Direct and indirect costs for families of children with autism spectrum disorder in China

Abstract

Autism spectrum disorder has gained international attention due to its prevalence and the extent to which it can affect families. As a disorder without quantifiable treatment effects, it is easily overlooked in the battle for resources. Estimating family economic burdens and the specific factors that may be associated with them could help in the identification of resources and the promotion of social justice. We examined the family costs from a national family survey with Children with autism spectrum disorder aged 2–6 years (N = 3236). A three-tiered model to quantify the costs was used. The families’ average annual direct cost per child was $24,869.0, including direct medical cost (inpatient, outpatient,drugs, etc.) of $6009.2 and direct nonmedical cost (rehabilitation or educational, rental, transportation, care, and others) of $18,859.8. The annual indirect costs (productivity loss from resignation and job adjustment) for families were $13,990.7. The total costs were $34,206.5. The results of the regression revealed that the mothers’ educational level was significantly associated with costs. Families with an interprovincial immigration background, a higher-than-average income, or children with more severe autism spectrum disorder had a greater possibility of higher direct, indirect, and overall costs. Autism spectrum disorder imposed a significant financial burden on the families of children with autism spectrum disorder.

Lay abstract

This is the first comprehensive national study to explore the direct and indirect costs for families of children with autism spectrum disorder in China. The increasing prevalence of autism spectrum disorder highlights a growing need for resources to provide care for families of children with autism spectrum disorder. The medical and nonmedical costs and parents’ productivity loss have caused a serious burden on their families. Our objective is to estimate the direct and indirect costs for the families of children with autism spectrum disorder in China. The target population was parents of children with autism spectrum disorder. We analyzed the costs using cross-sectional data from a Chinese national family survey with children aged 2–6 years (N = 3236) who were clinically diagnosed with autism spectrum disorder. Family data from 30 provinces in China were obtained. Cost items included direct medical costs, direct nonmedical costs, and indirect costs. In this study, we found that the largest part of family costs for autism spectrum disorder are nonmedical costs and productivity loss. Autism spectrum disorder has imposed a huge economic burden on parents having children with autism spectrum disorder in China, who need more support than the current health care system provides.

Keywords

autism spectrum disorder direct costs employment status indirect costs productivity loss

Autism spectrum disorder (ASD) is a range of neurodevelopmental disorders that are characterized by impairments in social interaction and communication and restricted, repetitive behaviors (American Psychiatric Association, 2013). According to the most recent evidence from China, the prevalence of ASD among children aged 6 to 12 years was 0.7% (Zhou et al., 2020). As a developing country with a large population and an ineffective social security system, China faces a health crisis. Previous research has found that the costs of medical health care (inpatient and outpatient services, emergency care, pharmacy, and other health care professionals) for people with ASD are 2–10 times higher than for the general population (Mandell et al., 2006; Vohra et al., 2017). Nonmedical costs, such as transportation and special education expenses make up a larger portion of the total costs (Järbrink et al., 2003). Furthermore, more than half of the caregivers of children with ASD have reported having to stop working due to childcare issues (Saunders et al., 2015). Earnings for families with autistic children were 28% lower than for families with healthy children (Cidav et al., 2012).

In Western countries, studies about cost of illness (COI) are common. COI research paradigm emerged in the 1960s, when a method for estimating illness costs that included both direct and indirect costs became the standard for future research (Rice, 1966). Ganz et al. used this method to calculate the costs of ASD in the United States (Ganz, 2007). The value of goods and services used is measured in direct costs, while the value of lost productivity due to ASD is measured in indirect costs. Although COI studies for ASD were not rare most studies provided estimates within specific domains, which underestimated the cost of ASD (Cidav et al., 2012; Lavelle et al., 2014). There were fewer than 10 studies that provided a comprehensive (including direct and indirect costs) estimate of costs: two from the United States (Ganz, 2007; Leigh & Du, 2015), three from the United Kingdom (Järbrink & Knapp, 2001; Knapp et al., 2009; MacKay et al., 2017), one from Sweden (Järbrink, 2007), one from Ireland (Roddy & O’Neill, 2019), one from South Korea (Hong et al., 2020) and one included estimated both the United States and the United Kingdom (Buescher et al., 2014). While these studies have contributed significantly to our understanding of this topic, they are limited in the following ways. First, many studies relied on assumptions for costs for which no published data existed. There were few studies on ASD-related costs based on real-world service use. Second, the indirect costs were only calculated for parents who took temporary or permanent leave or switched to part-time work (Hong et al., 2020; Roddy & O’Neill, 2019). Salary reduction due to changing the type of work (without reducing working hours) has not been thoroughly studied.

Although mental disability has received increased attention in China and some studies have used the COI framework to update the economic burden estimates for some mental disabilities, such as dementia (Xu et al., 2017) and cerebral palsy (B. Wang et al., 2008), there are few comprehensive cost estimates for ASD in China. Existing studies on the costs of ASD in China are either regional (Wu, 2018), or have a small sample size (Ou et al., 2015; Xiong et al., 2011), all of which lack a comprehensive perspective and pay no attention to very young children (Zhou et al., 2022). Preschool-age children are important because this is when early intervention programs are encouraged and most money is spent (Alnemary et al., 2017; Shepherd et al., 2018). Aside from describing the economic burden, it would be useful to identify factors associated with cost. Previous research found that parents with higher education or income level (DeRigne, 2012; Liptak et al., 2008) and children with a clinically more severe condition (Raz et al., 2013; Roddy & O’Neill, 2019) had a higher direct cost. However, the influencing factors in China are likely to differ from those in high-income countries and we do not know which specific factors in China may be associated with different types of costs.

