012). For example, home-based interventions can help generalize learned behaviors (e.g. daily living skills like using the toilet) in natural living settings; mainstream school-based intervention settings provide more social oppotunities for children with ASD to interact with neurotypical peers (Leaf et al. 2018).

Different formats may demand different supports from rehibilitation organizations and parents. For example, while the center-based and school-based programs can involve both 1:1 (between a therapist and a child) and group programs (between a therapist and a group of children; Denne et al. 2015, Dixon et al. 2017), the home-based programs typically are 1:1 between an adult and a child (Sneed and Samelson 2022). That means intervention delivery formats have different requirements for the number of therapists. Given the low staff-to-child ratio in China, there may be a higher proportion of the use of center-based and school-based interventions than home-based interventions. In addition, different from home-based and center-based interventions, mainstream school-based interventions must require the support of schools or even local communities (Grindle et al. 2012), which may present additional challenges for using this format of interventions. Understanding the use of these service delivery formats is informative for us to more efficiently allocate the resources of interventions, and to coordinate autism organizations, families, schools, and local communities to optimize interventions for children with ASD.

Parents’ involvement in interventions

In addition to professionals, some parents are also involved in children’s intervention programs. The involvement of parents is important as it has positive effects on improving children’s behaviors (Deb et al. 2020) and reliving parental stress levels (Derguy et al. 2016, Yan et al. 2022). Previous research has shown that many Chinese parents serve as complementary resources to implement interventions for their children (Liu et al. 2020). However, the percentage of Chinese parents’ involvement in ABA-based interventions has not been quantified. Moreover, it remains unclear about the motivations for parents’ involvement. Investigating these issues could improve our accurate understanding of the roles of parents in interventions, and it is also informative for professionals, researchers, and policymakers to integrate parental training programs into interventions.

Costs of ABA-based interventions

The cost for interventions includes both direct expenses, such as payments to organizations for services, and indirect costs, such as travel expenses to access interventions and accommodation rentals when organizations are located far from one's home (Wilson et al. 2018). The overall cost of autism interventions varies across regions but is generally high (Rogge and Janssen 2019). For example, early intensive behavior intervention, an ABA-based service program, typically requires 20-40 hours per week and spans over two years (Lovaas 1987). The annual cost of such a program ranges between $40,000 and $60,000 in the USA (Rogge and Janssen 2019). However, the cost of ABA-based interventions is generally covered by health insurance in the USA (L and M Policy Research 2014).

The cost of interventions is a more challenging issue for Chinese families. First, this cost is not included in the social medical insurance of the Chinese mainland. Although families can apply for medical subsidies, they typically only cover a small percentage of the total expenses (Xiong et al. 2011). For example, in Guangdong province, children with autism who are under 6 years old can apply for monthly subsidies ranging from ¥1200 to 2000 (≈$180-300) for various interventions, not limited to ABA (General Office of People’s Government of Guangdong Province 2018). Second, the cost of interventions is beyond the capacity of some low-income or less well-educated families (Zhao et al. 2023), and this financial barrier restricts their ability to access these interventions. Third, the cost is even more burdensome for families in rural areas, as they must travel to cities to access qualified services. This leads to significantly higher indirect costs associated with interventions such as traveling expenses and housing rentals (Zhao et al. 2023). A survey showed that 67.6% of Chinese parent respondents whose children were diagnosed with ASD did not live in cities, and 42.7% had to travel outside of their residency to access interventions (China Association of Persons with Psychiatric Disability and their Relatives 2014).

Collecting data on the costs of ABA-based interventions would enable families to better estimate the budget requirements, assist the government in making informed adjustments to social welfare policies, and aid researchers and service providers in developing more cost-effective programs.

Evaluations for ABA-based interventions

Parents’ evaluation of ABA-based intervention plays a pivotal role in influencing their preference and use of such service. Evaluations of ABA-based interventions include several aspects (e.g. Heitzman-Powell et al. 2014, Hidalgo et al. 2015), and here we focused on two critical ones: parents’ evaluations for the level of supportprofessionals (therapists or supervisors) provided to children during intervention work, and the effectiveness of interventions in generalizing children’s learned behaviors.

Understanding parents’ evaluations of professionals’ supportiveness is crucial for service providers, as it enables them to tailor intervention goals and procedures to optimize outcomes for their clients (Goin-Kochel et al. 2009, Kazdin 1980), and this understanding can also improve the relationships between clients and healthcare providers, thereby enhancing the overall well-being of clients (Beach et al. 2006, Muskat et al. 2015). Moreover, gaining insights into parents’ evaluations of professionals’ supportiveness can significantly enhance professionals’ intervention skills (Caplan et al. 2016, Robertson et al. 2003).

Additionally, in the field of ABA practice and research, the concept of "generalization" involves assessing whether behavior changes resulting from ABA-based interventions extend to new environments, behaviors, and interactions with different individuals (Cooper et al. 2019). This evaluation is a key indicator of the effectiveness of such interventions. For example, it is often to examine whether new skills that are trained in center-based settings can be generalized to natural settings such as home or school (Cooper et al. 2019). However, there have been debates about the effectiveness of ABA-based interventions (Keenan et al. 2015). A majority of parents and researchers have reported the effectiveness of ABA-based intervention in improving children’s behaviors, including increasing their academic skills and decreasing challenging behaviors (Rosales et al. 2021), and generalizing children’s learned behaviors (Gorycki et al. 2020, Leaf et al. 2022). However, it is worth noting that some concerns have been raised regarding the limited generalization effects of ABA-based interventions (Sandoval-Norton et al. 2019).

Notably, these issues are the subject of heated debates in countries where ABA-based interventions have been widely used over an extended period. Although these debates have been also seen in some social media in the Chinese mainland (Xiaoyaya 2016), empirical data are still lacking to provide insights into how Chinese parents evaluate the effectiveness of ABA-based interventions, especially with regard to the extent of generalization in terms of children’s learned behaviors or skills.

Roles of SES in the use of ABA-based interventions

Given the large population of individuals with ASD, wide variation in these families’ socioeconomic (SES) (e.g. annual income and/or parental education level) can be expected. The use of autism intervention services (not limited to ABA-based intervention service) may vary with the SES of families (Irvin et al. 2012; Sridhar et al. 2022, Wilson et al. 2018). For example, some studies have shown that compared to families with higher SES, families with lower SES are less likely to choose ABA-based interventions (Irvin et al. 2012, Alnemary et al. 2017). Possible reasons include that ABA typically demands extensive training, which can be challenging for parents with lower SES. They may have limited time for training or encounter additional barriers, such as transportation difficulties (Pickard and Ingersoll 2016). In addition, parents’ evaluations of services for children with autism may also vary with the family’s SES. For instance, Hidalgo et al. (2015) indicated that families with higher SES levels tended to give lower evaluations for the medical care their children received, perhaps because these families were more sensitive to the inconsistency between the services their children actually received and the best practices they expected. For these reasons, it is necessary to examine the use of ABA-based intervention across families with different SES levels, which may present a more accurate and comprehensive understanding of the use of this service within different populations. Moreover, this information will also help modify ABA-based intervention services to meet the varying needs of different families.

