Changes in spending and service use after a state autism insurance mandate

Abstract

Almost all states have insurance coverage mandates for childhood autism spectrum disorder treatment, yet little is known about how mandates affect spending and service use. We evaluated a 2011 Kansas law mandating comprehensive coverage of autism spectrum disorder treatments in the State Employee Health Plan. Data were extracted from the Kansas All-Payer Claims Database from 2009 to 2013 for enrollees of State Employee Health Plan and private health plans. The sample included children aged 0–18 years with >2 claims with an autism spectrum disorder diagnosis insured through State Employee Health Plan or a comparison group enrolled through private health plans. We estimated differences-in-differences regression models to compare trends among State Employee Health Plan to privately insured children. Average annual total spending on autism spectrum disorder services increased by US$912 (95% confidence interval: US$331–US$1492) and average annual out-of-pocket spending on autism spectrum disorder services increased by US$138 (95% confidence interval: US$53–US$223) among diagnosed children in the State Employee Health Plan relative to the comparison group following the mandate, representing 92% and 75% increases over baseline total and out-of-pocket autism spectrum disorder spending, respectively. Average annual quantity of outpatient autism spectrum disorder services increased by 15.0 services (95% confidence interval: 8.4–21.6) among children in the State Employee Health Plan, more than doubling the baseline average. Implementation of a comprehensive autism spectrum disorder mandate in the Kansas State Employee Health Plan was associated with substantial increases in service use and spending for autism spectrum disorder treatment among autism spectrum disorder–diagnosed children.

Keywords

access to care health insurance insurance mandate out-of-pocket spending utilization

Introduction

About 1 in 68 children in the United States are now identified as having an autism spectrum disorder (ASD), a doubling from the early 2000s (Centers for Disease Control and Prevention (CDC), 2016). Children with ASD may require extensive intervention in educational and clinical settings, including behavioral therapies, medications, and occupational/physical therapy (Myers and Johnson, 2007). These children frequently have physical and psychiatric comorbidities requiring specialty care. The costs associated with ASD are substantial. Conservative estimates suggest that children with ASD account for US$3020 higher mean annual medical costs than non-diagnosed children and US$182 higher out-of-pocket costs (in 2011 dollars; Lavelle et al., 2014). Caregivers of ASD-diagnosed children are more likely to report elevated health spending burden than are caregivers of children with other developmental disabilities (Zablotsky et al., 2014). The total cost of supporting an individual with ASD over his/her lifetime has been estimated to be US$2.4 million (Buescher et al., 2014).

Private health insurance plans have historically limited coverage of ASD treatments (Peele et al., 2002). However, virtually all states have now enacted some requirements requiring insurers to cover ASD treatments (National Conference of State Legislatures, 2012). Research on the impact of these state mandates is limited. A recent study using commercial insurance claims data finds that the diagnosed prevalence of ASD increased in states after the implementation of ASD mandates, potentially indicating greater screening and detection (Mandell et al., 2016). However, this study did not consider changes in spending or service use. Another study finds that a state mandate in Pennsylvania targeting private insurance plans was associated with decreased Medicaid spending, potentially reflecting reduced pressure on public systems due to improved private coverage (Stein et al., 2012). That study did not directly examine changes among privately insured children subject to the mandate. A study using national survey data finds families of ASD-diagnosed children residing in states enacting mandates did not experience improvements in access to care, financial burden, or unmet need compared to families in other states (Chatterji et al., 2015). However, the study data were collected very soon after the mandates were implemented and did not permit the authors to identify whether a child was enrolled in an insurance product subject to the state mandate, which would tend to reduce detection of changes.

In this study, we examine changes in service use and spending among ASD-diagnosed children following the enactment in 2011 of a law requiring the Kansas State Employee Health Plan (SEHP) to cover all services related to diagnosis and treatment of ASD among individuals under the age of 19 years. The provision limited annual spending to US$36,000 for children under the age of 7 years and US$27,000 for those aged 7–18 years (Kansas Legislature, 2010). The Kansas SEHP mandate was a legislative requirement for only 1 year (calendar year 2011), but the SEHP has subsequently continued the policy. The Kansas SEHP is self-insured, meaning it assumes the risk of all claims. During the 5-year study period, households could select from three similar commercial insurance products provided by major national carriers. Prior to 2011, the SEHP plans did not cover applied behavior analysis (ABA) therapy, a treatment regimen often requiring many hours of services per week over several years intended to alter social functioning, but ABA coverage was required after 2011 (Kansas State Employees Health Care Commission, n.d.).