As a result, we designed a nationwide survey to investigate the costs. Specifically, (1) we aimed to estimate the direct and indirect cost among children with ASD in China over a period of 12 months, and (2) we investigated the children’s demographics characters and social economic factors associated with the economic burden of ASD. A detailed description of the distribution of ASD costs, as well as the identification of the influencing factors can shed light on the driving force of the cost and inform policy-makers on how to mobilize resources to benefit vulnerable patients.

Methods

Data sources

This study used data from the Survey on Family Circumstances and Demand for Support and Resources among Autistic Children in China (FCDSR). It was a survey that was distributed to members of the AlsoLife online patient community. More than 200,000 parents of ASD children can share information about their conditions, treatments, symptoms, and comorbidities on that platform, which is the one of the largest online gathering place for parents with ASD children in China. The Quality Assurance staff at China Association of Rehabilitation of Disabled Persons (CARDP) reviewed the survey for editorial and technical suggestions, which aimed to describe the family information, treatment, rehabilitation subsidies, and health expenditure of ASD children. The survey was available online from 15 September to 30 September 2020. We did not use a sampling design because there is no nationwide ASD survey in China. A pilot field study (N = 20) was conducted to refine the instrument and data collection procedures, and the results indicated that respondents generally understood the questionnaire, so only minor wording changes were made.

Data collection

Families with children diagnosed with ASD were recruited if they met the following criteria: (1) the child was between the ages of 2 and 6 years and was diagnosed with or suspected (for children under 3 years old) of having ASD at a hospital; (2) the hospital had diagnostic qualifications and followed a DSM-5 (Diagnostic and Statistical Manual of Mental Disorders 5th ed.; American Psychiatric Association, 2013) standard, not only through scale measurement but also via medical professional diagnosis; and (3) the child did not have any other mental and physical comorbidities.

A total of 5014 households were investigated, with 3236 (64.53%) participating in the study. Figure 1 presents the selection process. Thirty provinces and a total of 380 cities in China were included. There is no large-scale survey about children with ASD in China; the second national sample survey of disabled people could be used as a reference. These samples are representative (see eTable 1 and eTable 2 in the Supplement for details).

Figure 1.

Flowchart.

Measures

We categorized all the costs into three classes: (1) direct medical costs, (2) direct nonmedical costs, and (3) indirect costs resulting from ASD-related loss of productivity. A detailed breakdown of the cost components was given in eTable 3 in the Supplement.

Direct medical costs were the costs for diagnostics, therapeutics, and care. Direct nonmedical costs were those associated with the illness, but not spent on health resources, such as rehabilitation treatment outside of hospitals, travel to health care providers, renting a house to get trained or treatment, receiving care and assistance, and so on. Based on the literature (Alnemary et al., 2017; Green et al., 2006; Lv et al., 2008; Shepherd et al., 2018; Y. G. Wang et al., 2019; F. Wang & Yang, 2017), a total of 24 nonmedical treatment services were chosen, as described in eTable 4. The types of services, frequency of use, and associated costs were obtained.

Indirect costs were defined as productivity loss in which sporadic work loss was associated with health service use and extended work loss due to a disability or death (Onukwugha et al., 2016). We assumed indirect costs incurred when family members quit or changed jobs to care for ASD children in this study. The productivity losses due to job changes that did not reduce working time for providing informal care were also included. Instead of using human capital approach, the direct opportunity cost approach was used (see eMethods in the Supplement for details). The cost of resignation was assumed to be the parent’s yearly labor wage if she or he had chosen to work (yearly labor wage multiplied by the duration of their leave over the 12-month period). The cost of job adjustment was determined by multiplying the reduced salary as a result of the job change by the length of time since the job change over a 12-month period (eTable 3). We encouraged parents to check documentation such as payroll records to reduce recall errors.

We made no distinction between direct costs and indirect costs related to medical care. It was difficult to provide an accurate estimate of the time parents spent on various informal care activities (Järbrink et al., 2003). Indirect costs were calculated from the perspective of employment burden, consistent with previous studies (Buescher et al., 2014; Roddy & O’Neill, 2019; Zhao et al., 2011). On a household basis, the total cost was calculated by adding the direct and indirect costs.

Data on the children’s demographic, family socioeconomic background and treatment history were collected. The age of the children was their age at the survey point. The regional variables were “eastern,” “central,” and “western.” The provinces in the eastern region were among the first to implement the coastal opening-up policy and had a high level of economic development. The provinces of the central region were economically underdeveloped, while those of the western region are even less so. The family’s interprovincial residence change was included. We classified family income into three categories. According to the data distribution, the below-average group had an annual income of less than $12,327 the around-average group had an annual income of between $12,327 and $23,112, and the above-average group had an annual income of more than $23,112.

Statistical analysis

We used three-tiered model to estimate the families’ direct medical costs, direct nonmedical costs, and indirect costs of the families. The indirect costs for family were the sum of the father’s indirect costs and the mother’s indirect costs in a family. The indirect costs were zero if the family members had no productivity loss. Logistic regression models were used to identify the factors influencing the costs. We divided the annual costs into quartiles, and calculated two separate multivariable logistic regression models—one for the upper quartile, and the other for the lower quartile of costs. Associations between predictors and independent variables were reported by odds ratios (ORs) and their 95% confidence intervals (CIs). All the estimated costs were converted to US dollar (US$) values in January 2021, when one US$ was equivalent to about 6.49 Chinese yuan. All statistical analyses were conducted using SPSS 22.0 for Windows (SPSS Inc., Chicago, IL, USA).

