Measuring family outcomes for young autistic children receiving interventions in China: The structure of the Family Outcomes Survey and predicting variables

Abstract

Given the importance of parent involvement and parent-implemented interventions in achieving maximum child outcomes, practitioners need valid measures to assess and monitor family outcomes in developing and delivering effective and sustainable interventions. This study examined the structure of the Family Outcomes Survey, Form A, using a sample of 467 caregivers of young children diagnosed with or at increased likelihood of autism in China and identified significant predictors of the five family outcomes. Findings supported a five-factor model but a poorer fit was reported when the overall family outcome factor was added to the model. Modeling shared variance between some items under Outcome 4, “having support systems,” and Outcome 5, “accessing the community” was found to significantly improve the model fit. Among the analyzed child and family characteristics, the number of hours that the caregiver spent with the autistic child, caregivers’ employment status, family income, and caregivers’ educational attainment were found to be significantly associated with Family Outcomes Survey, Form A scores. Implications of the findings are discussed.

Lay abstract

Efforts to measure, document, and monitor family outcomes can be helpful to practitioners in developing and delivering effective and sustainable interventions. Researchers have developed the Family Outcomes Survey, Form A, for measuring the outcomes experienced by families of children in the early intervention/early childhood special education system. Little has been reported on how well the five outcomes on the Family Outcomes Survey, Form A describe the experiences and expectations of families of autistic children in China. We conducted a survey using the Family Outcomes Survey, Form A, Chinese version with 467 caregivers of young autistic children in China. First, the five-outcome structure of the Family Outcomes Survey, Form A seemed to be appropriate for measuring family outcomes of autistic children in China. We also found that the Chinese caregivers of autistic children seemed to give general lower ratings on all five outcomes on the Family Outcomes Survey, Form A as compared to caregivers of children in early intervention/early childhood special education in Western countries like the United States and Australia. Furthermore, caregivers’ ratings on the five Family Outcomes Survey, Form A outcomes seemed to be related to their educational attainment, employment status, family income level, and how much time caregivers spent with their autistic child. This study supported the use of the Family Outcomes Survey, Form A, Chinese version with families of autistic children in China. We also discussed how the coronavirus disease 2019 pandemic could have impacted the family outcomes as reported by the Chinese caregivers.

Keywords

confirmatory factor analysis cross-cultural validation early childhood family outcome measures

Introduction

In recent years, nongovernment organizations and private agencies have been the main early intervention (EI) providers in China (Xu et al., 2018). While the child is usually the focus of EI services, it is important to realize that raising an autistic child may also have impacts on the functions of a family. Parents of autistic children may be facing physical, social, and emotional challenges more than parents of typically developing children. For example, parents of autistic children reported greater risks for psychological burden and various mental health issues, such as anxiety, depression, marital discord, and sleep problems (Hartley et al., 2017; Olson et al., 2021). Meanwhile, the coronavirus disease 2019 (COVID-19) outbreak led to school and intervention center closures in China, which resulted in millions of autistic children and related disabilities staying at home (Wang et al., 2020). Many of them might experience behavioral regression and extensive problem behaviors, which can be challenging for their families to handle (Wang et al., 2020). This has increased the urgent need to switch from the traditional, center-based EI service delivery model to one that supports parents in using effective intervention strategies with their children at home (McDevitt, 2021). In this context, it is important to measure family outcomes, in addition to child outcomes, in order to fully understand the effectiveness of EI services. Bailey and colleagues (2006, p. 228) conceptualized family outcomes as “benefits experienced by families as a result of services received.” In the current study, family outcomes are defined as benefits a family of an autistic child receives as a result of EI services.

In developing countries like China, measuring family outcomes is still not a common practice due to a lack of awareness, public resources, and validated measurement tools (Acar et al., 2019). Similarly, EI and preschool programs in the United States usually are not held accountable for family outcomes (Bailey et al., 2011). Thus, studies on autistic children have not routinely measured parental and family outcomes (Wainer et al., 2017). But in recent years, there have been increasing efforts to measure the efficacy of programs serving young children with disabilities and their families. Several studies (Kuhaneck et al., 2015; Landa, 2018; Vinen et al., 2018) have focused on evaluating the effectiveness of EI services for families of children with autism and related disabilities.

There is an emerging consensus (Waschl et al., 2021; Wicks et al., 2021) on the need to measure parental and family functioning to better understand the impacts of EI services in a wider context. A family-centered approach to EI services requires addressing family outcomes in various dimensions, such as focusing on family strengths, respecting family diversity and values, encouraging family decision-making and empowerment, communicating with families in an open and collaborative manner, adopting a flexible approach to service provision, and recognizing the value of informal support systems (Bailey et al., 2011). To advance research and clinical practice, it is necessary to use a validated measure when evaluating family outcomes. However, few psychometrically valid scales exist to assess family outcomes for young autistic children in China.

In the United States, one of the widely used measures of family outcomes in the EI sector is the Family Outcomes Survey–Revised (FOS-R; Bailey et al., 2011). The Simplified Chinese version of the FOS-R in this study is offered by the Early Childhood Outcomes Center with support from the Office of Special Education Programs, US Department of Education (available online https://ectacenter.org/eco/pages/familysurveys.asp). Previous cross-cultural studies of the FOS-R have reported good psychometric properties in Asian countries like Singapore (Poon et al., 2014) and Japan (Ueda et al., 2015). For example, Poon et al. (2014) evaluated the Bilingual version (i.e. English-Chinese and English-Malay) of the FOS-R with a total sample of 291 parents. Results suggested that the five individual subscales as well as the overall score showed high internal consistency (e.g. Cronbach’s α ranged from .84 to .91). Moreover, confirmatory factor analyses indicated that there was a fit between the current data set and the FOS-R structure proposed by the developers (Bailey et al., 2011). Overall, the participants reported moderately high attainment of family outcomes.

Research on predictors of family outcomes as measured on the FOS-R is only emerging. Adams and colleagues (2019) conducted analysis of variance (ANOVA) and reported significantly higher FOS-R-A Outcome 5 (accessing the community) score in families with a higher income, but not on the other family outcomes. This article (Adams et al., 2019) also reported no significant differences on any of the family outcomes upon parental employment status, mental health diagnosis, or child’s gender. A more recent study (Wicks et al., 2019) applied a hierarchical multiple regression model to examine how child characteristics (A), family resources (B), parental coping strategies (C) predicted FOS-R family outcomes (X) based on an double ABCX model of family adaptation (McCubbin & Patterson, 1983). Findings of this study (Wicks et al., 2019) indicated that greater autistic characteristics of the child and higher parent-child dysfunctional interaction significantly predicted lower scores on FOS-R-A Outcome 1 (understanding your child’s strengths, needs, and abilities); higher parent–child dysfunctional interaction and lower parental education predicted lower scores on Outcome 2 (knowing your rights and advocating for your child); higher parent-child dysfunctional interaction predicted lower scores on Outcome 3 (helping your child develop and learn); higher parental distress and lower parental education predicted lower scores on Outcome 4 (having support systems), and; higher parental distress, lower communication skills of the child, lower parental education, and lower family income predicted lower scores on Outcome 5 (accessing the community).