To evaluate changes after this mandate, we draw upon unique data from the Kansas All-Payer Claims Database that includes all ASD-diagnosed children in the SEHP from 2009 to 2013 and a comparison population of ASD-diagnosed, privately insured children during the same time period. Our data allow us to differentiate which children were subject to the mandate and to control for contemporaneous trends using this within-state comparison group. We hypothesized that the mandate would be associated with increased spending on ASD care and greater use of intensive treatment regimens. Although mandates can reduce the proportion of costs families are required to pay through reduced caps on services, we hypothesized that total out-of-pocket spending for insured services would increase due to greater service use.

Data and methods

Data

Through an agreement with the Kansas Department of Health, we obtained all inpatient, outpatient, and prescription drug claims data for ASD-diagnosed children enrolled in the SEHP or a private insurance plan in Kansas during calendar years 2009–2013. The private insurance plans were derived from a database of all private insurance plans operating in Kansas. Because the SEHP plans all follow a preferred provider organization (PPO) model, we likewise limited the comparison sample to privately insured children in PPO plans to increase comparability. Additional sample inclusion criteria for both populations were having at least two claims with an ASD diagnosis (Ninth Revision of International Classification of Diseases (ICD-9): 299.XX), with the ASD diagnosis being the first listed diagnosis in at least one claim in the study year, and age 0–18 years at the beginning of the study year. The unit of analysis was the child–year; children with multiple years of enrollment were only included in the sample for years for which they had claims with ASD diagnosis. For children meeting inclusion criteria in a given year, all inpatient, outpatient, and prescription drug claims for that child–year were included. The John Hopkins School of Public Health Institutional Review Board determined the study to be exempt.

Services used and spending

We aggregated all data to the level of child–year. For each child-year, we examined the total number of services and the number of prescriptions. Unique services were defined as all services provided in a setting (inpatient or outpatient) with the same starting date. Inpatient and outpatient data were combined because the private insurance claims did not reliably differentiate care across setting types. We defined ASD-related services as those services in which the claim carried a primary ASD diagnosis. We also identified prescription use and spending, but could not identify which medications were prescribed for ASD specifically and medication type was only available in the SEHP data.

We calculated total health care spending per child-year among children with a primary ASD diagnosis in that year by summing all spending for services. We also calculated total out-of-pocket payments by summing the total amount paid in copayments, deductibles, and coinsurance. To derive ASD-specific spending, we calculated these same amounts for all claims where ASD was the primary diagnosis. All spending amounts were inflation adjusted to 2013 dollars using the Consumer Price Index.

Covariates

Demographic measures were age and sex. We also examined physical and psychiatric comorbidities based on the presence of a primary diagnosis on a claim other than ASD for children in the sample in each study year. These included common physical comorbidities among children with ASD including asthma (ICD-9: 493.XX), auditory disorders (ICD-9: 380.XX, 389.XX), cardiac disorders (ICD-9: 427.XX-428.XX), gastrointestinal disorders (ICD-9: 555.XX-558.XX), seizures (ICD-9: 345.XX), and developmental delays (ICD-9: 315.XX; Kohane et al., 2012). Psychiatric comorbidities included attention-deficit hyperactivity disorder (ADHD; ICD-9: 314.XX), conduct or emotional disorders (ICD-9: 312.XX-313.XX), depression (ICD-9: 296.XX), anxiety disorders (ICD-9: 300.XX-311.XX), and psychotic disorders (291.XX-298.XX, except 296.XX). We identified four regions of residence based on three-digit zip code of residence: Kansas City, Topeka, and Wichita metro areas and all other regions in the state. Finally, since availability of pediatric care providers may be an important determinant of access to services, we included the ratio of pediatricians and psychiatrists (a proxy for mental health professionals) per 100,000 population for the baseline year, 2009, in each county of residence for children in the sample extracted from the Area Health Resource File.