To summarize, the current use of ABA-based intervention in the Chinese mainland was investigated by examining popularity, delivery patterns, costs, and parents’ evaluations, and exploring whether the use of ABA-based interventions varied with families’ SES. To explore these avenues, we used a survey to collect quantitative data from parents of children with ASD and conducted qualitative semi-structured interviews with part of the sample to further explore the reasons for their choices and evaluations of the ABA-based intervention service (Silverman 2016). This approach can combine the advantages of both quantitative and qualitative research and allows for a deeper understanding of the use of ABA-based interventions (Creswell 2002).

Method

The data reported here were part of a large project on early applied behavior analytic interventions for children with ASD). This study was approved by the School of Social Sciences, Education, and Social Work Ethics Committee at Queen’s University Belfast.

Participants

This study was conducted in several autism organizations located in Beijing and surrounding areas of China. Participants were parents/caregivers who were either residents of Beijing or from other cities and temporarily lived in Beijing to access treatments for their children (see Table 1 for the demographic information of participants). To be recruited for this study, participants must have at least one child diagnosed with ASD or “suspected ASD” from a doctor (e.g. psychiatrist or pediatrician). “Suspected ASD” represented that a child was not confirmed with an explicit ASD diagnosis by doctors, and the possible reasons included that a child was too young to receive an official diagnosis (usually under three years old; Zhao et al. 2023) or that the availability of diagnostic tools was limited in some hospitals (Sun et al. 2013). This group was also included in other research by Chinese researchers (Sun et al. 2013; Zhao et al. 2023; Zhou et al. 2014).

Table 1.

Demographic information of the surveyed parents.

	Parents (N = 85)
Gender
Female	71 (83.5%)
Male	14 (16.5%)
Average age (years)	33.95 (SD= 4.12)
Educational background
Junior high school or below	11(13%)
Technical secondary school	5 (5.9%)
Senior high school	6 (7.1%)
Technical college	19 (22.4%)
Undergraduate	32 (37.6%)
Master’s and above	10 (11.8%)
Overseas qualification	2 (2.4%)
Relationship to the child
Mother	71 (83.5%)
Father	13 (15.3%)
Others	1 (1.2%)
Marital status
Married and living with husband/wife	79 (92.9%)
A civil partner in a legally recognized civil partnership	1 (1.2%)
Married and separated from husband/wife	3 (3.5%)
Cohabiting with partner	2 (2.4%)
Parents’ working/employment status
Looking after home or family full-time	33 (38.8%)
Going to college/school full-time	11 (13%)
In paid employment or self-employed full-time or part-time	28 (32.9%)
Permanently unable to work because of the child’s disability	11(12.9%)
Others	2 (2.4%)
Family annual income
Less than ¥50,000*	28 (33.7%)
¥50,000 – 100,000	26 (31.3%)
¥100,000 -200,000	13 (15.7%)
¥200,000- 300,000	5 (6.0%)
¥300,000- 400,000	4 (4.8%)
¥400,000 -500,000	2 (2.4%)
More than ¥500,000	5 (6.0%)

*¥1 ≈ $0.15. E.g., ¥50,000 ≈ $7,500.

Participants in the quantitative survey

Participants recruited for the quantitative survey were parents whose children were receiving intervention services at autism organizations that provided a variety of services including ABA-based intervention programs. Eighty-five parents were included in the data analysis, and each of them had at least one child diagnosed with autism (71 boys, 83.5%; 14 girls, 16.5%). Sixty-three (74.1%) and 22 (25.9%) children had confirmed and suspected ASD diagnoses, respectively. The average age of these children was 58.4 months old (SD = 24.1). Fifteen parents who participated in the survey were excluded from analyses either due to incomplete survey responses or because their children had not yet received an autism-related diagnosis. Over half of the included parents (52.9%) were from northern China, and the remaining were from eastern, south-central, northeastern, northwestern, or southwestern China.

Table 1 shows these families’ demographic information. The distributions of participants’ educational level and annual family income were comparable with those in a report based on a national sample (China Association of Persons with Psychiatric Disability and their Relatives 2014) and other surveys about Chinese children with developmental disorders (Peking University Care Brain Health 2022). To estimate a family’s SES level, the educational level and family income were combined to generate a composite index, using the exploratory factor analysis with maximum likelihood estimators that can generate a factor score for each family (e.g. Braveman et al. 2005; Noble et al. 2015). A factor score lower than the mean level (zero) was classified as a low socioeconomic level, and a factor score equal to or above the mean level (zero) was classified as a middle/high socioeconomic level. Using this index, 41 and 42 participants in the current sample were classified as low and middle/high SES, respectively.

Participants in the qualitative semi-structured interview

There were two types of parents involved in the semi-structured interview, i.e. parent founder interviewees (n = 3; No. 8-10 of Table 2), and parent interviewees (n = 7; No.1-7 of Table 2). Among the interviewees, the three parent founders had their own autism organizations, whereas the seven parent interviewees did not have such organizations. The average age of all interviewees was 38.2 years old (25-58), and their children’s average age was 14.3 years old (3.3-30).

Table 2.

Information of parent interviewees.

No.	Name	Gender	Marriage status	Age	Education	Place of residence	Child’s age (yrs. and mos.) and gender
1	Qian	Female	Married	37	Technical College	Beijing (North)	6y, Female
2	Xiang	Female	Married	36	Bachelor	Hunan (South)	6y6m, Male
3	Chang	Female	Married	33	Primary School	Heilongjiang (North)	5y, Male
4	Bing	Female	Married	35	Junior High School	Guangzhou (South)	5y, Male
5	Ling	Female	Married	34	Bachelor	Shanxi (North)	3y4m, Male
6	Wuling	Female	Married	29	Technical Secondary School	Shanxi (North)	5y8m, Male
7	Meimei	Female	Married	25	Senior High school	Hebei (North)	4y3m, Male
8	Chun*	Female	Divorced	58	Bachelor	Beijing (North)	30y, Male
9	Jing*	Female	Married	42	Bachelor	Beijing (North)	13y, Female
10	Zheng*	Female	Married	53	Master	Qingdao (East)	24y, Male

Note. * A parent-founder interviewee indicate a participant had a child with ASD and found a local autism organization.