At present, there is no research reporting the psychometric properties or predictors of the FOS-R when used in China. Given the strong validity evidence, it would be important to further evaluate the construct validity with a Chinese sample and this may contribute evidence for the validity of a good measure to support EI program evaluation in mainland China. As summarized above, previous studies on the FOS-R were conducted in the United States (Bailey et al., 2011), Singapore (Poon et al., 2014), and Japan (Ueda et al., 2015).

Findings in the FOS-R studies conducted in other countries may or may not apply to Chinese families due to the differences in the available resources, beliefs, expectations, and parenting practices in different cultural contexts. For example, Chinese families having an autistic child were reported (Wang et al., 2011) to have more pessimistic expectations of their child due to the unique societal and cultural factors in China such as shortage of professionals and financial burden (Wang et al., 2011). A review study (Ng et al., 2021) identified five categories of burdens that Chinese caregivers of children with autism spectrum disorder (ASD) experience: physical burden, psychological burden, financial burden, social burden, and time burden. In addition to examining the reliability and validity evidence of the FOS-R on a Chinese sample, this study also aimed to investigate how children’s and families’ characteristics are associating with caregivers’ ratings on the FOS-R. The study addressed the following research questions:

Research Question 1 (RQ1): To what extent is the factor structure of the FOS-R-A supported by empirical data collected from parents of young autistic children in China?

Research Question 2 (RQ2): Do FOS-R-A scores show good internal consistency?

Research Question 3 (RQ3): To what extent are selected child and family characteristics associated with the family outcomes of autistic children in EI/early childhood special education (EI/ECSE)?

Method

Participants

Caregivers recruited from 12 intervention centers serving only autistic children in six cities in China (Beijing, Zhengzhou, Shanghai, Ningbo, Guangzhou, and Shenzhen) were included in this study. These intervention centers provided half- or full-day interventions in self-contained classrooms for autistic children and online training for their caregivers. Applied behavioral analysis was the primary approach to the interventions at these centers. Early interventionists, behavioral analysts, and therapists worked together with families to develop and implement individualized intervention programs. Most families paid the intervention fees in their entirety themselves (e.g. ¥3000 CNY/US$450 a month), while some received government subsidies, depending on the policies of each city.

Admission to the centers was based on either a medical diagnosis of autism or parental concerns thereof (e.g. in the case of families that wanted to start interventions while the children were still on the waiting list for a diagnostic evaluation). Children who were admitted to the center based on an increased likelihood of autism (e.g. on the waiting list for diagnosis) would be withdrawn from these centers once the diagnostic results rejected autism. Although the current study did not follow up on the participating children who were admitted without a diagnosis, the center staff estimated based on past experiences that a majority of children enrolled based on suspected condition would receive a formal diagnosis of autism within 1 year and continue to receive intervention from the center.

The diagnosis of autism in China is very different from practices in Western countries. Regarding the diagnostic instruments, Chinese practitioners rarely use the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2; Lord et al., 2000) and/or the Autism Diagnostic Interview-Revised (ADI-R; Rutter et al., 2003), due to the limited access and high cost (Pang et al., 2018). Another study (Huang et al., 2013) reported that the processes of diagnosing autism in China usually include interviewing the parents about the child’s developmental history and concerns, structured observation and multi-component assessments in the child’s development, and the use of an autism-specific screening instrument such as the Chinese version of the Child Autism Rating Scale (CARS; Scholper et al., 1988) or the Autism Behavior Checklist (ABC; Krug et al., 1980). Huang and colleagues (2013) further pointed out that autistic children without a co-occurring intellectual disability or language delay are less likely to receive a formal diagnosis in China due to lack of awareness of autism among the Chinese professionals (Huang et al., 2013). The whole autism diagnostic process for each child usually takes only 15–30 min and typically does not include any observation of the child in multiple settings and no multi-disciplinary diagnostic team is involved (Huang et al., 2013). In addition to the differences in instruments, procedures, diagnostic criteria, and personnel, perhaps it is more important to note that autism diagnostic resources are very scarce in China (Huang et al., 2013). It may be due to these reasons that the estimated number of autistic children has always been much lower in China, ranging from 1 in 255 (Wang et al., 2018) to 1 in 1205 (Jin et al., 2018), than the global estimation of 1 in 160 as reported by World Health Organization and the US estimation of 1 in 44 (Maenner et al., 2021).

The characteristics of the children and their families are listed in Table 1. Most (89.7%) of the participants were mothers. The caregivers’ ages ranged from 26 to 59 years (M = 34.98, SD = 4.54). A comparison of the educational profile of the sample with national census data (National Bureau of Statistics of China, 2021) indicated an overrepresentation of respondents with an associate’s degree or higher. The monthly income per capita in this sample was also significantly higher than the urban family income per capita in the national census (p < .001). Most (66.2%) participants reported that their children had been diagnosed with autism, while 33.8% indicated that their children were suspected of autism and were on the waiting list for diagnosis. Among the 309 child participants who were reported to have been diagnosed with autism, only 8 (2.6%) received their diagnosis from a private clinic, 1 (0.3%) received the diagnosis from a public special school, 238 (77%) received the diagnosis from a publicly funded hospital or healthcare clinic, and 62 (20.1%) did not respond to the question about where the child received the autism diagnosis. The children’s ages ranged from 4 to 97 months (M = 47.62, SD = 17.50). Most (68.5%) were boys. Although all children were receiving interventions at the 12 centers, few (18.4%) were also attending preschool for children with and without disabilities. Most participating families had three to five household members. The monthly family incomes ranged from 1000 to 20,000 CNY (M = 10,019, SD = 12,670). The caregivers reported spending 0–24 h (M = 3.91, SD = 2.25) daily with their children. Only one caregiver reported spending 0 h daily with their autistic child and this participant is included in the current sample for attending the parent training program, which demonstrates indirect involvement in the intervention for the child.

Table 1.

Demographic characteristics of participants (N = 467).

Demographic variables	N (%) or Mean (SD)
Child’s gender: male	320 (68.5%)
Diagnosed with autism	309 (66.2%)
Child is attending a preschool	86 (18.4%)
Caregiver’s relationship with autistic child: Mother	419 (89.7%)
Father	38 (8.1%)
Grandparent	1 (0.2%)
Other (e.g. aunt, domestic helper)	9 (1.9%)
Caregiver’s highest educational attainment:
Compulsory education (ninth grade) or below	18 (3.9%)
Secondary level (e.g. high school, vocational training)	41 (8.8%)
Associate degree	84 (18.0%)
Undergraduate degree	236 (50.5%)
Post-graduate degree	88 (18.8%)
Caregiver’s current employment status: Not employed	183 (39.2%)
Part-time employment	23 (4.9%)
Full-time employment	261 (55.9%)
Number of family members in the household: 2	21 (4.5%)
3	120 (25.7%)
4	139 (29.8%)
5	121 (25.9%)
6 or more	66 (14.1%)
Child’s age in months	47.62 (17.50)
Caregiver’s age	34.98 (4.54)
Family’s monthly income per capita in CNY	10,019 (12,670)
On a typical day, how many hours do you spend with your child?	3.91 (2.25)

Measures

The Mandarin Chinese version of the FOS-R-A (Bailey et al., 2011) was the measure of interest in this study. Little information is available about how the translation was done and whether there is any evidence supporting its linguistic equivalence to the original English version. In this study, two researchers, both native Chinese speakers who completed their doctoral training at US universities, reviewed the Chinese and English versions item by item and identified no significant adaptations or discrepancies in meaning. The Chinese FOS-R-A has the same 24 items as the English version, with four to six items per subscale addressing five family outcomes: understanding your child’s strengths, needs, and abilities (Outcome 1), knowing your rights and advocating for your child (Outcome 2), helping your child develop and learn (Outcome 3), having support systems (Outcome 4), and accessing the community (Outcome 5). Each of the 24 items was rated on a 5-point Likert-type scale from 1 (not at all) to 5 (completely). The FOS-A full-scale score and each of the five subscale scores were the mean of all items, which was calculated by dividing the summed score of all items by the number of items on the scale/subscale.