Consent and ethics approval

All families provided electronic informed consent before enrollment. All procedures involving human subjects/patients were approved by the ethics committee of Peking University Institutional Review Board and the approval number was IRB00001052-20016.

Community involvement

When deciding on the research topic and designing the questionnaire, we consulted four clinical psychologists, all of whom were ASD specialists. Furthermore, 57 parents with autistic children in rehabilitation institutions participated in interviews when we designed the survey (not the pilot testing group). This provided the survey’s original framework and compensated for previous research’s shortcomings, especially regarding indirect cost items noted by parents (e.g. many parents changed to a low-pay job, but they did not reduce their working hours). Parents were not involved in the study’s design or execution. The general results (no personal data) would be distributed on demand.

Results

Study population

A total of 3236 households were included in this survey. A total of 82.7% of the children were boys, and the mean age was 3.7 (SD: 1.0) years old, with middle function being the most common diagnosis (38.9%). A total of 67.0% of the mothers of ASD children had a higher education degree, while 66.9% of the fathers of ASD children had a higher education degree. 17.7% of interprovincial residence changes occurred. The eastern district accounted for 62.2% of the residences. The average household income was $22,650.2 (SD: $23,882.9; median: $15,408.3). Table 1 described the study population. See eTables 1 and 2 for information on specific provincial distribution and quality inspection.

Table 1.

Study population.

Characteristics	N/M	%/SD
Children’ demographics
Age	3.7	1.0
Sex
Male	2676	82.7
Female	560	17.3
ASD severity
LFA	709	21.9
MFA	1259	38.9
HFA	523	16.2
Undetermined	745	23.0
Only child
No	1598	49.4
Yes	1638	50.6
Social economic status
Household income
Below average	1154	35.7
Around average	1220	37.7
Above average	862	26.6
Higher education degree^a—father
No	1072	33.1
Yes	2164	66.9
Higher education degree^a—mother
No	1068	33.0
Yes	2168	67.0
Resident district
Eastern	1819	56.2
Central	1112	34.4
Western	305	9.4
Migrating inter provincial
No	2662	82.3
Yes	574	17.7
Average annual income (US$)	22,650.2	23,882.9

LFA: low-functioning autism; MFA: middle-functioning autism; HFA: high-functioning autism; N: number; M: mean; SD: standard deviation.

Including those who have obtained college degree, bachelor’s degree, master’s degree, or doctor’s degree.

Treatment service utilization in nonmedical setting

Applied Behavior Analysis (ABA) and sensory integration training were two of the most common rehabilitation methods. A total of 66.7% of the families used ABA-related treatment methods, while 44.4% of the families used sensory integration training. The social story method was also adopted by 19.8% of families. The detailed treatment service utilization is provided in eTable 4.

Annual direct costs

Direct medical costs averaged $6009.2 per year. The average cost of rehabilitation or education was $13,598.0. The average cost of living was $3364.1. The average cost of transportation was $1155.0. Care and assistance costs averaged $7501.9 and other costs averaged $4149.8. The average annual nonmedical costs amounted to $18,859.8. The annual total direct costs were $24,869.0 (Table 2). Medical costs accounted for 49.03% of annual household income. The costs of rehabilitation or education accounted for 84.35% of the annual household income. Living costs accounted for 29.13%. Transportation costs accounted for 8.57%. Care and assistance costs accounted for 41.23%, while other costs accounted for 31.14%. The total direct costs were 110.0% of the annual family income.

Table 2.

Average direct costs^a reported per family for the past 12 months.

Type	% of subjects	Mean (US$)	SD (US$)	Range (US$)	Median (US$)
Medical costs	69.5	6009.2	5775.0	0–38,520.8	1540.8
Nonmedical costs	96.3	18,859.8	16,923.1	0–214,231.0	12,307.5
Rehabilitation or education costs	92.1	13,598.0	13,083.1	0–77,041.6	6933.7
Rental costs	45.4	3364.1	3148.8	0–38,520.8	0.0
Transportation costs	71.0	1155.0	1302.0	0–38,520.8	380.2
Care and assistance costs	8.1	7501.9	5017.4	0–46,225.0	0.0
Other costs	42.8	4149.8	4178.2	0–38,520.8	0.0
Sum^b	97.2	24,869.0	20,416.0	0–231,124.8	15,408.3

ASD: autism spectrum disorder; SD: standard deviation.

Estimates are based solely on the annual expenditure reported by families.

Different households have different types of expenditure. Sum is not equal to the sum of sub items.

Annual indirect costs

The fathers’ and mothers’ work statuses were combined to create a cross-table of the family’s overall work status (Table 3). In 43.3% (1402) of the families, at least one member changed jobs or resigned, resulting in a loss of family productivity. The total annual indirect costs resulting from job changes or resignation were $13,990.7, accounting for 68.5% of household income.

Table 3.

Household indirect costs* (US$) under different work status combinations in families.