Sampling strategy

A purposive sampling strategy was primarily employed to maximize the representativeness of the sample. The participating parents had children who had opportunities to receive various services, including ABA-based intervention programs. The ABA-based intervention programs were defined as any programs that involved ABA-based interventions or the use of ABA strategies (e.g. DTT, NET, and FBA). The target behavior of the program can be comprehensive (covering multiple targeted areas) or target-specific (e.g. focusing on social skills); the intensity of the program can be intensive or non-intensive; and the setting of the program can be home-, center- or school-based. The majority of the survey data were collected from five large-scale autism organizations in Beijing and surrounding areas that provided a diverse range of treatment approaches. These organizations also had a large number of children receiving interventions. Parents in these organizations had opportunities to experience and compare various treatment approaches. Additionally, to increase the sample size of this study, a small number of data was collected online through social media in Beijing and surrounding areas.

In the semi-structured interview, participants were selected to ensure heterogeneity in their educational backgrounds and age levels. This included parents with both low (junior high school or below) and high educational levels (bachelor’s degree or above). Furthermore, we also purposively selected parent interviewees who had a child with ASD and also founded their autism organizations (i.e. parent founder interviewee), as well as parent interviewees who did not found an organization, because the two groups of parents may have different understandings and experiences about autism interventions, as well as different life experiences. For example, the parent founder interviewees were able to provide more comprehensive information regarding the use of ABA-based interventions in organizations.

Research material and procedure

The quantitative survey was modified from an Australian survey (Horiguchi 2014). The items of the survey reported for the present data analyses included: demographic information of parents (age, gender, education level, and annual income) and children (chronological age and gender), popularity, patterns of intervention delivery (i.e. formats of delivery, parents’ involvement in interventions), costs (i.e. direct expenses and traveling costs), and parents’ evaluations on ABA-based interventions (5-point Likert scales: generalization of children’s learned behaviors, and the level of support provided by professionals to children during intervention; see Appendix).

The directors of the five autism organizations were informed with the purposes and details of this study, along with consent forms. Then, the directors distributed the recruitment information, consent forms, and questionnaires to all families who were receiving interventions at organizations. Additionally, seven parents who previously completed the quantitative survey were also invited to participate in the semi-structured interview. Three parent founder interviewees were also recruited from the five autism organizations.

Data checking and clearance were implemented before data analyses. Given that participants’ data in the quantitative surveys were generated through selecting from prepared options for each item, no extra process was done to code these data. Two authors double-checked the data to minimize the potential input errors, and independently analyzed the data to identify and rectify any inconsistences or errors. Extending the quantitative survey, a qualitative semi-structured interview was designed to further ask parents to report their experiences, understandings, and opinions on autism- and ABA-related services. Moreover, additional questions were prepared for the three parent founders, including describing the characteristics of the services their organizations provided and the reasons for delivering these services. As a structured classification had been developed (See Figure 1), the qualitative data analysis of this study used content analysis, a systematic categorizing approach to describe and analyze phenomena (Krippendorff 1980). The procedure of content analysis consisted of three steps (Elo and Kyngäs 2008): (1) preparation, in which the first author organized the data and reviewed the texts; (2) organization. Themes (e.g. the theme of “experiences of patterns of the intervention delivery” was coded from the survey) and subthemes (e.g. subthemes: “involvement of parents in interventions” and “delivery format”) were initiated and each interviewee’s response to the subthemes was quantified to obtain an objective description. To build the trustworthiness of the qualitative themes, the first author discussed each theme and sub-themes with the second author, an expert in the field of ABA but did not conduct the interview directly, and (3) report generalization.

To establish the inter-rater reliability, five (50%) interviewees’ transcripts were randomly selected and then coded independently by two authors. The inter-rater reliability (percentage agreement) was calculated as the number of coding agreements divided by the sum of all agreements and disagreements (Campbell et al. 2013). An agreement was defined when both raters coded an interview as either containing a statement corresponding to a theme or containing no statements corresponding to a theme (e.g. Grindle et al. 2009). Our inter-rater agreement was 100% across all themes.

Results

Quantitative survey

Popularity of ABA-based intervention service

According to 82 parents’ valid reports, ABA-based intervention service was the most popular intervention approach (n = 65, 79%) among 13 types of approaches, followed by sensory integration (n = 51, 62%), play therapy (n = 30, 37%), and others (Table 3). Fifty-four parents (66%) reported that their children had received two or more services, and two (2%) had not yet received any of the services listed in the survey.

Table 3.

Popularity of ABA-based services (n = 82).

Services	Number of children	Percentage
1. ABA-based programs	65	79.3%
2. Sensory integration	51	62.2%
3. Play therapy	30	36.6%
4. TEACCH	17	20.7%
5. Music therapy	15	18.3%
6. Auditory integration	8	9.8%
7. Acupuncture and massage	4	4.9%
8. Biotherapy	3	3.7%
9. Occupational therapy	2	2.4%
10. Hyperbaric oxygen therapy	2	2.4%
11. Arts learning	1	1.2%
12. Mind reading	1	1.2%
13. Animal-assisted	0	–

Non-parametric Chi-square tests showed that any choices of autism services were not different between the low (n = 39) and the middle/high (n = 41) SES groups (χ² ranged from 0.16 to 3.20, n.s.). We also examined if the choices of any autism services differed for children with an identified diagnosis of autism or with suspected autism, and found no difference between the two groups (χ² ranged from 0.01 to 3.54, n.s.), except for the choice over the play therapy (χ² = 4.39, p = 0.04; more families with the identified diagnosis than with suspended diagnosis chose the play therapy). However, given that this difference was the only one among all 13 types of intervention approaches and that only 30 parents chose play therapy, this result should be interpreted with caution.

Additionally, for parents who chose the ABA-based service, we did not find differences between the identified and suspected autism groups in any of the following aspects, including costs (χ² = 1.57, n.s., n = 64), parents’ evaluations (χ² = 1.09, n.s., n = 62, for supportiveness of professionals; χ² = 2.92, n.s., n = 63, for behavior generalization), and choices of delivery formats (9 delivery formats available. χ² ranged from 0.32 to 3.26, n.s., n = 63; except that more families in the suspected group (n = 3) than the diagnosed group (n = 1) chose the ABA playgroup. This difference was not interpreted due to the very small numbers). As the difference between the identified and suspected groups was not significant in most aspects, the two were not distinguished in the following analyses or discussions.

Patterns of ABA-based intervention delivery

Intervention delivery formats. Sixty-five parents reported the delivery formats of ABA-based intervention services they chose (Table 4). The top three formats were home-based 1:1 programs (n = 37, 59%), center-based 1:1 programs (n = 35, 56%), and center-based group programs (n = 27, 43%). In contrast, the formats not often used were kindergarten/school support from ABA consultants (n = 0), kindergarten/school support from ABA therapists (n = 1, 2%), and school-based 1:1 programs (n = 1, 2%). Thirty-nine children (62%) had tried two or more formats. The choices over delivery formats were not different between low (n = 28) and middle/high SES groups (n = 36; χ² ranged from 0.03 to 2.29, n.s.).