In addition to the Chinese FOS-R-A, demographic data on the children, the caregivers completing the questionnaires, and the families were collected. Information about the children included date of birth, gender, whether and where they had received an autism diagnosis, and whether they were attending a preschool that was open to children with and without autism. Information on the caregivers included relationship with the child, highest educational attainment, employment status, and average daily hours interacting with the autistic child. Information on the families included household size, monthly income, number of employed family members, and siblings of the autistic child.

Procedures

This study collected de-identified FOS-R-A and demographic data from the intervention centers’ archived client surveys and child profiles documented from March to June 2020, during the COVID-19 pandemic. Exempt human subjects’ research status was granted by the University of Massachusetts before the data were extracted. The 12 intervention centers were approached by the researchers via a personal network. Once a center accepted the invitation, the center staff extracted and de-identified the data needed and sent them to the researchers.

As part of the routine evaluation, all parents enrolled in the online parent training program in the 12 intervention centers were invited to participate in a parent survey using a third-party online survey platform in China, like Survey Monkey. This routine parent survey included the measures in this study and other questions about their experiences with and perceptions of the intervention services. According to reports from the center staff, parents generally took 10–30 min to complete the whole online survey. Among the 781 families approached, 485 agreed to participate, resulting in a participation rate of 62.1%. Of those who participated, 18 (3.71%) were excluded due to submitting incomplete questionnaires (i.e. missing values in the FOS-R-A). Thus, a total of 467 questionnaires were used in the analyses. Only one questionnaire was collected for each child.

Family members, community providers, and agency leaders from the 12 intervention centers serving autistic children where recruitment took place were actively involved in reviewing the FOS-R-A for cultural sensitivity for caregivers in China as well as in the data collection process. The findings have been shared with these stakeholders to receive feedback.

Analysis

Confirmatory factor analysis

To examine the factor structure of the FOS-R A (Chinese version), confirmatory factor analysis (CFA) was conducted using Mplus 8.6 (Muthén & Muthén, 2018). A robust maximum likelihood estimator, which is appropriate for non-normal data distributions, was used. Factor structures reported in previous studies (Bailey et al., 2011; Poon et al., 2014; Waschl et al., 2021) were tested: (1) five factors (understanding your child’s strengths, needs, and abilities; knowing your rights and advocating for your child; helping your child develop and learn; having support systems; and accessing the community); (2) five factors with a higher order general family outcome factor; (3) five factors with a mini testlet factor to account for the shared variance of items on the fifth factor measuring practical aspects of care; and (4) five factors with a mini testlet factor and a higher order general family outcome factor.

Model fit was assessed according to standard model fit criteria: a comparative fit index or Tucker–Lewis index of ⩾.95, a root mean square error of approximation of ⩽.06, and a standardized root mean square residual of ⩽.08 (Brown, 2006; Hu & Bentler, 1999). In cases in which fit cutoffs were not met, alternative models were explored to examine whether the model fit could be improved.

Internal consistency reliability

Cronbach’s α was calculated for each of the five FOS-R-A (Chinese version) factors in R 4.0.5 (R Core Team, 2021) using the psych package (Revelle, 2020).

Descriptive statistics and multiple regression

IBM SPSS Statistics 25 was used for descriptive statistics, ANOVA, and multiple regression analyses. In the linear regression analyses, the overall FOS-R-A score and five subscale scores served as the outcome variables. The mean score on each subscale was computed as the subscale score. Then, the average of the five subscale scores was calculated as the overall FOS-R-A score. Within-subjects ANOVA was conducted to identify statistically significant differences among the five-outcome (subscale) scores. Effect sizes for group differences were defined as small (d = 0.10), medium (d = 0.30), and large (d = 0.50). These outcome variables were regressed on eight predicting variables, including four continuous variables (child’s age in months, number of hours that the caregiver typically spent with the child, monthly family income, and number of household members) and five dichotomous variables (whether the child was a boy, whether the child had been diagnosed with autism, whether the child was attending preschool, and whether the caregiver had at least an associate’s degree). Effect size of f ² = 0.02 was defined as small, f ² = 0.15 was defined as medium, and f ² = 0.35 was defined as large effect size.

Results

Descriptive statistics

Item level

Table 2 lists the means and standard deviations, skewness, and percentages of the responses to the 24 FOS-R-A items. All but three items (items 18, 19, and 22) were negatively skewed. Item 24 (“Our food, clothing, and housing needs are met”) showed the highest mean score (3.78), followed by items 3 (“We understand our child’s delays and/or needs”) and 4 (“We are able to tell when our child is making progress”). Item 24 also showed the highest percentage (27.6%) of completely responses. Item 19 (“Our child participates in social, recreational, or religious activities that we want”) had the lowest mean score (2.64), followed by items 22 (“Our childcare needs are met”) and 8 (“We know what options are available when our child leaves the program”). Item 22 also showed the highest percentage (21.4%) of not at all responses.

Table 2.

Item responses to the FOS-R-A.