Statistical analysis

We calculated descriptive statistics to compare ASD-diagnosed child-years in the SEHP to those with private insurance in 2009–2010, comparing differences in means using pairwise t-tests. To examine the effects of the Kansas SEHP ASD mandate on spending and service use outcomes, we compared changes after the mandate became effective on 1 January 2011 among children in our treatment group enrolled in the SEHP to changes among children in our comparison group with private insurance. We used a differences-in-differences (DD) approach, which allows us to isolate changes in these outcomes that are unique to the targeted population, net of other secular trends that may have occurred during the same period.

In this model, we include an indicator for whether the child is in the SEHP versus private, an indicator for whether the calendar year is before (2009–2010) or after (2011–2013) the insurance mandate was implemented, and an interaction term of the treatment group and post-period indicator variables. This interaction term is the main coefficient of interest as it represents the change in SEHP enrollees relative to the privately insured enrollees. We also included in the model a year time trend and child-level variables including age, sex, physical and mental health comorbidities, indicators for region, and the county-level measures of pediatrician and psychiatrist availability.

DD models were estimated using general linear models, which allow us to fit models to the appropriate distribution of each outcome. For spending models, we fit our data to a gamma distribution using a log link function. The exceptions were spending models related to prescription drug spending, for which ordinary least squares regression models achieved better model fit. The quantity of outpatient services and medications filled was modeled using negative binomial regression. Standard errors were clustered to account for repeated child observations. To increase our confidence in our quasi-experimental approach, we compared pre-period trends in outcomes in the 2 years prior to implementation of the ASD mandate among enrollees in the treatment and comparison groups. Based on these tests, we could not reject the null hypothesis that trends were parallel.

Results

Demographic characteristics

Across all years of the study, we identified a total of 343 children under the age of 19 years meeting our study inclusion criteria in the SEHP who contributed a total of 663 child-years of data, and we identified 1429 children in the privately insured sample contributing 1937 child-years.

In the years prior to the mandate, there were 784 child-years in the private insurance program and 212 in the SEHP sample (Table 1). Children in the SEHP and privately insured groups were similar across a variety of dimensions. They had relatively similar proportion of male, mean age, and comorbidity prevalence. SEHP children were significantly more likely to have gastrointestinal disorders. Larger percentages of SEHP children lived in the metro area of Topeka (the state capital), whereas privately insured children were more likely to live near Wichita and Kansas City. Children in the privately insured sample resided in counties that, on average, had almost double the availability of pediatricians and psychiatrists than did children in the SEHP sample.

Table 1.

Characteristics of ASD-diagnosed child-years in the Kansas SEHP and comparison of private insurance sample, 2009–2010.

	Private	SEHP	p value
	N = 784	N = 212
	Mean (N)	Mean (N)
Male	0.81 (635)	0.79 (619)	0.657
Age (SD, years)	10.78 (4.24)	10.18 (4.79)	0.072
Psychiatric comorbidities
Attention-deficit/hyperactivity disorder	0.28 (220)	0.26 (204)	0.688
Conduct or emotional disorder	0.11 (86)	0.14 (110)	0.227
Mood disorder	0.11 (86)	0.14 (110)	0.131
Anxiety disorder	0.14 (110)	0.14 (110)	0.998
Psychotic disorder	0.01 (8)	0.02 (16)	0.608
Physical comorbidities
Asthma	0.04 (31)	0.06 (47)	0.274
Auditory/language	0.08 (63)	0.05 (39)	0.238
Cardiac disorder	0.01 (8)	0.03 (24)	0.155
Gastrointestinal disorder	0.02 (16)	0.05 (39)	0.008
Seizure	0.05 (39)	0.03 (24)	0.144
Developmental delays	0.06 (47)	0.09 (71)	0.192
Region of state
Kansas City	0.33 (259)	0.25 (196)	0.019
Topeka	0.11 (86)	0.34 (267)	<0.001
Wichita	0.23 (180)	0.09 (71)	<0.001
Other Area	0.32 (251)	0.33 (259)	0.864
County physician supply
Pediatricians per 100,000 (SD)	40.89 (47.05)	22.65 (33.84)	<0.001
Psychiatrists per 100,000 (SD)	50.25 (57.33)	27.34 (44.19)	<0.001

SEHP: State Employee Health Plan; ASD: autism spectrum disorder; ICD-9: Ninth Revision of International Classification of Diseases; CI: confidence interval; OOP: out of packet; SD: standard deviation.