Table 4.

Delivery format of ABA-based programs (n = 65).

Modes	Number of children	Percentage
1. Home-based, 1:1 program	37	58.7%
2. Center-based, 1:1 program	35	55.6%
3. Center-based, group program	27	42.9%
4. Multidisciplinary program (ABA + Speech + OT)	4	6.3%
5. ABA playgroup	4	6.3%
6. ABA sports program	3	4.8%
7. Mainstream school-based, group program	2	3.2%
8. Mainstream school-based, 1:1 program	1	1.6%
9. Kindergarten/school support from ABA therapies	1	1.6%
10. Kindergarten/school support from ABA consultation	0	–

Parents’ involvement in interventions

The majority of parents (n = 51, 79%) worked as assistant or complementary therapists for their children at home. The content of the intervention usually included teaching children basic living skills (e.g. toileting and personal hygiene skills) and some cognitive skills (mostly using DTT to teach cognitive and social skills). Twenty-five parents (38%) spent more than five hours weekly. Potential reasons for their involvement included: familiarity with the child (n = 39, 81%), willingness to improve skills and knowledge in ABA (n = 30, 63%), challenges in accessing professional therapists (n = 20, 42%), and financial pressure (n = 19, 40%). A higher proportion of parents in the low SES group (25 out of 28) worked as assistant therapists copared to the middle/high SES group (25 out of 36; χ² = 3.63, p = 0.06).

Costs of ABA-based interventions

Financial costs and resources

Fouty-eight (75%) families annually spend at least ¥25,000 (≈$3,750) on ABA-based interventions. The expense was not statistically different between the middle/high (30 out of 36 families spent ¥25,000 ≈ $3,750/year) and low SES groups (17 out of 27 families spent ¥25,000 ≈ $3,750/year; χ² = 11.35, p = 0.12). Regarding the financial resources (Table 5), a majority of the families (n = 60, 94%) relied on salaries to pay for the ABA-based interventions. Some families also used loans from relatives or friends (n = 10, 16%), personal investment (n = 9, 14%), or governmental financial support (n = 7, 11%). Notably, eight families (13%) indicated that they spent more than 60% of their annual family income on interventions, and another 14 families (22%) spent about 30%-60% of their annual income. In short, over one-third of families spent 30% or more of their family income on children’s interventions.

Table 5.

Economic resources for ABA-based services (n = 65).

Services	Number of families	Percentage
1. Personal salary	60	93.8%
2. Loans from relatives/friends	10	15.6%
3. Personal investment	9	14.1%
4. Governmental financial support	7	10.9%
5. Additional (part-time) employment	5	7.8%
6. Sold assets	2	3.1%
7. Loans from banks	1	1.6%
8. Retirement pension	1	1.6%
9. Remortgaged house	1	1.6%
10. Cashed in leave entitlement	0	–
11. Inheritance	0	–
12. Fundraising	0	–

Traveling cost

Twenty-eight parents (44%) took long-distance travel to access ABA-based interventions. More families reported long-distance travel in the middle/high SES group (21 out of 36) than in the low SES group (7 out of 27; χ² = 9.95, p =0.04).

Evaluations for ABA-based interventions

Supportiveness of professionals’ intervention work

The leveld of professionals’ support for children’s interventions was evaluated on a 5-point Likert scale ranging from 1 (Not at all) to 5 (Very much). The average score was 3.68 (SD = 1.16). Seventeen parents (27%; low SES = 11; middle/high SES = 6) reported the highest level of support (scale score = 5), and 22 (36%; low = 12; middle/high = 9) reported a moderate level of support (score = 4). The other 12 (19%; low = 3; middle/high = 9), eight (13%; low = 0; middle/high = 8), and three (5%; low = 1; middle/high = 2) parents gave the evaluation at the levels of somewhat (score = 3), a little (score = 2), and not at all (score = 1), respectively. The low SES group (M = 4.19) gave a significantly higher evaluation for the supportiveness of professionals than the middle/high SES group (M = 3.26; t = 3.31, p = 0.002). This result was further confirmed by a Chi-square test that showed the distributions of the selection over the five options differed between the two SES groups (χ² = 12.60, p = 0.01).

Extent of generalization

The extent to which children’s behaviors, learned during ABA-based interventions, generalized into their daily lives was evaluated by parents on a 5-point Likert scale ranging from 1 (Not at all) to 5 (Very much). The average score was 3.57 (SD = 0.98). Eleven parents (18%; low SES = 4; middle/high SES = 6) reported the highest level of generalization (scale score = 5), and 24 (38%; low = 11; middle/high = 13) reported a moderate extent (scale score = 4). The other 19 (30%; low = 7; middle/high = 12), eight (13%; low = 4; middle/high = 4), and one (2%; low = 1) parents rated the generalization level as somewhat (score = 3), a little (score = 2), and not at all (score = 1), respectively. Parents’ evaluation on generalization was not different between low (M = 3.48) and middle/high (M = 3.60) SES groups (t = −0.47, n.s.). This result was further confirmed by a Chi-square test that showed the distributions of the selection over the five options did not differ between the two SES groups (χ² = 1.88, n.s.).

Qualitative interview

Each interviewee was coded in the format of pseudonym-age, and the three parent founder interviewees had an additional marker (i.e. name-age-parent founder).

Views of the popularity of the treatment approaches

Consistent with quantitative data showing the high popularity of the use of ABA-based interventions, all seven parent interviewees reported that their children were receiving or had received these interventions. During the interviews, they also reported that they heard about ABA-based interventions through other parents (n = 1), internet (n = 1), or doctors (n = 5). Notably, six interviewees reported they “got lost” in choosing interventions or therapies. Chang recalled doctors’ recommendations of various treatment approaches to her.

“The doctor [from local municipal hospital] suggested several treatment approaches for my child, including mouse nerve growth factor, cerebral protein, sensory integration therapy, and brain electrotherapy… He also suggested that I join a waitlist for ABA-based interventions …Another doctor [from a provincial capital hospital] prescribed medication and cerebral protein, asserting that ‘[they are] beneficial for brain health’ … However, I feel some of the recommendations are false claims.” (Chang-33)

Interviewees reported that they had tried several approaches, including ABA-based interventions (e.g. DTT), sensory integration, auditory integration, music therapy, arts, or medical therapy. Chun (parent founder-58) said that many children tried diverse treatment approaches before coming to their school to receive ABA-based intervention, which was the main approach used in the school.

“This is autism Jianghu … There were a lot of people “passing off fish eyes as pearls”. We had no predecessors [to learn from].” (Chun-58-parent founder)

“Jianghu” (江湖, jiāng hú) literally translates to “rivers and lakes” and metaphorically refers to charlatans and quacks in the field of autism treatment. The phrase “passing off fish eyes as pearls” means there are many people who claim to be professionals in autism interventions but actually have no or little experience or knowledge about autism services, and the professionals also show a great disparity in the qualities of interventions.