FOS-R-A scores	% of not at all	% of a little	% of somewhat	% of almost	% of completely	M (SD)	Skewness (standard error)
FOS-R-A full scale						3.26 (0.73)	−0.08 (0.11)
Outcome 1: Understanding your child’s strengths, needs, and abilities						3.60 (0.86)	−0.62 (0.11)
1. We know the next steps for our child’s growth and learning	4.90%	13.30%	35.88%	33.80%	12.20%	3.35 (1.02)	−0.33 (0.11)
2. We understand our child’s strengths and abilities	4.7	9.0	27.6	42.6	16.1	3.56 (1.02)	−0.65 (0.11)
3. We understand our child’s delays and/or needs	3.0	7.3	23.6	43.9	22.3	3.75 (0.98)	−0.72 (0.11)
4. We are able to tell when our child is making progress	3.0	8.8	22.7	42.6	22.9	3.74 (1.01)	−0.69 (0.11)
Outcome 2: Knowing your rights and advocating for your child						3.09 (0.95)	−0.11 (0.11)
5. We are able to find and use the services and programs available to us	6.2	17.1	33.2	33.4	10.1	3.24 (1.05)	−0.29 (0.11)
6. We know our rights related to our child’s special needs	7.3	17.8	33.8	29.8	11.3	3.20 (1.09)	−0.22 (0.11)
7. We know who to contact and what to do when we have questions or concerns	12.0	21.8	29.3	27.8	9	3.00 (.1.16)	−0.12 (0.11)
8. We know what options are available when our child leaves the program	15.0	20.1	36.8	20.6	7.5	2.85 (1.13)	−0.02 (0.11)
9. We are comfortable asking for services and supports that our child and family need	12.0	16.7	30	27.4	13.9	3.15 (1.21)	−0.22 (0.11)
Outcome 3: Helping your child develop and learn						3.41 (0.82)	−0.19 (.011)
10. We are able to help our child get along with others	4.3	20.6	38.8	27.6	8.8	3.16 (0.99)	−0.05 (0.11)
11. We are able to help our child learn new skills	1.3	12.2	38.1	37.5	10.9	3.45 (0.89)	−0.17 (0.11)
12. We are able to help our child take care of his/her needs	1.5	9.9	29.6	43.0	16.1	3.62 (0.92)	−0.42 (0.11)
13. We are able to work on our child’s goals during everyday routines	1.7	12.4	39	34.9	12	3.43 (0.92)	−0.15 (0.11)
Outcome 4: Having support systems						3.10 (0.91)	−0.06 (0.11)
14. We are comfortable talking to family and friends about our child’s needs	10.7	22.9	34.7	23.3	8.4	2.96 (1.11)	−0.02 (0.11)
15. We have friends or family members who listen and care	7.5	22.1	32.1	26.6	11.8	3.13 (1.11)	−0.07 (0.11)
16. We are able to talk with other families who have a child with similar needs	7.3	17.1	27.8	30.0	17.8	3.34 (1.17)	−0.29 (0.11)
17. We have friends or family members we can rely on when we need help	6.9	21.2	30.2	27.4	14.3	3.21 (1.14)	−0.12 (0.11)
18. I am able to take care of my own needs and do things I enjoy	12.4	24.0	34.5	21.8	7.3	2.88 (1.11)	0.02 (0.11)
Outcome 5: Accessing the community						3.11 (0.88)	−0.06 (0.11)
19. Our child participates in social, recreational, or religious activities that we want	19.9	25.7	30.6	18.2	5.6	2.64 (1.15)	0.17 (0.11)
20. We are able to do things we enjoy together as a family	6.9	22.7	32.8	29.6	8.1	3.09 (1.06)	−0.12 (0.11)
21. Our medical and dental needs are met	14.8	19.1	31.0	26.1	9	2.96 (1.18)	−0.12 (0.11)
22. Our childcare needs are met	21.4	22.7	28.9	18.6	8.4	2.70 (1.23)	0.17 (0.11)
23. Our transportation needs are met	7.1	12.8	24.4	32.5	23.1	3.52 (1.18)	−0.50 (0.11)
24. Our food, clothing, and housing needs are met	3.4	9.0	21.6	38.3	27.6	3.78 (1.06)	−0.71 (0.11)

Note: FOS-R-A = Family Outcomes Survey, Revised, Form A. M = mean. SD = standard deviation.

Subscale and overall scale levels

Of the five family outcomes, Outcome 1 (understanding your child’s strengths, needs, and abilities) had the highest mean score (M = 3.60, SD = 0.86), followed by Outcome 3 (helping your child develop and learn; M = 3.41, SD = 0.82). Both were significantly higher than the other three mean outcome scores (p < .001) and significantly different from each other (p < .001, effect size (d) = 0.144). No significant differences were found between Outcomes 2 (knowing your rights and advocating for your child; M = 3.09, SD = 0.95), 4 (having support systems; M = 3.10, SD = 0.91), and 5 (accessing the community; M = 3.11, SD = 0.88). All five mean subscale scores were negatively skewed. The overall mean FOS-R-A score was 3.26, falling between 3 (somewhat) and 4 (almost), with a slightly negatively skewed distribution.

Confirmatory Factor Analysis

Of the initially hypothesized models, the five-factor model with a mini factor accounting for shared variance between items related to practical care on the fifth factor demonstrated the best fit. However, upon inspection of the item loadings on the mini factor, only two items (23 and 24) showed loadings of >.20, indicating limited meaningfulness of this factor. Thus, alternative but similar models that accounted for the shared variance between items 23 and 24 and items 21 and 22 were explored. A model that accounted for this shared variance by including correlations between the disturbances of these items, as well as accounting for the shared variance between items 14 and 15 (M5), showed a better fit than the five-factor plus mini-factor model (M3) and the five-factor plus mini factor model with residual correlations between items 14 and 15 (M7). This provided support for the five factors of the FOS-R-A (Chinese version) without an additional mini factor/testlet factor, but with some additional shared variance modeled between certain pairs of items (Figure 1).

Figure 1.

Confirmatory factor analysis model tested for five factors modified to account for shared variance between items 14 and 15, 23 and 24, and 21 and 22 (M5). o1 = understanding child’s strengths, abilities, and special needs; o2 = knowing their rights and advocating for their children; o3 = helping their child develop and learn; o4 = having support systems; o5 = accessing desired services, programs and activities in the community.

Models that included a higher order overall family outcome factor (M4 and M6) showed a poorer fit than their respective lower order five-factor models without an overall family outcome factor, indicating that the FOS-R-A (Chinese version) is best conceptualized in terms of five distinct but related dimensions rather than an overall family outcome (Table 3).

Table 3.

Confirmatory factor analysis fit indices.

Model	χ²	df	p	χ²/df	CFI	TLI	RMSEA	SRMR
M1: five factor	791.27	242	<.001	3.27	.90	.89	.07	.06
M2: five factor, higher order	859.33	247	<.001	3.48	.89	.88	.07	.07
M3: five factor + mini factor	566.71	239	<.001	2.37	.94	.93	.05	.05
M4: five factor + mini factor, higher order	651.25	242	<.001	2.69	.93	.92	.06	.06
M5: five factor modified^a	541.71	239	<.001	2.27	.95	.94	.05	.05
M6: five factor modified^a, higher order	627.76	244	<.001	2.57	.93	.92	.06	.06
M7: five factor + mini factor modified^b	539.40	238	<.001	2.27	.95	.94	.05	.05

Note. ^aResidual correlations modeled between items 23 and 24, 21 and 22, 14 and 15; ^bResidual correlations modeled between items 14 and 15.

CFI: Comparative fit index; TLI: Tucker–Lewis index; RMSEA: root mean square error of approximation; SRMR: standardized root mean square residual.

Internal consistency

Cronbach’s α ranged from good to excellent for all five FOS-R-A (Chinese version) factors: Outcome 1, α = .88; Outcome 2, α = .90; Outcome 3, α = .91; Outcome 4, α = .87; Outcome 5, α = .86.

Multiple regression

First, all predicting variables were entered into the model and only those that statistically significantly predicted the outcome variables were retained, resulting in the six models listed in Table 4. In Model 1, caregiver’s time spent with the child every day, unemployment, and family income were statistically significant in explaining the full-scale FOS-R-A score, F (3, 405) = 10.66, p < .001, with 7.9% of variance in FOS-R-A score explained. In Model 2, caregivers’ time spent with the child every day and possessing a college degree were statistically significant in explaining the Outcome 1 score, F (2, 462) = 6.3, p < .01, with 3.9% of variance in Outcome 1 score explained. In Model 3, caregivers’ time spent with the child every day was statistically significant in explaining the Outcome 2 score, F (1, 463) = 10.83, p < .01, with 3.1% of variance in Outcome 2 score explained. In Model 4, caregivers’ time spent with the child every day, unemployment, and family income were statistically significant in explaining the Outcome 3 score, F (3, 405) = 6.28, p < .001, with 5.4% of variance in Outcome 3 score explained. In Model 5, caregivers’ time spent with the child every day and unemployment were statistically significant in explaining the Outcome 4 score, F (2, 462) = 11.75, p < .001, with 6% of variance in Outcome 4 score explained. In Model 6, caregivers’ time spent with the child every day, unemployment, and family income were statistically significant in explaining the Outcome 5 score, F (3, 405) = 12.77, p < .001, with 8.4% of variance in Outcome 5 score explained. The six regression models showed small to medium effect sizes f ² ranging from 0.32 (Outcome 2) to 0.092 (Outcome 5).