Work status		Mothers				Sum
Work status		Unchanged^a	Overtime	Flexible^b	Resignation	Sum
Fathers	Unchanged^a	1777 [27,796.6] (−0.0)	16 [22,311.2] (−0.0)	166 [21,340.5] (−9738.0)	418 [21,910.6] (−17,842.8)	2377 [23,836.6] (−12,496.1)
	Overtime	36 [20,847.4] (−0.0)	5 [18,489.9] (−0.0)	239 [17,226.5] (−10,107.8)	133 [16,255.7] (−19,645.6)	413 [17,257.3] (−13,590.1)
	Flexible^b	110 [21,633.2] (−12,650.2)	8 [28,043.1] (−12,018.4)	158 [21,941.4] (−19,414.4)	90 [16,240.3] (−31,140.2)	366 [20,909.0] (−19,969.1)
	Resignation	32 [18,520.8] (−20,277.3)	7 [17,827.4] (−13,204.9)	22 [13,697.9] (−27,365.1)	19 [11,155.6] (−21,263.4)	80 [15,624.0] (−21,848.9)
Sum		1955 [26,517.7] (−14,530)	43 [22,295.8] (−12,511.5)	585 [20,631.7] (−11,864.4)	660 [19,830.5] (−19,922.9)	3,236 [22,650.2] (−13,990.7)

The indirect costs are shown in the form of negative value in parentheses, and the households with 0 loss are not averaged in the total.

Unchanged means that after the child is diagnosed, the working status of the father or mother remains unchanged (still continuing the past work or still not working).

Flexible includes “reduce workload,” “shorten work time,” or “engaged in Child Rehabilitation work” or “took Long leave.”

The data outside the brackets are the number of households.

The data inside the square brackets are the annual household income (US$) for that kind of household.

The data inside the round brackets are the indirect costs (US$) for that kind of household.

Changes in employment status and indirect costs by sex were presented in eTables 5 and 6 from a personal rather than a family perspective. Resignation to provide childcare was much higher among mothers than fathers (20.4% vs 2.5%), and the likelihood of overwork was much higher for fathers than for mothers (12.8% vs 1.1%). Mothers who resigned involuntarily were significantly more likely to have higher previous salaries than mothers who were voluntarily unemployed. There were no similar findings for fathers (see eTable 7 for details).

Total costs and subsidies

The total costs were $34,206.5, which accounted for 151.0% of household income. The total costs were not the sum of direct and indirect costs as we calculated only families whose costs were greater than zero. Most medical costs were not reimbursed by the government. In China, the only subsidy was rehabilitation assistance, which averaged $3226.5 per family and accounted for less than 10% of total costs (Figure 2). A total of 54.2% of the families had received subsidies in the previous year. The main reason for not receiving subsidies was that they were ineligible (eTable 8). A total of 83.6% of families believed that these costs put them under a lot of financial stress, and the families’ main source of anxiety was financial stress (eTable 9). The majority of families (82.9%) had no savings at all. According to the interviews and survey responses on “how to resolve the continuous economic pressure in the future,” many families raised funds by selling their apartments and borrowing from relatives.

Figure 2.

Costs and subsidies.

Predictors of costs

In terms of direct costs, having a well-educated mother was associated with a greater likelihood of having a high cost (OR = 1.71, 95% CI: 1.38–2.12). A higher family income was associated with a greater likelihood of having a high cost (OR = 1.31, 95% CI: 1.04–1.64 for around average income; OR = 3.12, 95% CI: 2.47–3.94 for above average income). Compared with families with high-functioning ASD children, families with middle-functioning ASD children had a greater likelihood of having a high cost (OR = 1.28, 95% CI: 1.01–1.64). An increased chance of high costs was linked to having experience with immigration (OR = 2.69, 95% CI: 2.19–3.31) (Table 4, Model 1). For indirect costs, having a well-educated mother was associated with a lower likelihood of having a high cost (OR = 0.72, 95% CI: 0.61–0.86). Middle- or low-functioning children were more likely to have families with high costs (OR = 1.36, 95% CI: 1.13–1.63 for middle-functioning children; OR = 1.24, 95% CI: 1.00–1.55 for low-functioning children). Having experience with interprovincial immigration was associated with a higher likelihood of high cost (OR = 1.91; 95% CI: 1.49–2.43) (Table 4, Model 2). For total costs, higher family income, having middle- or low-functioning ASD children, and having an interprovincial immigration experience were associated with a higher likelihood of high costs (Table 4, Model 3). In this study, the sex and age of the children were not associated with the costs.

Table 4.

Multivariable logistic regression models for three costs.

	Model 1: Direct cost			Model 2: Indirect cost			Model 3: Total cost
	95% CI			95% CI			95% CI
Variables	OR	Low	High	OR	Low	High	OR	Low	High
Mother’s education
No higher education	1.00
Higher education	1.71	1.38	2.12	0.72	0.61	0.86
Household income
Below average	1.00
Around average	1.31	1.04	1.64				1.23	1.01	1.50
Above average	3.12	2.47	3.94				1.94	1.58	2.40
Birth order
1st	1.00
2nd	0.86	0.72	1.03	0.85	0.71	1.01
3rd+
ASD severity
High function	1.00
Middle function	1.28	1.01	1.64	1.36	1.13	1.63	1.30	1.05	1.69
Low function				1.24	1.00	1.55	1.16	1.01	1.42
Undetermined
District
Eastern
Central
Western				0.78	0.58	1.04
Immigrate
No	1.00
Yes	2.69	2.19	3.31	1.91	1.49	2.43	2.29	1.80	2.91
Rehabilitation setting
Institution	1.00
Home	0.53	0.43	0.66	0.84	0.70	1.02	0.57	0.47	0.70

OR: odds ratio; CI: confidence interval.

Independent variables were entered using the stepwise forward method by likelihood ratio, with p(in) = 0.20 and p(out) = 0.25. The variables that did not enter any of the models were children’s sex and age, fathers’ education degree.

Dependent variables: Model 1: Upper quartile of direct costs in comparison with all other quartiles.

Model 2: Upper quartile of indirect costs in comparison with all other quartiles.

Model 3: Upper quartile of total costs in comparison with all other quartiles.