After comparing various approaches, parents gained experiences and formed their own opinions. Some preferred ABA-based interventions because “…it [ABA] is not only for children with ASD but can be used by anyone.” (Meimei-25)

Experiences of patterns of the intervention delivery

Intervention delivery formats

The center-based ABA intervention service was reported by six parent interviewees (Qian-37, Xiang-36, Chang-33, Bing-35, Ling-34, and Wuling-29), in which their children had received the 1:1 and/or group programs. Parents understood that the 1:1 program helped children quickly learn cognitive skills in a structured environment (i.e. instruction usually occurred at a table with a high rate of presentation of trials) and the group program allowed children to interact with others in social activity classes.

Interestingly, our quantitative survey showed that home-based interventions had been used often, and all seven parent interviewees reported that they utilized home-based 1:1 intervention almost every day. In our survey, we asked parents whether they used home-based interventions but did not distinguish whether the therapists in the home-based interventions were parents themselves or professionals from organizations. Given the very low staff-to-child ratio in China, the high proportion of home-based intervention may be largely due to parents’ involvement (further discussion on the motivations for parents’ involvement in interventions is presented later).

In the home-based interventions, parent interviewees used ABA-based intervention skills they learned to teach children about social rules and real-life manners at home and in community settings. They understood that an advantage of the home-based interventions was that they could facilitate the generalization of children’s learned behaviors into the natural environment.

“At home, our teaching focuses on social rules, such as how to interact with others, and some principles of etiquette during social interactions… However, when it comes to academic skills, I will resort to table [training].” (Xiang-36)

Table training means that the teaching occurs on a table between an adult and a child. Meimei and her husband also took turns teaching their son at home because “it helps generalize his skill to the natural environment.”

Parents’ involvement in interventions

All seven parent interviewees worked as assistant therapists. Importantly, the interviewees thought that the involvement of parents as therapists was a “double-edged sword.” The advantages included that parents had opportunities to learn how to manage their children’s behaviors in daily circumstances. In addition, parents’ involvement can also relieve their financial burden as they can work with their children themselves for some time, which shortens the intervention programs they must pay for. However, their involvement also had disadvantages, as they had to temporarilly suspend their employment for their children’s intervention programs.

“[The intervention] skills are applicable in our daily life. For example, if the child misbehaves and hits others while playing downstairs, I know how to guide him. Teachers cannot be around to monitor our child all the time.” (Chang-33)

“I sometimes feel upset about my child… My entire life has been disrupted and I am unable to work.” (Wuling-29)

The three parent founders described their training programs that were aimed at improving parents’ knowledge and skills of ABA-based interventions and other approaches. Chun’s (parent founder −58) school provided entry-level knowledge of ABA-based interventions and other approaches like structured Treatment and Education of Autistic and Communication Handicapped Children (Mesibov et al. 2005) for parents of children with autism (mostly aged 3-6 years old). In the ABA-based training program, parents were required to bring their children to school to learn basic intervention skills through an 11-week course. The school had around 50 parents enrolled each term, and many of them came from places out of Beijing. Chun considered this training model “a natural outcome of adapting to Chinese society and meeting the needs of our families” because many parents wanted to help children by themselves in addition to professionals.

Training programs offered by Jing’s (parent founder −42) center usually lasted for about three-and-a-half months. The program included classes like individualized ABA training (e.g., DTT), sensory integration facilities, arts, and physical education, with each session lasting 30-35 min. Parents were required to assist their children in some courses, for example, assisting children with their lunch, escorting them to the toilet, or prompting children’s interactions with peers in social interaction classes. Additionally, six (Qian-37; Xiang-36; Chang-33; Bing-35; Ling-34 and Wuling-29) of the seven parent interviewees reported that their children had previously participated in intervention programs similar to the one offered by Jing’s center. These programs offered a variety of classes, which included different approaches, such as ABA-based intervention classes. Each class typically ran for 30 to 35 minutes and spanned over several months, usually around three months..

Zheng’s (parent founder −53) school had about 300 students (most aged 3-6), and most of them enrolled in a course for three months. The school provided classes including ABA-based intervention, sensory integration, speech and language therapy, and academic courses like math. The school included individualized (1:1) intervention (between a teacher and a student) and group training classes. The group training involved teachers, parents, and students, where the parents were required to assist or prompt children’s behaviors. Parents received ABA knowledge training sessions every weekend.

Financial cost and resources

All seven parent interviewees reported the primary source of funds for their treatment expenses was their personal income. For instance, Qian and Ling reported that they spent $1,200 and $1,500 per month, respectively. In addition, two parents sought external financial supports, with one recieving support from family members (Meimei-25) and the other resorting to credit cards (sometimes even leading to credit card overuse; Wuling-29).

“Given our annual income, it is actually quite ridiculous. We hardly have any savings left because last year we spent $11,250 in just five months, which almost wiped out all our savings… Fortunately, his grandparents generously offered us financial support and are still helping us this year.” (Meimei-25)

Traveling cost

All seven parents (No. 1-7) who were not organization founders had long-distance travels to Beijing for interventions. Many regions, especially small towns or the countryside, did not have enough qualified service providers, and thus many families had to travel to cities far from home to access services.

“I understand that the age range of 3 and 6 is a critical period for a child with ASD. Therefore, I prioritize providing my child with professional training outside of our home during this period. However, if I were to train him at home, I might only be able to dedicate 50% of my energy to his development. That is why I would prefer to enroll him in interventions outside of our home.” (Wuling-29)

Perceptions of ABA-based interventions

Supportiveness of professionals’ intervention work

The parent interviewees showed inconsistent evaluations of professionals’ intervention work with their children. Two parents (Ling-34 and Xiang-36) thought professionals’ work was supportive. Xiang said: “He [her son] likes this teacher and always follows her instructions.” The other four parents had some concerns about professionals’ work, even though they agreed that their children’s skills were improved (see the following subtheme). Specifically, three parent interviewees (Qian-37, Wuling-29, and Meimei-25) thought that the teaching abilities varied across therapists in organizations. Chang reported that some therapists’ intervention work was not well organized.

“The therapist is the most important part of ABA …Professional therapists have excellent opportunities for career development. They often stay at one center for one or two years and then they leave and open their own centers after they have gained sufficient experience and expertise. This is why China’s autism organizations often appear disorganized, as there is a lack of management and planning. While there is a large quantity of therapists, the quality of teaching is poor.” (Meimei-25)

“One teacher works with five children in a social game class. The teacher could barely pay attention to my son and he was often left behind.” (Chang-33)

Children’s behavioral changes

All parent interviewees reported positive changes in children in prerequisite skills (e.g. waiting, turn-taking, and sharing skills), academic and language skills, emotional expression, safety awareness, and social interactions, after receiving the ABA-based interventions. In addition, sleeping and feeding problems and self-injurious behaviors decreased after interventions.