Table 4.

Summary of stepwise multiple regression results predicting mean scores on the FOS-R-A (n = 373).

Variables	ß	95% confidence interval		p value	R ²	f ²
Model 1 DV: Full-scale FOS-R-A mean					0.079	0.086
IV: Hours spent with child^a	0.062	0.028	0.096	0.000
IV: Unemployed^b	−0.307	−0.464	−0.151	0.000
IV: Income in CNY 1000^a	0.007	0.001	0.013	0.016
Model 2 DV: Outcome 1 M: Understanding your child’s strengths, needs, and abilities					0.039	0.041
IV: Hours spent with child^a	0.046	0.005	0.086	0.029
IV: College degree^b	0.295	0.099	0.490	0.003
Model 3 DV: Outcome 2 M: Knowing your rights and advocating for your child					0.031	0.032
IV: Hours spent with child^a	0.077	0.031	0.122	0.001
Model 4 DV: Outcome 3 M: Helping your child develop and learn					0.054	0.057
IV: Hours spent with child^a	0.058	0.019	0.096	0.004
IV: Unemployed^b	−0.261	−0.440	−0.083	0.004
IV: Income in CNY 1000^a	0.008	0.001	0.014	0.019
Model 5 DV: Outcome 4 M: Having support systems					0.060	0.064
IV: Hours spent with child^a	0.056	0.012	0.100	0.012
IV: Unemployed^b	−0.429	−0.629	−0.230	0.000
Model 6 DV: Outcome 5 M: Accessing the community					0.084	0.092
IV: Hours spent with child^a	0.069	0.029	0.109	0.001
IV: Unemployed^b	−0.380	−0.565	−0.195	0.000
IV: Income in CNY 1000^a	0.010	0.003	0.017	0.003

Note. ^aIndicates a continuous variable, ^b Indicates a dichotomous variable.

Each additional hour was associated with a 0.05 (Outcome 1) to 0.08 (Outcome 2) increase in FOS-R-A scores. In addition, caregivers’ unemployment was significantly associated with a 0.26 (Outcome 3) to 0.43 (Outcome 4) reduction in scores. For every increase of ¥1,000 CNY in monthly family income, a significant increase of 0.007 (overall scale) to 0.10 (Outcome 5) was found. A college degree was significantly associated with an increase of 0.30 (95% CI: 0.01–0.49, p < 0.01) in the mean Outcome 1 score.

Discussion

This study aimed to evaluate the construct validity of the Chinese version of the FOS-R-A and explore the factors associated with family outcomes using a large sample of children diagnosed with or at increased likelihood of autism in China. A CFA analysis was conducted to examine the fit of the FOS-R-A model consisting of five subscales: understanding your child’s strengths, needs, and abilities (Outcome 1), knowing your rights and advocating for your child (Outcome 2), helping your child develop and learn (Outcome 3), having support systems (Outcome 4), and accessing the community (Outcome 5). The regression analysis results indicate that family income and the caregiver’s time spent with the child, employment status, and education level are associated with family outcomes. These results support the increasing emphasis on family engagement and family outcome measures in the literature (e.g. Kuhaneck et al., 2015; Wainer et al., 2017) and have research and practical implications for the use of the FOS-R-A.

Evidence for reliability and construct validity of using the FOS-R-A with Chinese parents of autistic children

This study provided evidence for good internal consistency reliability (i.e. Cronbach’s α > .080) of the FOS-R-A (Chinese version) total score and subscale scores. In addition, construct validity of the FOS-R-A (Chinese version) as measure of five family outcomes for children diagnosed with or at increased likelihood of autism in China was supported; the five-factor modified model (M5), which showed the best fit with the Chinese data, supports the original five-outcome structure of the FOS-R-A as reported by Bailey et al. (2011) and Poon et al. (2014). This provides new evidence for the cross-cultural applicability of the FOS-R-A structure in China. The fact that Model 5 showed a better fit than the second-order models (M2, M4, and M6) is consistent with Waschl et al. (2021) and supports the use of the five family outcomes as independent subscales.

Moreover, Model 5 is slightly different from Model 3, which has been reported to show the best fit with a Singaporean data set (Waschl et al., 2021). This is indicative of the impact of the cultural context on family outcomes. In our study, a concerningly high proportion of caregivers reported that their medical and dental needs were not at all met. The proportion reporting that their childcare needs were not at all met was even higher. Both were higher than the proportions reporting that their transportation and food, clothing, and housing needs were not met. In the Singaporean data, the not at all response rates in these four items were lower and more consistent. Research has revealed significant shortages in medical, dental, and childcare resources in China (Jiang et al., 2017; Ma et al., 2012; Sullivan & Wang, 2020; Xu & Zhang, 2014), which may have contributed to the high rate of unmet needs in this study. However, given that families of a high socioeconomic status (SES), who are expected to be better resourced, were overrepresented in this sample, further research is needed to investigate whether this high rate of unmet medical, dental, and especially childcare needs was due to discrimination against children diagnosed with or at increased likelihood of autism rather than a general scarcity of resources. Another direction for future research is to collect data from a low socioeconomic sample in China to identify disparities in resource allocation and differentiated needs.

Another feature of this five-factor modified model was the correlated errors of items 14 and 15, which is different from what has been reported in previous studies of the FOS-R-A (e.g. Waschl et al., 2021). From the wording used in the items, one may infer why these items might be characterized by greater shared variance than that explained by Outcome 4 (having support systems). For participants who felt comfortable talking about their children’s needs with friends or family (item 14), their friends or family were also likely to listen and care (item 15).

Potential cultural impacts on family outcomes

A comparison of the five FOS-R-A subscale scores in the China, Singapore (Poon et al., 2014), Japan (Ueda et al., 2015) and US (Bailey et al., 2011) data sets is listed in Figure 2. As reported in previous cross-cultural studies (e.g. Poon et al., 2014; Ueda et al., 2015), Chinese caregivers in general seem to give lower ratings than US caregivers. Another study (Adams et al., 2019) reported a mean subscale score range of 3.56–4.06 in an Australian sample but is not included in Figure 2 because it did not report the specific scores. It has been suggested that East Asian cultures favor moderate responses (e.g. Hamamura et al., 2008). However, the lower scores may also indicate a generally low attainment of family outcomes in China (as in Singapore and Japan) compared to Western countries, such as the U.S. and Australia. Acar et al. (2021) reviewed the literature on parent involvement in developmental disabilities and noted that some aspects of the collectivist culture of Asian countries, such as paternalistic tendencies and shame avoidance, may be barriers to families’ active involvement in their children’s interventions and the achievement of positive family outcomes.

Figure 2.

Mean scores of the five subscales on the Family Outcomes Survey, Revised, Form A, across four countries.