Discussion

The prevalence of ASD has increased significantly in the last decades, indicating a growing need for resources to care for people with ASD. However, few COI analysis of ASD has been conducted in China to date. This was the first study to comprehensively examine family costs of ASD using a national dataset in China.

The annual direct costs were $24,869.0, with nonmedical costs accounting for three-quarters of the total. The annual direct costs were higher than those reported in Israeli ($8289, 4–10 years, outpatient service, respite care, medication use, and others; Raz et al., 2013) and in Ireland (€9489.60, 2–18 years, living, care and assistance, education, medical use, and others; Roddy & O’Neill, 2019). The indirect cost amounted to $13,990.7 per year, which was higher than those reported in the United States ($6207.70, Montes & Halterman, 2008), but was lower to the cost reported in Australia ($29,200, Horlin et al., 2014,). It is inappropriate to make direct comparisons between findings due to differences in health care systems and cost estimation methods, but a brief comparison of cost drivers is worthwhile. Direct nonmedical costs (particularly rehabilitation costs) and indirect costs were the main cost divers in our study, accounting for 48.5% and 36.0% of total costs, respectively, which was highly consistent with previous studies (Buescher et al., 2014; Horlin et al., 2014; Järbrink et al., 2003; MacKay et al., 2017). We discovered that direct nonmedical cost was roughly three times of medical cost, which was consistent with previous research findings that annual nonmedical costs were significantly higher than medical costs (Lavelle et al., 2014).

We explored the key factors relating to direct and indirect costs. Families with a well-educated mother or an above-average income were more likely to have higher direct costs. The positive effects of income and education were consistent with previous studies (Anderson et al., 2007; DeRigne, 2012; Liptak et al., 2008). Less educated parents may have had lower access to specialist services resulting in lower direct costs (Liptak et al., 2008; Thomas et al., 2007). In contrast, having a well-educated mother reduced the indirect costs. This was consistent with a previous finding that families with a high education background could afford to hire someone to care for their autistic children, indicating a negative relationship between education and indirect costs (Ou et al., 2015). The disadvantaged socioeconomic groups have a greater economic burden (Boyden et al., 2022; Wiggins et al., 2020). This study also found that the severity of children’s ASD had a significant impact on costs. Middle- and low-functioning children were more likely to incur higher cost. This was consistent with previous findings (Raz et al., 2013; Roddy & O’Neill, 2019). Low-functioning children needed more training and company, putting a greater financial strain on their families. We found no significant correlation between costs and children’s sex or age, which contradicted previous research (Barrett et al., 2012; Hong et al., 2020). The relatively narrow age range in our sample made identifying some determinants difficult, and cultural differences might play a role.

The findings of this study confirmed the gender division of labor. The proportion of quitting jobs was 2.5% for fathers and 20.4% for mothers, with mothers who resigned to care for autistic children earning significantly more than the average. There were no similar findings in fathers. This meant that mothers with autistic children faced higher opportunity costs. Previous research has found that childcare demands have a negative impact on married women’s labor supply (Mahringer & Zulehner, 2015; Viitanen, 2005). In the United States, 34.9% of families stated that they required additional income to cover the costs of children with ASD (Kogan et al., 2008). In this study, the proportion of fathers who overworked was 12.8% and 1.1% for mothers, indicating that fathers were still the breadwinners.

It was worth noting that rehabilitation resources were unevenly distributed in China (Zhang & Coyte, 2020), and many families forced to relocate in order to access available treatment resources. In our study, 39.8% of families relocated to better accommodate their children’s treatment. Interprovincial migration accounted for 17.7% of total migrations, resulting in higher living costs and a lower likelihood of obtaining subsidies and medical copayments (Gong et al., 2012). For families undertook interprovincial migration, 51.9% migrated from the central and western regions to the eastern region. Beijing, Guangdong, and Zhejiang were the top three migration destination provinces, all of which belonged to eastern districts. Jiangxi, Anhui, and Henan were the top three provinces in terms of migration outflow, all of which were central provinces. According to the China Disabled Persons’ Federation’s list of designated service institutions for the rehabilitation of autistic children, there were 2304 designated ASD rehabilitation institutions in China in 2019, with the central region accounting for only 22.2%. Previous research has found that the resources distribution within a country affects health output (Parish et al., 2012), and China’s insufficient allocation of resources to the central and western regions may result in unfairness.

The family did not bear all of the financial burdens,which could be measured from different perspectives (Beecham, 2014). In China, health care budget allocations were heavily skewed toward somatic diseases, with mental disorders receiving less than 1% of total health care spending (World Health Organization (WHO), 2010). A much larger proportion of the population has excessively high health care expenses in comparison with their annual disposable income than in developed countries (Xu et al., 2003). China’s health care financing systems are fragmented and inefficient (Hanson et al., 2022; Meng, 2022). Most families can only receive social medical assistance, as detailed in eBackground. In our study, nearly half of the families still did not receive rehabilitation subsidies, and those who received were still under a lot of stress (eTable 9). Consistent with previous research (Wang et al., 2012), we found that the families paid majority of the costs out of their own pockets. The government should provide more welfare to families of children with do not have to ASD, especially to unemployed caregivers, so that they give up their income to provide care, nor postpone the responsibility of care to make up for income.

Limitations

Several limitations of the study should be noted. First, the sample was drawn from a network survey. All families were invited to participate in and completed an electronic questionnaire, which did not allow for control stratification in sampling. Although the sex ratio and family location distribution are consistent with the main data of the Second National Survey of Disabled People (SNSDP), we will adopt more rigorous sampling methods to improve our research in the future. Second, parents were asked to recall the costs, income, and service utilization, which might result in recall bias or error. Obtaining more objective indicators is critical for future improvement. Third, some academics have proposed new cost classifications in recent years, such as intangible costs (Chevreul et al., 2013). Future studies should examine costs from various perspectives.