“I have noticed that she smiles more often now and seems to have developed a broader range of interests. She is also better at expressing her feelings and emotions and can now communicate with others to some extent, using either eye contact or language.” (Qian-37)

“He used to run aimlessly back and forth on the street without any regard for safety. But now, I feel more confident in letting him walk independently. He can follow my instructions, and I can tell him to ‘come here’ and ‘walk with Mummy step by step’, which he can now do successfully.” (Wuling-29)

Discussion

Using a quantitative survey and a qualitative semi-structured interview, the present study investigated the implementation of ABA-based services in Beijing and surrounding areas of China. The results showed that the ABA-based interventions were the most popular approach compared with other approaches. The delivery of ABA-based intervention was mainly center- (i.e. autism organizations/clinics/schools) or home-based, and interventions were rarely used in mainstream school-based environments. The costs of interventions (not limited to ABA-based intervention) were a financial challenge to many families. A high proportion of parents were involved in interventions as assistant therapists. In addition, most parents had positive evaluations of the effectiveness of the ABA-based interventions in improving children’s behaviors and were satisfied with the work professionals engaged in with their children. Furthermore, some aspects regarding the use of ABA-based interventions (i.e. traveling costs and parents’ evaluations on the level of support professionals provided to children) varied with families’ SES levels.

Popularity of ABA-based interventions

Autism organizations in the Chinese mainland typically offer multiple intervention approaches (Shenzhen Autism Society 2013). According to our survey, ABA-based interventions were widely adopted by parents (79%) and were the most popular choice among 13 types of intervention approaches. Furthermore, the preference for the ABA-based interventions did not differ between families with low or middle/high SES levels, consistent with a study with a larger sample in the USA (Pickard and Ingersoll 2016). In addition, the finding that ABA-based interventions ranked as the most popular intervention is consistent with previous research (Sun et al. 2013) but at odds with others (Shenzhen Autism Society 2013). The majority of the sample in Sun et al. (2013) was from the north of China, as was the case for our study, whereas over half of the sample of the Shenzhen survey was from the south. These findings together suggest the use of the ABA-based intervention service may vary across regions.

This regional discrepancy may be due, in part, to that ABA-based interventions were introduced in northern Chinese regions first (Zhou et al. 2014), and that the north has more resources available than other regions. For instance, the international qualification for professionals providing behavior-analytic practice and consultation services includes Board Certified Behavior Analysts/-Doctoral level (BCBA/BCBA-D) and Board Certified assistant Behavior Analysts (BCaBA) (Shook and Favell 2008). There are significantly more registered BCBA/BCBA-D and BCaBA certificants in northern China (e.g. 196 in Beijing) than the southern region (46 and 43 in two of the biggest cities in the south, Guangzhou and Shenzhen, respectively; Behavior Analyst Certification Board 2022). These may result in the higher popularity of ABA in our survey. Discrepancies across regions have been also reported in Western studies (Monz et al. 2019). All of these findings suggest researchers collect data from a variety of regions and have a larger and nationally-represented sample to gain a more accurate and comprehensive understanding of the use of ABA-based interventions in the Chinese mainland.

The results from the survey showed that parents typically chose multiple intervention approaches, and our qualitative interviews further showed that many parents experienced difficulties in choosing from intervention approaches, consistent with previous reports (Shenzhen Autism Society 2013, Wucailu Autism Research Institute 2017, Peking University Care Brain Health 2022). Our interviews revealed that many parents lacked knowledge about evidence-based intervention practices (e.g. the quotations from Chang-33 and Chun-parent founder-58; Sullivan and Wang 2020). Unfortunately, they got into the “Jianghu” of autism (metaphorically refers to charlatans and quacks in the field of autism) without foundamental knowledge of this field. Most parent interviewees reported that they received recommendations from doctors, who were the primary information resource about autism interventions. However, as noted in the quotations, some doctors suggested various intervention approaches but were unclear about which specific approach would be most suitable for a particular child with ASD. Parents’ reports suggest that doctors who diagnose children with autism are often the first professionals in the field of autism that these families encounter, and the opinions provided by doctors significantly influence the treatment decisions of these families. Therefore, doctors play a critical role in providing therapeutic guidance, highlighting the need for specialized training for medical professionals including psychiatrists and pediatricians. We recommend that medical professionals proactively engage in continuing education programs focusing on evidence-based approaches including ABA (e.g. Sam et al. 2020). These programs should emphasize the customization of interventions based on the individual characteristics of children. Moreover, medical schools in China can provide evidence-based intervention programs for medical students, or incorporate these contents into their curriculums. By doing so, doctors can provide more individualized recommendations to families for making therapeutic decisions, ultimately reducing their time and financial commitments required.

Furthermore, kindergarten or preschool environments are also important sources, as attentive teachers can also refer children to doctors. Thus, strategically disseminating therapeutic and educational information in these settings will effectively reach families with special needs. Additionally, we emphasize the effect of school psychologists. These professionals have a comprehensive understanding of available options in the field, and can also bridge families with special needs to hospitals, universities, school settings, and local communities (Forlin 2010, Wong et al. 2014). They can be a valuable source for parents who are unable to find suitable resources themselves.

Patterns of ABA-based intervention delivery

Intervention delivery formats

The choices over intervention delivery formats were not different between families with different SES levels. Center-based 1:1 (59%) and home-based 1:1 programs (56%) were the main formats of intervention delivery, consistent with some previous findings (Horiguchi 2014). It was surprising to see the frequent use of the home-based 1:1 format, given the very low staff-to-child ratio in China. However, the qualitative semi-structured interviews suggested that this high proportion might be due to parents’ involvement in interventions at home (Our survey did not distinguish whether the therapists of the home-based 1:1 program were professionals or parents). The motivations for parents’ involvement were examined in our interviews and will be discussed later.

In addition, the center-based group programs were used by about 40% of parents in our survey. This format is important because it provides children with ASD more opportunities to interact with their peers, which facilities children’s learning and practice of social communication skills (Leaf et al. 2018). However, the qualitative interview showed that some parents were concerned about some therapists’ capacities for handling a group of children. Therapists are required to improve their abilities to effectively manage group activities and implement evidence-based interventions. In addition, therapists can also recruit group members based on children’s skill assessment and include typically developing children as role models in social skills groups (e.g. Leaf et al. 2010).