In this study, Outcome 3 (helping your child develop and learn) had the second highest mean score (only lower than that of Outcome 1), which was significantly higher than those of Outcomes 2 (knowing your rights and advocating for your child), 4 (having support systems), and 5 (accessing the community). This seems to be a unique feature in the Chinese data set. One possible explanation is the parent training component in the intervention centers from which the sample was drawn, which may have enhanced caregivers’ perceived confidence and skills in helping their children develop and learn (Outcome 3). The COVID-19 pandemic context for data collection might be another reason for the higher ratings on Outcome 3 (helping your child develop and learn). Due to school shutdown and work-from-home measures, parents were likely spending more time at home with their children than they did before the pandemic, which as reported echoed findings in previous studies (Amorim et al., 2020; Mumbardó-Adam et al., 2021). Work-from-home and school shutdown might have provided an opportunity for many participants in this study to apply what they have learned from the parent training in supporting their child at home. However, evaluating the effect of parent training is beyond the scope of the current study. Further research using longitudinal family outcome data collected before and after parent training may help to elucidate the effect of including a parent training component in autism-specific interventions on family outcomes.

The restrictions to social, recreational, and other public activities during the COVID-19 pandemic might also help in explaining the lower ratings on Outcome 5 (accessing the community). Barriers for autistic children and their families to participate in community activities has been reported (e.g. Tint et al., 2017) and the low ratings on Outcome 5 (accessing the community) in this study aligned with the literature. However, the social distancing measures and lockdown policies during the COVID-19 pandemic could have made access and participation even more challenging for the children and families.

Child and family predictors of family outcomes

First of all, the child factors examined in the current study such as children’s age, gender, and confirmed diagnosis of autism did not significantly predict family outcomes. More research is needed to investigate the relationships between family outcomes and other child characteristics. For example, Wicks et al. (2019) reported the child’s autism characteristics and communication skills, parental distress, and parent–child dysfunctional interaction as significant predictors of FOS-R-A. However, these variables were not included in the current study. The other significant predictors reported in Wicks et al. (2019), including parental education and family income, were included in the current study. In future research, it is reasonable to hypothesize that these predictors of parenting stress could also impact parents’ ratings on family outcomes.

It is noteworthy that caregivers who spent more time with their children gave higher scores on the FOS-R-A. This finding, together with those in a previous study reporting a significant positive correlation between parent involvement in school and parental school satisfaction (Zablotsky et al., 2012), highlights the importance of family involvement that has been emphasized in the literature (e.g. Involving parents in the intervention for their child was reported to associate with many positive outcomes in the child’s academic achievement (Fan & Chen, 2001), reduced behavioral needs (Domina, 2005), and social-emotional development (Sheridan et al., 2013). The finding of this study contributed to the literature by adding the evidence that involving caregivers in the interactions with and care for their autistic child was associated with improved family outcomes as measured by the FOS-R-A. Based on this finding, it is promising to apply a family-centered approach to EI services in China (e.g. Hu & Yang, 2013; Hu, 2012).

The finding that family income was one of the predictors of Outcome 5 (accessing the community) is consistent with findings on two cohorts (ages 4–5 and 9–10 years) of autistic children in Australia (Wicks et al., 2019). However, in our study, family income was also associated with Outcome 3 (helping your child develop and learn), which was not significant in Wicks et al.’s study. This is probably because our sample was drawn from nongovernment intervention centers, and most enrolled families were paying the intervention fees themselves. Given the shortage of publicly funded autism intervention services in China (Hu & Yang, 2013), family income may play a more critical role in family outcomes. In the Australian study (Wicks et al., 2019), parents’ education was a significant predictor of Outcome 2 (knowing your rights and advocating for your child). In our study, parental education was not a significant predictor of Outcome 2 but was significantly predicting Outcome 1 (understanding your child’s strengths, needs, and abilities). A possible explanation may be related to differences in political systems. As McCabe (2007) observed, advocating for autistic children is more difficult for parents in China than in Western countries because of China’s top-down approach to policymaking. A higher education level may not necessarily be associated with knowing one’s rights or advocacy in such a system but could be helpful in accessing information about child development and autism and better understanding their child. Although the regression models used in this study identified significant predictors of FOS-R-A scores, the explained variance remained below 10%. Future research is needed to examine the relationships between FOS-R-A (Chinese version) and other child and family characteristics, such as children’s levels of autism characteristics and behavioral functioning, parental stress, parenting efficacy, and family functioning.

Limitations

It is worth noticing that the data in this study were collected during the COVID-19 pandemic when many cities in China exercised school shutdown and online schooling (e.g. Filfield, 2020), as well as work-from-home measures (Liang, 2020). Findings from this study, therefore, should be interpreted with considerations of the possible impacts from the pandemic and the subsequent social distancing measures. One such impact may be found in the low percentage (18.4%) of child participants attending a preschool, either physically going to the school site or attending online. It was reported that about 44% of children aged three to six with disabilities were enrolled in preschools (China Ministry of Public Health et al., 2003), which is much higher than 18.4% in the current sample. It is possible that young autistic children experienced more barriers in school enrollment than children with other types of disabilities due to the lack of policy or legislation support and professional resources for this population (Huang et al., 2013). However, preschool education in China is funded primarily via parent fees, unlike public school education provided for children aged 7–16 years (Wu et al., 2012). The massive number of preschools that went out of business during the COVID-19 pandemic (e.g. Zuo, 2021) could also have contributed to the low preschool attendance rate of 18.4% in this sample.

The fact that about one-third (33.8%) of the child participants in this study had not yet received a formal diagnosis of autism can be a major cause for concern. As a study to validate the Chinese FOS-R-A for use with families of young autistic children in China, including children who are on the waitlist for diagnosis in the study sample may result in compromised representativeness of the study sample and limit the generalization of the findings. However, the authors decided to include these children in the study based on the limited autism diagnostic resources and the urgent need for intervention services and supports of the children and families who for some reason have not yet received an autism diagnosis. As there is no law or legislation in China to mandate public intervention services for autistic children (Huang et al., 2013) and the public awareness of autism in China is low (Yu et al., 2020), it is reasonable to expect at least in the near future that a substantial proportion of children receiving EI in China have not received a formal autism diagnosis, just as the sample in this study.

Another concern of potential sample bias is raised by the relatively higher SES backgrounds of the participants in this study in their educational attainment and family income. Generalizing the findings to Chinese families with lower SES backgrounds may require extra caution. The high SES sample in this study might be a result of the recruitment method of extracting retrospective data from 12 autism intervention centers funded primarily on the fees paid by the families. Diagnosis and intervention services for autistic children and their families are very limited in China, families lacking the resources may experience barriers in accessing such services (Zhou et al., 2022). Future research is needed to investigate expectations and perceptions of families of autistic children from the rural areas and lower SES backgrounds.

Conclusion

Using validated measures to evaluate family outcomes is vital to the quality and sustainability of interventions for young autistic children and their families. In China, a developing country with a large population and limited autism intervention resources, caregivers and families play a key role in advocating for and even providing (e.g. via parent training and parent-mediated interventions) the needed interventions and other resources. Based on the responses of 467 caregivers, the five-factor structure of the FOS-R-A appears to be a good fit for this population, with minor modifications for a better adjustment to the cultural context. A higher family income and caregivers’ active involvement, college education, and employment are significant predictors of better family outcomes. However, this evidence on predictors of family outcomes may not be generalized to populations from lower SES backgrounds or living in resource-lacking areas of China. The study shed light on areas highlighted as needing improvement in terms of empowering caregivers to know their rights and advocate for their child, having support systems, and accessing the community. EI services for autistic children and their families in China may apply the family-centered approach (Dunst et al., 2007) in better addressing these gaps in needs. Using the validated Chinese FOS-R-A to monitor and evaluate the needs for support of the family can be helpful to the improvement of EI services for this population.