Authors’ analysis of data from the Kansas All Payers Claims Database (APCD). The p value from pairwise t-test for difference in means. N is shown in parentheses, except for continuous measures where the SD is shown. Psychiatric and physical comorbidities defined using ICD-9 codes from primary diagnosis fields on claims. ICD-9 codes for comorbidities: asthma (ICD-9: 493.XX), auditory disorders (ICD-9: 380.XX, 389.XX), cardiac disorders (ICD-9: 427.XX–428.XX), gastrointestinal disorders (ICD-9: 555.XX–558.XX), seizures (ICD-9: 345.XX), developmental delays (ICD-9: 315.XX), attention-deficit hyperactivity disorder (ADHD; ICD-9: 314.XX), conduct or emotional disorders (ICD-9: 312.XX–313.XX), depression (ICD-9: 296.XX), anxiety disorders (ICD-9: 300.XX–311.XX), and psychotic disorders (291.XX-298.XX, except 296.XX).

Unadjusted trends in spending and number of services

Mean, unadjusted spending related to ASD increased from US$952 in 2009 to US$2823 in 2013 in the SEHP, while increasing from US$710 to US$1497 among privately insured children (Figure 1). The annual time trends for both groups were statistically significant (p < 0.001). The mean number of services in the SEHP population related to ASD care among diagnosed children rose from 8.9 in 2009 to 35.3 in 2013, with a dramatic increase occurring in 2011, while the mean rose from 5.5 to 6.4 among privately insured children (Figure 2). Only the trend for the SEHP was statistically significant over time. Within the SEHP, children who were in the 75th percentile of the service use distribution in the year experienced much greater change during the study period, while those at the 50th or 25th percentile of the service use distribution experienced little or no change (Figure 3).

Figure 1.

Trends in mean spending on ASD services comparing SEHP and privately insured children.

Figure 2.

Trends in mean number of claims for ASD treatment among SEHP and privately insured children with diagnosed ASD.

Figure 3.

Trends in number of ASD services at the 25th, 50th, and 75th percentiles in SEHP.

Spending and service use on ASD treatments

Differences-in-differences estimates shown in Table 2 indicate that while average annual spending on ASD services increased in both the SEHP and privately insured populations after the mandate, the increase was significantly greater in the SEHP (difference-in-differences estimate, US$912 (95% confidence interval (CI): US$331–US$1492)). This represents a 92% increase relative to the baseline mean for the SEHP (i.e. US$912/US$994). Mean annual out-of-pocket spending paid to insurers for these visits also significantly increased among SEHP children relative to privately insured (US$138 (95% CI: US$53–US$223)), a 75% increase over the SEHP baseline mean for out-of-pocket spending. The proportion of ASD spending that was out-of-pocket prior to the mandate was about 18% (i.e. US$183/US$994), and this remained about the same (17%) after the mandate. The average annual number of services per child with a primary ASD diagnosis increased in the SEHP, while remaining virtually unchanged among privately insured children. The change in the SEHP relative to the change among privately insured children was 15.0 services (95% CI: 8.4–21.6). This represents a more than doubling relative to the baseline mean of 8.9 visits in the SEHP.

Table 2.

Adjusted estimates of the effects of the Kansas SEHP autism mandate on spending and service use among children with ASD.