Notably, mainstream school-based programs were rarely used (less than 7%), even though evidence exists that these programs benefit children’s acquisition and generalization of intellectual, social, and language skills (Grindle et al. 2012; Reed et al. 2007). These programs have been widely used in some countries (e.g. Suhrheinrich et al. 2020), suggesting that the involvement of ABA-based and other interventions in inclusive education is viable (Li et al. 2023). The very low frequency in our study suggests that ABA-based interventions have not been well integrated into inclusive education in the Chinese mainland (Liao et al. 2020). In fact, other autism-related treatment approaches (e.g. structured teaching) have not been sufficiently used in mainstream schools either (Li et al. 2022, Liu et al. 2016). This is likely because the current education settings and relevant policies have not been well prepared to include these interventions in mainstream schools. Given the unique advantages of mainstream school-based intervention programs, we encourage service providers, researchers, and educational policy-makers to explore how to integrate ABA-based interventions in mainstream education (e.g. Grindle et al. 2012).

Parents’ involvement in interventions

A significant characteristic of using ABA-based interventions in our sample was that most parents (79%) were involved in interventions as assistant therapists, and one-third of parents even worked more than five hours per week as assistants. This proportion in our sample is higher than that reported with samples from Western countries (e.g. 52% Dillenburger et al. 2012). Moreover, our data showed that parents with low SES were more likely to be involved in interventions, compared to parents with middle/high SES.

According to our interviews, parents’ involvement was due to their familiarity with the needs and characteristics of their children. Parents also wanted to improve their skills and knowledge in the field of ABA. Additionally, for approximately 40% of respondents, financial pressures and challenges in accessing professional therapists motivated parents to work as assistant therapists. This is consistent with our findings on the discrepancy between the two SES levels. In short, parents’ decision to engage in training is driven by both subjective reasons (leveraging their familiarity with their children and a strong desire to do more for them) and practical reasons (reducing the cost of treatment and compensating for the shortage of professionals).

For these reasons, it is appropriate to consider parents’ involvement a viable alternative or complement to current intervention procedures. Our interviews showed that service providers had noticed this interest in parents and offered training programs to educate parents in ABA knowledge and practical skills. For example, Chun’s autism school offered entry-level knowledge in ABA-based interventions for parents through an 11-week-long course. In the future, service providers also need to design more programs to give parents opportunities to practice newfound skills (e.g. prompting a child to follow instructions; McCabe et al. 2023), and furnish themwith constructive feedback (Steiner et al. 2012).

It is noteworthy that our interviews also revealed some shortcomings of parents’ involvement as assistant therapists. For example, some parents had to take time away from their jobs/work due to their involvement in interventions (as indicated by the three parent founders). It was indeed hard for them to balance work and interventions. There are several ways to alleviate the time pressure on parent who act as therapists, especially for those who may not desire this role but find no other options. For example, some time-efficient training methods, such as online or video-based parent-training programs (Bearss et al. 2018, McDevitt 2021), have been developed to help parents learn intervention skills at home, allowing for more flexible scheduling. Nevertheless, to further ease the burden on parents, it reamins essential to provide training for more qualified professionals through diverse resources.

Cost of ABA-based interventions

The cost of using ABA-based intervention service is a heavy financial burden for many families in our sample, and this finding is consistent with previous studies (Zhou et al. 2022). Seventy percent of the parents in our sample reported that they spent at least ¥25,000 (≈ $3,750) per year on interventions, which was already higher than the median national annual disposable income per capita in China (¥19,281 ≈ $2,892) in 2015 (National Bureau of Statistics 2016). Moreover, intervention costs are not covered by health insurance, with families only receiveing limited reimbursement. In addition, more than 40% of families had to travel to major metropolitan areas far from their homes to access qualified interventions, greatly inflating costs (e.g. travel expenses and accomondation rentals).

Additionally, there was no significant difference between families with low and middle/high SES concerning the direct costs of ABA-based interventions (e.g. payments to organizations). However, compared to families with low SES, more middle/high SES families traveled long distances to access ABA-based interventions, likely because of their improved financial circumstances, which enable them to cover these indirect costs. In other words, this implies that low SES families face more challenges to access high-quality intervention services. In light of this situation, policymakers should coordinate and balance regional development, especially aiding families from remote and rural areas.

Furthermore, we strongly suggest researchers and service providers explore cost-effective intervention service models to alleviate the financial strain on families. For example, low-intensity ABA-based intervention (5.5- 18.8 h/week) can lead to positive outcomes for children with ASD and this can also help reduce costs (Eldevik et al. 2012, Grindle et al. 2012, Haraguchi et al. 2020). Another way to reduce costs is to apply ABA-based interventions in a wider range of settings where staff are available, such as special needs schools (Foran et al. 2015), autism-specific educational programs (Lambert-Lee et al. 2015), and mainstream education (Grindle et al. 2012). This would require no additional staff cost once staff have been trained in ABA. Finally, a recent national policy of China, the Fourteenth Five-Year Plan (2021-2025), has emphasized the importance of supporting people with disabilities, including ASD (State Council 2021). As part of this, the government continues promoting inclusive education for children with ASD and building more autism-specific schools. More families will benefit from these policies, and their financial burdens will be relieved.

Evaluations for ABA-based interventions

Our survey indicated that most parents gave positive evaluations on the generalization of children’s behaviors and the supportive nature of professionals’ work during interventions. The interviews further showed that parents believed that their children gained new skills and that their challenging behaviors decreased after ABA-based interventions. These findings are consistent with some Western studies (Dillenburger et al. 2012; Rosales et al. 2021).

Despite the overall positive evaluations, parents also pointed out some potential limitations of ABA. For example, while ABA-based interventions are evidence-based, some have questioned the long-term effect of behavior generalization after interventions (Magiati et al. 2014). In the present study, some parents rated the extent of generalization as “low” or “very low” (15%). In fact, some Chinese parents also expressed concerns through social media about the limited generalization of some ABA-based programs in children’s learned behaviors and skills (Xiaoyaya 2016). In addition, a few parents provided less favorable evaluations of the intervention, as some reported that ABA professionals’ interventions were “little supportive” or “not supportive at all” (18%). These negative evaluations are assocaited with various factors, and one of them may be the limited capacities of ABA professionals. This is supported by the qualitative interview, which indicated that four parent interviewees had some concerns about the professionals’ work, such as the large variability in teaching abilities among therapists. Another possible reason is the shortage of qualified staff available for supervising interventions. Due to the limited number of supervisors capable of designing more advanced programs and flexibly adjust programs, children may quickly achieve a plateau and show no further improvement, which subsequently impacts the effectiveness of interventions (Mandell et al. 2013) and raises parents’ concerns about the outcomes of ABA-based interventions.

Overall, although most parents expressed positive assessments of ABA-based interventions, it is crucial to give due considerations to their negative evaluations because the quality of professionals’ work strongly influences children’s behavioral outcomes (Pellecchia et al. 2015, Durlak and DuPre 2008). It is imperative to train more well-qualified supervisors and therapists who can develop more appropriate and flexible intervention programs that meet the unique needs of children with ASD.