Footnotes

Acknowledgements

The authors thank the Dami and Xiaomi Child Development Center for their support in recruiting the participants in this research. They also thank the caregivers who participated in the study.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The first and the third authors of the manuscript are serving as consultants for some of the centers from which data were collected. Duties include providing advice for program development and evaluation.

Data availability statement

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Huichao Xie

References

Acar

Chen

C. I.

Xie

(2021). Parental involvement in developmental disabilities across three cultures: A systematic review. Research in Developmental Disabilities, 110, 103861. https://doi.org/10.1016/j.ridd.2021.103861

Acar

Zeng

Chen

C. I.

Xie

(2019). A review of EI/ECSE systems across four settings: Procedures and lessons learned. Asia-Pacific Journal of Research in Early Childhood Education, 13(3), 37–63. http://doi.org/10.17206/apjrece.2019.13.3.37

Adams

Keen

Heussler

H. S.

Wicks

Roberts

(2019). Family outcomes for families of 4–5-year-old children on the autism spectrum who have received early childhood intervention in Australia. Infants & Young Children, 32(3), 186–200. https://doi.org/10.1097/IYC.0000000000000143

Amorim

Catarino

Miragaia

Ferreras

Viana

Guardiano

(2020). The impact of COVID-19 on children with autism spectrum disorder. Revista de Neurologia, 71(8), 285–291. https://doi.org/10.33588/rn.7108.2020381

Bailey

D. B.

Bruder

M. B.

Hebbeler

Carta

Defosset

Greenwood

, et al. (2006). Recommended outcomes for families of young children with disabilities. Journal of Early Intervention, 28, 227–251.

Bailey

D. B.

Raspa

Olmsted

M. G.

Novak

S. P.

Sam

A. M.

Humphreys

B. P.

Nelson

Robinson

Guillen

(2011). Development and psychometric validation of the Family Outcomes Survey-Revised. Journal of Early Intervention, 33(1), 6–23. https://doi.org/10.1177/1053815111399441

Brown

T. A.

(2006). Confirmatory factor analysis for applied research. New York, NY: The Guilford Press.

China Ministry of Public Health, China Ministry of Public Security, China Disabled Persons’ Federation, China National Bureau of Statistics, and United Nations Children’s Fund. (2003). The sampling survey on disability in 0-6 years-old children in China in 2001. China Statistics Press.

Domina

(2005). Leveling the home advantage: Assessing the effectiveness of parental involvement in elementary school. Sociology of Education, 78, 233–249. http://doi.org/10.1177/003804070507800303

10.

Dunst

C. J.

Trivette

C. M.

Hamby

D. W.

(2007). Meta-analysis of family-centered help giving practices research. Mental Retardation and Developmental Disabilities Research Reviews, 13, 370–378.

11.

Fan

Chen

(2001). Parental involvement and students’ academic achievement: A meta analysis. Educational Psychology Review, 13, 1–22. http://doi.org/10.1023/A:1009048817385

12.

Filfield

(2020, February 17). In China, 200 million kids have gone back to school. The Washington Post. https://www.washingtonpost.com/world/asia_pacific/in-china-200-million-kids-have-gone-back-to-school-online/2020/02/17/e5cc6f10-5131-11ea-80ce-37a8d4266c09_story.html

13.

Hamamura

Heine

S. J.

Paulhus

D. L.

(2008). Cultural differences in response styles: The role of dialectical thinking. Personality and Individual Differences, 44(4), 932–942. https://doi.org/10.1016/j.paid.2007.10.034

14.

Hartley

S. L.

Papp

L. M.

Mihaila

Bussanich

P. M.

Goetz

Hickey

E. J.

(2017). Couple conflict in parents of children with versus without autism: Self-reported and observed findings. Journal of Child and Family Studies, 26(8), 2152–2165. https://doi.org/10.1007/s10826-017-0737-1

15.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55.

16.

Yang

(2013). Early intervention practices in China: Present situation and future directions. Infants & Young Children, 26(1), 4–16. https://doi.org/10.1097/IYC0b013e3182736cd3

17.

X. Y.

Wang

Xiao

(2012). Family quality of life of Chinese families of children with intellectual disabilities. Journal of Intellectual Disability Research, 56(1), 30–44.

18.

Huang

A. X.

Jia

Wheeler

J. J.

(2013). Children with autism in the People’s Republic of China: Diagnosis, legal issues, and educational services. Journal of Autism and Developmental Disorders, 43, 1991–2001. https://doi.org/10.1007/s10803-012-1722-6

19.

Jiang

Pang

L. J.

Sun

(2017). Early childhood teacher education in China. Early Childhood Education in Chinese Societies, 19, 85–100. https://doi.org/10.1007/978-94-024-1004-4_6

20.

Jin

Yang

Liu

Huang

Jin

(2018). Prevalence of DSM-5 autism spectrum disorder among school-based children aged 3–12 years in Shanghai, China. Journal of Autism and Developmental Disorders, 48, 2434–2443.

21.

Krug

D. A.

Arick

Almond

(1980). Behavior checklist for identifying severity handicapped individuals with high levels of autistic behavior. Journal of Child Psychology and Psychiatry, 21(3), 221–229.

22.

Kuhaneck

H. M.

Madonna

Novak

Pearson

(2015). Effectiveness of interventions for children with autism spectrum disorder and their parents: A systematic review of family outcomes. American Journal of Occupational Therapy, 69(5), 1–14. https://doi.org/10.5014/ajot.2015.017855

23.

Landa

R. J.

(2018). Efficacy of early interventions for infants and young children with, and at risk for, autism spectrum disorders. International Review of Psychiatry, 30(1), 25–39.

24.

Liang

L.-H.

(2020, March 9). Millions of Chinese employees are working from home for the first time, as the nation tackles Covid-19. Will it change conservative corporate culture? BBC. https://www.bbc.com/worklife/article/20200309-coronavirus-covid-19-advice-chinas-work-at-home-experiment

25.

Lord

Risi

Lambrecht

Cook

E. H.

Leventhal

B. L.

DiLavore

P. C.

… Rutter

(2000). The autism diagnostic observation schedule-generic: A standard measure of social and communication deficits associated with the spectrum of autism. Journal of Autism and Developmental Disorders, 30(3), 205–223.

26.

Shi

X. C.

D. Y.

(2012). The poor oral health status of former heroin users treated with methadone in a Chinese city. Medical Science Monitor: International Medical Journal of Experimental and Clinical Research, 18(4), 51–55. https://doi.org/10.12659/msm.882611

27.

Maenner

M. J.

Shaw

K. A.

Bakian

A. V.

Bilder

D. A.

Durkin

M. S.

Esler

Cogswell

M. E.

(2021). Prevalence and characteristics of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 sites, United States, 2018. MMWR Surveillance Summaries, 70(11), 1–16. https://doi.org/10.15585/mmwr.ss7011a1

28.