	Group	Predicted means		Est.	p value	95% CI
	Group	2009–2010	2011–2013	Est.	p value	95% CI
Annual ASD service use and spending
ASD services spending	SEHP	US$994	US$2116	US$912	0.002	(US$331, US$1492)
ASD services spending	Private	US$899	US$1109
ASD OOP services spending	SEHP	US$183	US$361	US$138	0.001	(US$53, US$223)
ASD OOP services spending	Private	US$261	US$301
Services for ASD treatment	SEHP	8.9	23.9	15.0	<0.001	(8.4, 21.6)
Services for ASD treatment	Private	6.0	6.0
Annual non-ASD service use and spending
Non-ASD services spending	SEHP	US$4122	US$3095	−US$663	0.500	(−US$2591, US$1265)
Non-ASD services spending	Private	US$6141	US$5778
Non-ASD services OOP	SEHP	US$538	US$552	US$71	0.440	(−US$110, US$252)
Non-ASD services OOP	Private	US$1797	US$1623
Prescription spending (all drugs)	SEHP	US$3020	US$1976	−US$622	0.224	(−US$1591, US$346)
Prescription spending (all drugs)	Private	US$1941	US$1651
Prescription OOP	SEHP	US$740	US$533	−US$170	0.047	(−US$338, −US$2)
Prescription OOP	Private	US$333	US$313
Non-ASD services	SEHP	34.15	32.39	0.05	0.992	(−9.83, 9.93)
Non-ASD services	Private	21.82	20.01
Non-ASD behavioral services	SEHP	35.5	35.2	1.9	0.707	(−8, 11.8)
Non-ASD behavioral services	Private	25.6	23.4
Non-ASD non-behavioral services	SEHP	13.3	13.7	−4.1	0.358	(−12.9, 4.7)
Non-ASD non-behavioral services	Private	18.5	23.0
Drug fills (All drugs)	SEHP	29.9	29.3	1.8	0.687	(−6.9, 10.5)
Drug fills (All drugs)	Private	21.5	19.1

SEHP: State Employee Health Plan; ASD: autism spectrum disorder; CI: confidence interval; OOP: out of packet.

Each outcome is derived from a separate differences-in-differences regression model. Main estimate represents the interaction of being in the SEHP and observed after the 2011 policy change. Predicted means represent regression-adjusted predicted margins for each group. Models include age, sex, physical and mental health comorbidities, county-level ratio of pediatricians and psychiatrists, and region covariates. Sample sizes are 343 children in the SEHP who contributed a total of 663 child-years of data and 1429 children in the privately insured sample contributing 1937 child-years.

Spending and service use on non-ASD treatments and prescription drugs

We investigated potential changes in spending and service use on non-ASD treatment and on prescription drugs, as these categories of service use could have been indirectly affected by the mandate (Table 2). We considered all non-ASD services (i.e. services where the primary diagnosis code was not ASD) and further subdivided this into non-ASD services for other behavioral health diagnoses and non-ASD, non-behavioral health diagnoses. Prescription drugs included claims and spending for all therapeutic categories.

There were notable differences in the level of service use and mean annual spending in the SEHP relative to the privately insured population in the pre-mandate period: non-ASD spending was higher in the private insurance group than in the SEHP group prior to the mandate. Our models indicate large, but not statistically significant, reductions in mean annual spending on non-ASD services and prescription drugs. There were no significant changes in utilization of non-ASD services, or in the number of non-ASD services disaggregated by either behavioral or non-behavioral. We did, however, observe a significant decrease in mean prescription drug out-of-pocket spending (−US$170 (95% CI: −US$338 to −US$2)). There was no corresponding change in the total number of fills for prescription drugs, suggesting the out-of-pocket spending change was likely driven by changes in price of drugs per fill rather than quantity consumed.

Discussion

We examined the impact of a 2011 mandate that required the Kansas SEHP to provide comprehensive coverage for the treatment of ASD. Comparing SEHP-enrolled children with an ASD diagnosis to within-state counterparts in private insurance plans not subject to a mandate during this period, we find that ASD-related spending rose in both groups over time, but that the increase was more marked among the SEHP-enrolled populations. Spending increases outside of the SEHP may reflect broader national trends. For example, research indicates that the Mental Health Parity and Addiction Equity Act implemented in 2010 led to a national increase in spending related to ASD treatment over this same time period (Stuart et al., 2017). Importantly, our study design allows us to isolate the changes in outcomes among the SEHP group due to the mandates separate from other secular trends in service use and spending due to the parity law or other factors.