Finally, parents with middle/high SES gave relatively lower scores for the quality of professionals’ work than parents with low SES. This result is consistent with the study in the USA (Hidalgo et al. 2015). It may be that the middle/high SES families possess a better understanding of what constitutes optimal practice, which subsequently results in higher expectations and more stringent criteria for professional services (Hidalgo et al. 2015).

Limitations and suggestions for future research

The present study has several limitations. One is about the relatively small sample size, and participants were primarily from the northern regions of China (Beijing and surrounding areas). A larger sample would be more representative of the families with children with ASD. Moreover, the participants of our study were predominantly sourced from a few specific organizations in or around Beijing. Although these organizations offered various treatment approaches and many of these participants had sought interventiosn from multiple resources, it is still necessary for future studies to recruit a more diverse range of families from various organizations and areas, as well as those with different experiences about autism treatments (e.g. recruiting participants through medical professionals or at school settings). Based on the larger and more representative sample, we can further examine whether the present findings on the popularity of treatment approaches and the delivery format of ABA-based interventions can be generalized to the broader population of Chinese families with children with autism.

Additionally, this study did not refer to the duration and intensity of the ABA-based intervention sessions that children typically received. It is possible that the duration and intensity of receiving intervention services were influenced by parents’ perceptions of its effectiveness and/or their financial capacities. This information would give parents a more accurate understanding of ABA-based intervention services, helping them estimate the time and financial resources required for these interventions. The qualitative reviews showed that the interventions for children were relatively short (typically lasting around three months), and this may be due to a lack of qualified staff to supervise these programs, making it difficult for therapists and parents to maintain long-term interventions. Therefore, it is important to gather information about the duration and intensity of ABA-based intervention programs to better understand clients’ needs and obstacles.

Finally, this study focused on parents’ experiences and evaluations of the effectiveness of ABA-based interventions but did not include children’s actual outcomes after interventions. It is plausible that parents’ evaluations may not align consistently with their children’s actual outcomes because their subjective evaluations may be biased by their knowledge, experiences, and expectations (Goin-Kochel et al. 2009). To attain a more accurate evaluation of the effectiveness of ABA-based interventions, it is strongly recommended to incorporate multiple indicators to assess the overall effects of the intervention. These indicators should encompass parental assessments, therapists’ evaluations, and objective measurements of children’s actual performance using professional scales or experimental methods.

Conclusions

The present study depicted the use of ABA-based interventions for children with ASD in Beijing and surrounding areas, China. The results indicated that parents of these children typically employed multiple autism-related intervention approaches, with ABA-based intervention being the most popular one. Parents also gave overall positive evaluations regarding the effectiveness of ABA-based interventions and the level of support professionals provided to children during interventions. ABA-based interventions were primarily administered at organizations (i.e. autism clinics/schools) and children’s homes, but the ulitization in mainstream schools was very limited. A significant proportion of parents were involved in interventions as assistant therapists. However, the cost of interventions (not limited to ABA-based interventions) posed a financial challenge for many families. Additionally, traveling expenses and parents’ evaluations of the support provided by professionals to children varied with families’ SES levels. Overall, these findings provide valuable insights for service providers, policymakers, and researchers to enhance ABA-based intervention services for Chinese children with ASD.

Appendix.

Sample Questions of Parent Survey

1. Approximately how much do you currently spend on your child/children’s ABA-based programs per year? (Per child, including the costs of materials)

   • Less than ¥1000
   • ¥1000 – 5000 (including ¥5000)
   • ¥5000 -10,000 (including ¥10,000)
   • ¥10,000-15,000 (including ¥15,000)
   • ¥15,000-20,000 (including ¥ 20,000)
   • ¥20,000 -25,000 (including ¥ 25,000)
   • ¥25,000 -30,000 (including ¥ 30,000)
   • More than ¥30,000 (including ¥30,000)
2. How do you pay/have you paid for ABA programs? (Choose all the options you have used)

   • Government financial support
   • Salary
   • Investment
   • Took on additional employment
   • Cashed in leave entitlements
   • Loan
   • Borrowed funds from family/friends
   • Retirement pension
   • Inheritance
   • Sold assets
   • Fundraising
   • Gifted money
   • Remortgaged house
   • Others (Specify:___________)
3. Where does your child undertake ABA-based programs

   • Local town
   • Local city
   • Local province
   • Other cities other than the place of your residential province
   • Other provinces
   • HK or Macaw
   • Overseas
   • Others
4. What kind of services has your child accessed? (Choose all the options your child has accessed)

   • ABA-based programs
   • Sensory integration
   • Play therapy
   • Treatment and Education of Autistic and Related Communication Handicapped Children (TEACCH)
   • Music therapy
   • Auditory integration
   • Acupuncture and massage
   • Biotherapy
   • Occupational therapy
   • Hyperbaric oxygen therapy
   • Arts learning
   • Mind reading
   • Animal-assisted
5. Types of your child’s current ABA program (Choose all the options your child has accessed)

   • Home-based 1:1 program
   • Centre-based 1:1 program
   • Mainstream School-based 1:1 program
   • Center-based group program
   • Mainstream school-based group program
   • Multi-disciplinary program (ABA + Speech + Occupational Therapy)
   • Kindergarten/school ABA consultation
   • Kindergarten /school support from ABA therapists
   • ABA playgroup
   • ABA Sports program
   • Others (Specify: __________)

6. Do you run ABA therapy sessions for your child as a parent-therapist?

   • Yes

   • No (Skip to Q7)

     •   6.1 On average, how many hours of ABA therapy do you do per week? (only count session hours) _____Hours

     •   6.1 Reasons for becoming a parent-therapist

       •    • Familiarity with the child
       •    • To improve skills and knowledge in ABA
       •    • Financial reasons
       •    • Challenges in recruiting therapists
       •    • Others (Specify: _________)
7. To what extent do you find your ABA program’s professionals are supportive to children during interventions

   • Not At All
   • A little
   • Somewhat
   • Moderately
   • Very much
8. To what extent do you find that the skills addressed in therapy sessions are generalized to the child’s daily life?

   • Not At All
   • A little
   • Somewhat
   • Moderately
   • Very much

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Funding Statement

This study was funded by Queen’s University Belfast/China Scholarship Council Ph.D. studentship to the first author under the supervision of the second author. This study was funded by the National Natural Science Foundation of China (32000762), the Science and Technology Program of Guangzhou, China, Key Area Research and Development Program (202007030011), and the Fundamental Research Funds for the Central Universities, Sun Yat-sen University (2023qntd53) to the first author.

Disclosure statement

No potential conflict of interest was reported by the authors.

Author contributions

Author Contributions: Y.L. Conceptualization, Project administration, Investigation, Coding, Analysis, & Writing the original draft. K.D Conceptualization & Supervision. X.L and X.Y. Analysis & Coding. Y. M. Conceptualization, Analysis, Writing & editing

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.