Conclusion and implications

To the best of our knowledge, this is the first comprehensive COI study for ASD families in China, as well as one of the few studies that provides a thorough investigation into personal characteristics and sociodemographic factors associated with ASD costs. Our research design reduces underestimation of costs because the detailed classification of costs. ASD is associated with significant financial and employment burdens in Chinese families with preschool-aged children, particularly in low-income families, families with low-functioning children, and families who have migrated. Families with autistic children require more assistance than the current health care system can provide. Addressing these families’ needs should not only provid additional support and resources, but also carefully reevaluate regional policies. Policy changes should include adding paid leave or related financial supports, reducing subsidies barriers and balancing regional rehabilitation resources. Coordination between agencies and professionals is required. In the future, better differentiating costs and classifying comorbidities will aid in the development of more reasonable policies.

Supplemental Material

sj-docx-1-aut-10.1177_13623613231158862 – Supplemental material for Direct and indirect costs for families of children with autism spectrum disorder in China

Supplemental material, sj-docx-1-aut-10.1177_13623613231158862 for Direct and indirect costs for families of children with autism spectrum disorder in China by Yanan Zhao, Yanan Luo, Rong Zhang and Xiaoying Zheng in Autism

Footnotes

Acknowledgements

The authors are grateful to Liu Daiyue and Xu Disha at ALSOLIFE, He Ping at Peking Universtiy for their contributions. The authors are grateful to the anonymous reviewers for their detailed comments and supports. They also thank all the parents who participated in the study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: All phases of this study were supported by Beijing Municipal Science & Technology Commission grant, Z181100001518005. The Key Realm R&D Program of Guangdong Province, 2019B030335001.

ORCID iD

Yanan Zhao

Supplemental material

Supplemental material for this article is available online.

References

Alnemary

F. M.

Aldhalaan

H. M.

Simon-Cereijido

Alnemary

F. M.

(2017). Services for children with autism in the Kingdom of Saudi Arabia. Autism, 21(5), 592–602. https://doi.org/10.1177/1362361316664868

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-5®).

Anderson

Dumont

Jacobs

Azzaria

(2007). The personal costs of caring for a child with a disability: A review of the literature. Public Health Reports, 122(1), 3–16. https://doi.org/10.1177/003335490712200102

Barrett

Byford

Sharac

Hudry

Leadbitter

Temple

Aldred

Slonims

Green

; PACT Consortium. (2012). Service and wider societal costs of very young children with autism in the UK. Journal of Autism and Developmental Disorders, 42(5), 797–804. https://doi.org/10.1007/s10803-011-1306-x

Beecham

(2014). Annual Research Review: Child and adolescent mental health interventions: A review of progress in economic studies across different disorders. Journal of Child Psychology and Psychiatry, 55(6), 714–732. https://doi.org/10.1111/jcpp.12216

Boyden

J. Y.

Hill

D. L.

Nye

R. T.

Bona

Johnston

E. E.

Hinds

Friebert

Kang

T. I.

Hays

Hall

Wolfe

Feudtner

; PPCRN SHARE Project Group. (2022). Pediatric palliative care parents’ distress, financial difficulty, and child symptoms. J Pain Symptom Manage, 63(2), 271–282. https://doi.org/10.1016/j.jpainsymman.2021.08.004

Buescher

A. V.

Cidav

Knapp

Mandell

D. S.

(2014). Costs of autism spectrum disorders in the United Kingdom and the United States. JAMA Pediatrics, 168(8), 721–728. https://doi.org/10.1001/jamapediatrics.2014.210

Chevreul

Prigent

Bourmaud

Leboyer

Durand-Zaleski

(2013). The cost of mental disorders in France. European Neuropsychopharmacology, 23(8), 879–886. https://doi.org/10.1016/j.euroneuro.2012.08.012

Cidav

Marcus

S. C.

Mandell

D. S.

(2012). Implications of childhood autism for parental employment and earnings. Pediatrics, 129(4), 617–623. https://doi.org/10.1542/peds.2011-2700

10.

DeRigne

L. A.

(2012). The employment and financial effects on families raising children with special health care needs: An examination of the evidence. Journal of Pediatric Health Care, 26(4), 283–290. https://doi.org/10.1016/j.pedhc.2010.12.006

11.

Ganz

M. L.

(2007). The lifetime distribution of the incremental societal costs of autism. Archives of Pediatrics & Adolescent Medicine, 161(4), 343–349. https://doi.org/10.1001/archpedi.161.4.343

12.

Gong

Liang

Carlton

E. J.

Jiang

Wang

Remais

J. V.

(2012). Urbanization and health in China. Lancet, 379(9818), 843–852. https://doi.org/10.1016/S0140-6736(11)61878-3

13.

Green

V. A.

Pituch

K. A.

Itchon

Choi

O’Reilly

Sigafoos

(2006). Internet survey of treatments used by parents of children with autism. Research in Developmental Disabilities, 27(1), 70–84. https://doi.org/10.1016/j.ridd.2004.12.002

14.

Hanson

Brikci

Erlangga

Alebachew

De Allegri

Balabanova

Blecher

Cashin

Esperato

Hipgrave

Kalisa

Kurowski

Meng

Morgan

Mtei

Nolte

Onoka

Powell-Jackson

Roland

Wurie

(2022). The Lancet Global Health Commission on financing primary health care: Putting people at the centre. Lancet Global Health, 10(5), e715–e772. https://doi.org/10.1016/S2214-109X(22)00005-5

15.