The changes in service use observed in our study are large relative to the baseline levels of spending and service use observed in our sample and may indicate that children with diagnosed ASD were better able to access clinically appropriate services after the mandate. For example, prior literature finds that ASD-diagnosed children in private insurance plans receive about five visits per year for key services such as occupational and physical therapy, speech therapy, and behavioral modification (and these rates are much lower than for Medicaid-enrolled youth with ASD; Wang et al., 2013). Thus, an increase of 15 visits with an ASD diagnosis could enable more contact with key providers in these areas. Similarly, a RAND report using 2013 commercial claims data found that mean reimbursement rates for ABA varied from US$25 to US$160 per hour for a bachelor’s level provider and US$37 to US$197 per hour for a masters- or doctoral-level provider (Maglione et al., 2016). Our estimates thus imply that average spending increases might mean a child could receive anywhere from 6 to 37 additional hours of ABA services from a bachelor’s level provider or 5 to 25 additional hours of ABA from master-trained clinician over the course of a year. The mean amount charged per new visit in our sample was US$61 (i.e. US$912/15 visits), which appears to fall in the middle of these estimates.

It is unclear what factors drive increased utilization. One likely possibility is that increasing the spending cap led families with children who were previously constrained to use more services. Another possibility is that some of the increase in spending and service use with an ASD diagnosis code may be related to reclassification of services that were already being provided to children prior to the mandate. For example, a child previously receiving counseling with an intellectual disability diagnosis may be classified with an ASD diagnosis after the mandate in order to take advantage of better coverage. We found large, but not statistically significant, reductions in non-ASD spending and prescription drug spending after the mandate. Finally, better access to services for ASD-diagnosed youth may reduce spending in other areas due to improved medical management (e.g. reduced visits to the emergency department). Identifying the contribution of these factors is important for further research.

Effects of the mandate in Kansas might be stronger than in other states because the Kansas mandate applied to any ASD-related treatments and not just to a specific treatment modality such as applied behavioral analysis. The largest increases in spending occurred among the highest utilizing children, consistent with the hypothesis that families of these children might have previously been constrained by service limits. However, the Kansas mandate also was weaker than mandates in other states such as California, Indiana, Massachusetts, and Minnesota in that it still included a cap on the maximum amount of spending per year (Fifield, 2016). One of the key limitations of this study is that it relies on data from a single state, with relatively small sample sizes. These limitations reduce the generalizability of the study and also the ability to detect potentially important changes.

This study has some other important limitations. First, as noted, we cannot determine whether the observed increase in ASD-related service use was mainly due to the initiation of new treatments versus the shifting of cost of treatments that had previously been paid for entirely outside of health insurance from households to the SEHP. Cost-shifting may occur because many families pay for treatments entirely out-of-pocket when insurance will not authorize the treatments or because the treatments take place in settings that do not seek reimbursement from insurance (e.g. school-based programs). Moreover, we do not have access to payments that are made by households outside of insurance plans. These payments could decrease, lowering total out-of-pocket spending. Understanding these mechanisms can help identify whether autism mandates alleviate financial burden and improve access to care. Second, while claims data provide a detailed record of spending and utilization, limitations in the private insurance data do not allow us to reliably evaluate the setting where treatment took place. The private insurance data also do not provide information on the therapeutic class of medications prescribed, meaning that we cannot examine trends in the medications most commonly prescribed for ASD (e.g. stimulants and antipsychotic drugs). Third, the data also do not allow us to infer whether the quality of treatments improved after the mandate or whether treatment promoted greater functioning among diagnosed children. Fourth, because procedure codes were only available for SEHP-enrolled children, we could not test whether the increase in the use of specific procedures, such as daily counseling, was unique to this population or occurred to some extent among privately insured children outside of the SEHP. Fifth, because our dataset only includes children who had at least one claim with a primary diagnosis of ASD in each year, we cannot track children in years when they do not have claims with an ASD diagnosis, even though such children might continue to receive ASD services. Sixth, our data did not provide specific months of enrollment, and so, we cannot exclude children who were enrolled in plans for only short periods of time. Accordingly, we may underestimate the changes in expenditures for children who are enrolled for the full year. Finally, although we applied identical procedures to the SEHP and private insurance databases, differences in data collection and processing methods by each entity could lead to discrepancies between the two populations.

Conclusion

The Kansas SEHP experience demonstrates that coverage mandates can be effective in increasing receipt of services among youth with diagnosed ASD. With a growing number of states implementing these mandates, experiences may differ depending upon factors such as the scope and enforcement of the mandate, the response of insurance companies to new requirements, and the level of awareness among physicians, patients, and parents of children with ASD. As was found in Kansas, new mandates may create opportunities for children to take advantage of a broader array of services. Whether these policies lead to better quality of care and improved outcomes among these children will be an important benchmark for future evaluation.

Supplemental Material

AUT728205_Lay_Abstract – Supplemental material for Changes in spending and service use after a state autism insurance mandate

Supplemental material, AUT728205_Lay_Abstract for Changes in spending and service use after a state autism insurance mandate by Brendan Saloner and Colleen L Barry in Autism

Footnotes

Acknowledgements

The authors gratefully acknowledge assistance from the Kansas Division of Health Care Finance. All views represented in this paper represent only those of the authors. Dr Barry gratefully acknowledges funding to support her work on this study from the NIMH (R01MH096848).

Funding

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

References

Buescher

AVS

Cidav

Knapp

et al . (2014) Costs of autism spectrum disorders in the United Kingdom and the United States. JAMA Pediatrics 168(8): 721–728.

Centers for Disease Control and Prevention (CDC) (2016) Facts about ASD. Available at: http://www.cdc.gov/ncbddd/autism/facts.html (accessed 9 April 2016).

Chatterji

Decker

Markowitz

(2015) The effects of mandated health insurance benefits for autism on out-of-pocket costs and access to treatment. Journal of Policy Analysis and Management 34(2): 328–353.

Fifield

(2016) Coverage for Autism Treatment Varies by State. Washington, DC: Pew Charitable Trusts.

Kansas Legislature (2010) State Employee Benefit Programs: Autism Coverage; Time Limit. Available at: http://www.kslegislature.org/li/b2017_18/statute/075_000_0000_chapter/075_065_0000_article/075_065_0024_section/075_065_0024_k/ (accessed 2 November 2017).

Kansas State Employees Health Care Commission (n.d.) Report on Insurance Coverage for Autism Spectrum Disorder Pilot. Topeka, KS: Kansas State Employees Health Care Commission.

Kohane

McMurry

Weber

et al . (2012) The co-morbidity burden of children and young adults with autism spectrum disorders. PLoS ONE 7(4). Available at: https://iths.pure.elsevier.com/en/publications/the-co-morbidity-burden-of-children-and-young-adults-with-autism- (accessed 19 February 2017).

Lavelle

Weinstein

Newhouse

et al . (2014) Economic burden of childhood autism spectrum disorders. Pediatrics 133: e520–e529.

Maglione

Kadiyala

Kress

et al . (2016) Applied Behavior Analysis (ABA) Benefit: Comparison with Medicaid and Commercial Benefits. Santa Monica, CA: RAND. Available at: http://www.rand.org/content/dam/rand/pubs/research_reports/RR1300/RR1334/RAND_RR1334.pdf

10.

Mandell

Barry

Marcus

et al . (2016) Effects of autism spectrum disorder insurance mandates on the treated prevalence of autism spectrum disorder. JAMA Pediatrics 170: 887–893.

11.

Myers

Johnson

(2007) Management of children with autism spectrum disorders. Pediatrics 120(5): 1162–1182.

12.

National Conference of State Legislatures (2012) Insurance coverage for autism. Available at: http://www.ncsl.org/research/health/autism-and-insurance-coverage-state-laws.aspx (accessed 1 January 2015).

13.

Peele

Lave

Kelleher

(2002) Exclusions and limitations in children’s behavioral health care coverage. Psychiatric Services. Available at: http://ps.psychiatryonline.org/doi/abs/10.1176/appi.ps.53.5.591 (accessed 9 July 2017).

14.

Stein

Sorbero

Goswami

et al . (2012) Impact of a private health insurance mandate on public sector autism service use in Pennsylvania. Journal of the American Academy of Child and Adolescent Psychiatry 51(8): 771–779.

15.

Stuart

McGinty

Kalb

et al . (2017) Increased service use among children with autism spectrum disorder associated with mental health parity law. Health Affairs 36(2): 337–345.

16.

Wang

Mandell

Lawer

et al . (2013) Healthcare service use and costs for autism spectrum disorder: a comparison between Medicaid and private insurance. Journal of Autism and Developmental Disorders 43(5): 1057–1064.

17.

Zablotsky

Kalb

Freedman

et al . (2014) Health care experiences and perceived financial impact among families of children with an autism spectrum disorder. Psychiatric Services 65(3): 395–398.

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

0.11 MB