McCabe

(2007). Parent advocacy in the face of adversity: Autism and families in the People’s Republic of China. Focus on Autism and Other Developmental Disabilities, 22(1), 39–50. https://doi.org/10.1177/10883576070220010501

29.

McCubbin

H. I.

Patterson

J. M.

(1983). The family stress process. Marriage & Family Review, 6, 7–37. doi:10.1300/J002v06n01_02

30.

McDevitt

S. E.

(2021). While quarantined: An online parent education and training model for families of children with autism in China. Research in Developmental Disabilities, 109, 103851. https://doi.org/10.1016/j.ridd.2020.103851

31.

Mumbardó-Adam

Barnet-López

Balboni

(2021). How have youth with Autism Spectrum Disorder managed quarantine derived from COVID-19 pandemic? An approach to families perspectives. Research in Developmental Disabilities, 110, 103860. https://doi.org/10.1016/j.ridd.2021.103860

32.

Muthén

L. K.

Muthén

B. O.

(2018). Mplus user’s guide (8th ed.). Muthén & Muthén.

33.

National Bureau of Statistics of China. (2021). The seventh national population census. http://www.stats.gov.cn/english/

34.

C. S. M.

Fang

Wang

Zhang

(2021). Potential factors of parenting stress in Chinese parents of children with autism spectrum disorder: A systematic review. Focus on Autism and Other Developmental Disabilities, 36(4), 237–248.

35.

Olson

Chen

Ibarra

Wang

Mash

Linke

Kinnear

Fishman

(2021). Externalizing behaviors are associated with increased parenting stress in caregivers of young children with autism. Journal of Autism and Developmental Disorders, 52, 975–986. https://doi.org/10.1007/s10803-021-04995-w

36.

Pang

Lee

C. M.

Wright

Shen

(2018). Challenges of case identification and diagnosis of autism spectrum disorders in China: A critical review of procedures, assessment, and diagnostic criteria. Research in Autism Spectrum Disorders, 53, 53–66.

37.

Poon

K. K.

Ooi

Bull

Bailey

D. B.

Jr. (2014). Psychometric validation of the Family Outcome Survey-Revised in Singapore. Research in Developmental Disabilities, 35(7), 1534–1543. https://doi.org/10.1016/j.ridd.2014.03.047

38.

R Core Team. (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/

39.

Revelle

(2020). psych: Procedures for personality and psychological research. Northwestern University. https://www.scholars.northwestern.edu/en/publications/psych-procedures-for-personality-and-psychological-research

40.

Rutter

LeCouteur

Lord

(2003). Autism diagnostic interview-revised manual. Los Angeles, CA: Western Psychological Services.

41.

Schopler

Reichler

R. J.

DeVellis

R. F.

Daly

(1980). Toward objective classification of childhood autism: Childhood Autism Rating Scale (CARS). Journal of Autism and Developmental Disorders, 10(1), 91–103.

42.

Sheridan

S. M.

Ryoo

J. H.

Garbacz

S. A.

Kunz

G. M.

Chumney

F. L.

(2013). The efficacy of conjoint behavioral consultation on parents and children in the home setting: Results of a randomized controlled trial. Journal of School Psychology, 51, 717–733. http://doi.org/10.1016/j.jsp.2013.09.003

43.

Sullivan

O. A.

Wang

(2020). Autism spectrum disorder interventions in mainland China: A systematic review. Review Journal of Autism and Developmental Disorders, 7(3), 263–277. https://doi.org/10.1007/s40489-019-00191-w

44.

Tint

Maughan

A. L.

Weiss

J. A.

(2017). Community participation of youth with intellectual disability and autism spectrum disorder. Journal of Intellectual Disability Research, 61(2), 168–180.

45.

Ueda

Yonemoto

Bailey

D. B.

Jr (2015). Psychometric validation of the Family Outcomes Survey-Revised in Japan. Research in Developmental Disabilities, 39, 55–66. https://doi.org/10.1016/j.ridd.2015.01.003

46.

Vinen

Clark

Paynter

Dissanayake

(2018). School age outcomes of children with autism spectrum disorder who received community-based early interventions. Journal of Autism and Developmental Disorders, 48(5), 1673–1683.

47.

Wainer

A. L.

Hepburn

McMahon Griffith

(2017). Remembering parents in parent-mediated early intervention: An approach to examining impact on parents and families. Autism, 21(1), 5–17. https://doi.org/10.1177/1362361315622411

48.

Wang

S. B.

Zhang

C. H.

Ungvari

G. S.

Cao

X. L.

J. P.

Hou

C. L.

Jia

F. J.

Xiang

Y. T.

(2018). The prevalence of autism spectrum disorders in China: a comprehensive meta-analysis. International Journal of Biological Sciences, 14(7), 717–725.

49.

Wang

Zhang

Zhao

Zhang

Jiang

(2020). Mitigate the effects of home confinement on children during the COVID-19 outbreak. The Lancet, 395(10228), 945–947. https://doi.org/10.1016/S0140-6736(20)30547-X

50.

Wang

Michaels

C. A.

Day

M. S.

(2011). Stresses and coping strategies of Chinese families with children with autism and other developmental disabilities. Journal of Autism and Developmental Disorders, 41(6), 783–795.

51.

Waschl

Poon

K. K.

Chen

Xie

(2021). Assessing family outcomes over time: Longitudinal measurement invariance of the Family Outcomes Survey–Revised. Topics in Early Childhood Special Education. Advance online publication. https://doi.org/10.1177/0271121421990838

52.

Wicks

Paynter

Adams

(2021). Exploring the predictors of family outcomes of early intervention for children on the autism spectrum: An Australian cohort study. Journal of Early Intervention, 43(1), 3–23. https://doi.org/10.1177/1053815119883413

53.

K. B.

Young

M. E.

Cai

(2012). Early child development in China: Breaking the cycle of poverty and improving future competitiveness. The World Bank.

54.

Zhang

S. C.

(2014). Chinese pediatricians face a crisis: Should they stay or leave? Pediatrics, 134(6), 1045–1047. https://doi.org/10.1542/peds.2014-1377

55.

Yang

Yao

Chen

Zhuang

Wang

Zhang

Lee

G. T.

(2018). A pilot study of a culturally adapted early intervention for young children with autism spectrum disorders in China. Journal of Early Intervention, 40(1), 52–68. https://doi.org/10.1177/1053815117748408

56.

Stronach

Harrison

A. J.

(2020). Public knowledge and stigma of autism spectrum disorder: Comparing China with the United States. Autism : The International Journal of Research and Practice, 24(6), 1531–1545.

57.

Zablotsky

Boswell

Smith

(2012). An evaluation of school involvement and satisfaction of parents of children with autism spectrum disorders. American Journal on Intellectual and Developmental Disabilities, 117, 316–330. http://doi.org/10.1352/1944-7558-117.4.316

58.

Zhou

R. Y.

B. X.

Wang

J. J.

(2022). Difficulties in the diagnosis and treatment of children with autism spectrum disorder in China. Journal of Autism and Developmental Disorders, 52(2), 959–961.

59.

Zuo

(2021, September 1). Over 5,200 private Chinese kindergartens went out of business in 2020. South China Morning Post. https://www.scmp.com/news/people-culture/social-welfare/article/3147161/over-5200-private-chinese-kindergartens-went-out