Hong

Lee

S. M.

Park

Yoon

S. J.

Kim

Y. E.

I. H.

(2020). Prevalence and economic burden of autism spectrum disorder in South Korea using national health insurance data from 2008 to 2015. Journal of Autism and Developmental Disorders, 50(1), 333–339. https://doi.org/10.1007/s10803-019-04255-y

16.

Horlin

Falkmer

Parsons

Albrecht

M. A.

Falkmer

(2014). The cost of autism spectrum disorders. PLOS ONE, 9(9), e106552. https://doi.org/10.1371/journal.pone.0106552

17.

Järbrink

(2007). The economic consequences of autistic spectrum disorder among children in a Swedish municipality. Autism, 11(5), 453–463. https://doi.org/10.1177/1362361307079602

18.

Järbrink

Fombonne

Knapp

(2003). Measuring the parental, service and cost impacts of children with autistic spectrum disorder: A pilot study. Journal of Autism and Developmental Disorders, 33(4), 395–402. https://doi.org/10.1023/a:1025058711465

19.

Järbrink

Knapp

(2001). The economic impact of autism on Britain. Autism, 5(1), 7–22. https://doi.org/10.1177/1362361301005001002

20.

Knapp

Romeo

Beecham

(2009). Economic cost of autism in the UK. Autism, 13(3), 317–336. https://doi.org/10.1177/1362361309104246

21.

Kogan

M. D.

Strickland

B. B.

Blumberg

S. J.

Singh

G. K.

Perrin

J. M.

van Dyck

P. C.

(2008). A national profile of the health care experiences and family impact of autism spectrum disorder among children in the United States, 2005-2006. Pediatrics, 122(6), 1149–1158. https://doi.org/10.1542/peds.2008-1057

22.

Lavelle

T. A.

Weinstein

M. C.

Newhouse

J. P.

Munir

Kuhlthau

K. A.

Prosser

L. A.

(2014). Economic burden of childhood autism spectrum disorders. Pediatrics, 133(3), e520–e529. https://doi.org/10.1542/peds.2013-0763

23.

Leigh

J. P.

(2015). Brief report: Forecasting the economic burden of autism in 2015 and 2025 in the United States. Journal of Autism and Developmental Disorders, 45(12), 4135–4139. https://doi.org/10.1007/s10803-015-2521-7

24.

Liptak

G. S.

Benzoni

L. B.

Mruzek

D. W.

Nolan

K. W.

Thingvoll

M. A.

Wade

C. M.

Fryer

G. E.

(2008). Disparities in diagnosis and access to health services for children with autism: Data from the National Survey of Children’s Health. Journal of Developmental and Behavioral Pediatrics, 29(3), 152–160. https://doi.org/10.1097/DBP.0b013e318165c7a0

25.

C. X.

Zhang

Liu

(2008). A study on the current situation of treatment and rehabilitation services for autistic children in Tianjin. Chinese Journal of Rehabilitation Medicine, 23(4), 353–355. https://doi.org/1001-1242(2008)-04-0353-03 (in Chinese).

26.

MacKay

Knapp

Boyle

J. M.

Lemmi

Connolly

Rehill

(2017). The microsegmentation of the autism spectrum: Economic and research implications for Scotland. The Scottish Government. https://doi.org/10.13140/RG.2.2.18206.23365

27.

Mahringer

Zulehner

(2015). Child-care costs and mothers’ employment rates: An empirical analysis for Austria. Rev Econ Household, 13(4), 837–870. https://doi.org/10.1007/s11150-013-9222-9

28.

Mandell

D. S.

Cao

Ittenbach

Pinto-Martin

(2006). Medicaid expenditures for children with autistic spectrum disorders: 1994 to 1999. Journal of Autism and Developmental Disorders, 36(4), 475–485. https://doi.org/10.1007/s10803-006-0088-z

29.

Meng

Q. Y.

(2022). Strengthening public health systems in China. Lancet Public Health, 7(12), e987–e988. https://doi.org/10.1016/S2468-2667(22)00206-7

30.

Montes

Halterman

J. S.

(2008). Association of childhood autism spectrum disorders and loss of family income. Pediatrics, 121(4), e821–e826. https://doi.org/10.1542/peds.2007-1594

31.

Onukwugha

McRae

Kravetz

Varga

Khairnar

Mullins

C. D.

(2016). Cost-of-illness studies: An updated review of current methods. Pharmacoeconomics, 34(1), 43–58. https://doi.org/10.1007/s40273-015-0325-4

32.

J. J.

Shi

L. J.

Xun

G. L.

Chen

R. R.

Luo

X. R.

Zhang

F. Y.

Zhao

J. P.

(2015). Employment and financial burden of families with preschool children diagnosed with autism spectrum disorders in urban China: Results from a descriptive study. BMC Psychiatry, 15(1), 3. https://doi.org/10.1186/s12888-015-0382-4

33.

Parish

S. L.

Rose

R. A.

Dababnah

Yoo

Cassiman

S. A.

(2012). State-level income inequality and family burden of US families raising children with special health care needs. Social Science Medicine, 74(3), 399–407. https://doi.org/10.1016/j.socscimed.2011.10.035

34.

Raz

Lerner-Geva

Leon

Chodick

Gabis

L. V.

(2013). A survey of out-of-pocket expenditures for children with autism spectrum disorder in Israel. Journal of Autism and Developmental Disorders, 43(10), 2295–2302. https://doi.org/10.1007/s10803-013-1782-2

35.

Rice

D. P.

(1966). Estimating the cost of illness [In Health economics series No. 6. PHS Pub. No. 947-6]. US Government Printing Office.

36.

Roddy

O’Neill

(2019). The economic costs and its predictors for childhood autism spectrum disorders in Ireland: How is the burden distributed? Autism, 23(5), 1106–1118. https://doi.org/10.1177/1362361318801586

37.

Saunders

B. S.

Tilford

J. M.

Fussell

J. J.

Schulz

E. G.

Casey

P. H.

Kuo

D. Z.

(2015). Financial and employment impact of intellectual disability on families of children with autism. Families, Systems and Health, 33(1), 36–45. https://doi.org/10.1037/fsh0000102

38.

Shepherd

Csako

Landon

Goedeke

(2018). Documenting and understanding parent’s treatment choices for their child with autism spectrum disorder. Journal of Autism and Developmental Disorders, 48(4), 988–1001. https://doi.org/10.1007/s10803-017-3395-7

39.

Thomas

K. C.

Ellis

A. R.

McLaurin

Daniels

Morrissey

J. P.

(2007). Access to care for autism-related services. Journal of Autism and Developmental Disorders, 37(10), 1902–1912. https://doi.org/10.1007/s10803-006-0323-7

40.

Viitanen

T. K.

(2005). Cost of Childcare and Female Employment in the UK. Labor, 19(1), 149–170. https://doi.org/10.1111/j.1467-9914.2005.00325.x

41.

Vohra

Madhavan

Sambamoorthi

(2017). Comorbidity prevalence, healthcare utilization, and expenditures of Medicaid enrolled adults with autism spectrum disorders. Autism, 21(8), 995–1009. https://doi.org/10.1177/1362361316665222

42.

Wang

Chen

Zhang

Guo

Hailey

(2008). A preliminary study into the economic burden of cerebral palsy in China. Health Policy, 87(2), 223–234. https://doi.org/10.1016/j.healthpol.2008.01.001

43.

Wang

Yang

G. X.

(2017). On the investigation into and the analysis of the intervention and rehabilitations status of ASD in China. Medicine and Philosophy, 38(10), 49–54. https://doi.org/1:10.12014/j.issn.1002-0772.2017.10b.15 (in Chinese).

44.

Wang

Zhou

Xia

Sun

C. H.

L. J.

Wang

J. L.

Akemi

(2012). Parent-reported health care expenditures associated with autism spectrum disorders in Heilongjiang province, China. BMC Health Services Research, 12(1), 7. https://doi.org/10.1186/1472-6963-12-7

45.

Wang

Y. G.

Wang

Y. K.

Huang

B. Y.

X. J.

(2019). Research on the application and influence factors of intervention methods and assistive teaching devices in autism rehabilitation agencies. Disability Research, 2, 50–55. https://doi.org/2095-0810(2019)34-0050-06 (in Chinese).

46.

Wiggins

L. D.

Durkin

Esler

Lee

L. C.

Zahorodny

Rice

Yeargin-Allsopp

Dowling

N. F.

Hall-Lande

Morrier

M. J.

Christensen

Shenouda

Baio

(2020). Disparities in documented diagnoses of autism spectrum disorder based on demographic, individual, and service factors. Autism Research, 13(3), 464–473. https://doi.org/10.1002/aur.2255

47.

World Health Organization. (2011). The global burden of disease.

48.

L. J.

(2018). Status and thoughts on family disease burden of children with autism spectrum disorders. Chinese Journal of School Health, 39(3), 321–324. https://doi.org/10.16835/j.cnki.1000-9817.2018.03.001 (in Chinese).

49.

Xiong

Yang

Hou

Liu

Zhang

Jiao

(2011). Investigation of raising burden of children with autism, physical disability and mental disability in China. Research in Developmental Disabilities, 32(1), 306–311. https://doi.org/10.1016/j.ridd.2010.10.003

50.

Wang

Wimo

Fratiglioni

Qiu

(2017). The economic burden of dementia in China, 1990-2030: Implications for health policy. Bulletin of the World Health Organization, 95(1), 18–26. https://doi.org/10.2471/BLT.15.167726

51.

Evans

Kawabata

Zeramdini

Klavus

Murray

(2003). Household catastrophic health expenditure: A multicountry analysis. Lancet, 362(9378), 111–117.

52.

Zhang

Coyte

P. C.

(2020). Inequality of opportunity in healthcare expenditures: Evidence from China. BMC Health Services Research, 20(1), 379. https://doi.org/10.1186/s12913-020-05252-z

53.

Zhao

Y. J.

Tan

L. C.

S. C.

W. L.

Seah

S. H.

Lau

P. N.

Luo

Wee

H. L.

(2011). Economic burden of Parkinson’s disease in Singapore. European Journal of Neurology, 18(3), 519–526. https://doi.org/10.1111/j.1468-1331.2010.03210.x

54.

Zhou

Yan

Zou

Luo

Huang

Guan

Chen

Mao

Xia

Zhang

Zhou

. . . Wang

(2020). Prevalence of autism spectrum disorder in china: A nationwide multi-center population-based study among children aged 6 to 12 years. Neuroscience Bulletin, 36(9), 961–971. https://doi.org/10.1007/s12264-020-00530-6

55.

Zhou

Wang

Huang

(2022). Brief report: Health expenditures for children with autism and family financial well-being in China. Journal of Autism and Developmental Disorders, 52(8), 3712–3717. https://doi.org/10.1007/s10803-021-05214-2

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB