Healthcare service utilization and cost among transition-age youth with autism spectrum disorder and other special healthcare needs

Abstract

Youth with autism spectrum disorder often have complex medical needs. Disruptions of healthcare during the transition from pediatric to adult healthcare may put youth with autism spectrum disorder at higher risk of medical emergencies and high medical costs. We conducted a study among transition-age youth (14–25 years old) receiving healthcare at Kaiser Permanente Northern California during 2014–2015. We examined the differences in healthcare utilization and costs among youth with autism spectrum disorder (n = 4123), attention deficit and hyperactivity disorder (n = 20,6015), diabetes mellitus (n = 2156), and general population controls (n = 20,615). Analyses were also stratified by age and sex. Youth with autism spectrum disorder had the highest utilization of outpatient primary care, mental health, and psychotropic medications and the lowest utilization of obstetrics/gynecology and urgent care. Costs for youth with autism spectrum disorder were higher than those for attention deficit and hyperactivity disorder and general population peers and lower than for diabetes mellitus. Utilization patterns varied by age. Transition-age youth with autism spectrum disorder generally utilize healthcare at higher rates relative to attention deficit and hyperactivity disorder and general population peers but at similar or lower rates than diabetes mellitus peers, indicating this group’s complex combination of psychiatric and medical healthcare needs. The relatively high utilization of psychiatric services and low utilization of women’s health services in transition-age youth with autism spectrum disorder may have implications for long-term health and warrants additional research.

Lay abstract

Youth with autism spectrum disorder often have complex medical needs. Disruptions of healthcare during the transition from pediatric to adult healthcare may put youth with autism spectrum disorder at higher risk of medical emergencies and high medical costs. To understand healthcare utilization during the transition years, we conducted a study among transition-age youth (14–25 years old) receiving healthcare at Kaiser Permanente Northern California during 2014–2015. We examined differences in healthcare utilization and costs among youth with autism spectrum disorder (n = 4123), attention deficit and hyperactivity disorder (n = 20,6015), diabetes mellitus (n = 2156), and general population controls (n = 20,615). Analyses were also stratified by age and sex. Youth with autism spectrum disorder had the highest utilization of outpatient primary care, mental health, and psychotropic medications and the lowest utilization of obstetrics/gynecology and urgent care. Costs for youth with autism spectrum disorder were higher than those for attention deficit and hyperactivity disorder and general population peers and lower than for diabetes mellitus. Healthcare utilization patterns varied by age. Transition-age youth with autism spectrum disorder generally used healthcare at higher rates relative to attention deficit and hyperactivity disorder and general population peers but at similar or lower rates than diabetes mellitus peers, indicating this group’s complex combination of psychiatric and medical healthcare needs. The relatively high utilization of psychiatric services and low utilization of women’s health services in transition-age youth with autism spectrum disorder may have implications for long-term health and warrants additional research.

Keywords

autism spectrum disorders health services transition

Introduction

Autism spectrum disorder (ASD) is characterized by difficulties with social communication, sensory processing, and behavior and affects approximately 1%–2% of children in the United States (Baio et al., 2018). Individuals with ASD also experience high rates of co-occurring medical and psychiatric conditions throughout the lifespan (Bauman, 2010; Croen et al., 2015; Davignon et al., 2018). The transition to adulthood is a vulnerable period for adolescents with ASD and their families as numerous developmental and societal changes, including loss of educational and developmental support services, coincide (Roux et al., 2015). Strain on youth and their families is further compounded by evolving medical needs that often go unmet when the adolescent’s care team shifts from pediatrics to adult medicine (Cadman et al., 2012; Cidav et al., 2013; Lounds et al., 2007).

Healthcare utilization and expenditures are generally greater for children and adults with ASD than for individuals with typical development or with other neurodevelopmental conditions such as attention deficit and hyperactivity disorder (ADHD; Croen et al., 2006; Mandell et al., 2006; Shimabukuro et al., 2008; Weiss et al., 2018; Zerbo, Qian et al., 2018). Previous work suggests that the transition of adolescents with ASD is marked by changes in patterns of healthcare utilization including a shift in expenditures away from outpatient care and toward prescription medications (Shimabukuro et al., 2008). In addition, utilization of emergency departments (EDs) often increases among young adults with ASD and other special healthcare needs (SHCN; Liu et al., 2017; Lunsky et al., 2015), perhaps reflecting unstable insurance coverage and inadequate access to primary and preventive care during the transition years (Okumura et al., 2007).

The quality and consistency of healthcare interactions during adolescence and early adulthood can have long-term impacts on an individual’s engagement with the health system during adulthood (Cheak-Zamora et al., 2013; Crowley et al., 2011). However, little is known about how healthcare service utilization patterns among youth with ASD compare with youth with other chronic behavioral or medical conditions, and how these patterns differ across the transition period. Using comprehensive data from a large, integrated healthcare delivery system in California, we examined healthcare utilization and costs among transition-age youth with ASD in comparison to age-matched peers with other SHCN and to peers without these conditions.

Methods

Study population

The study sample was drawn from Kaiser Permanente Northern California (KPNC), a large integrated healthcare system serving over 4.3 million members. The sociodemographic distribution of the KPNC membership is broadly similar to the local and state-wide California population, though the extremes of the income distribution are underrepresented (Gordon, 2015). Insurance coverage is predominantly private payer (90% paid through employer groups or individual plans) with a smaller fraction (10%) of member plans paid by public funds (e.g. Medicaid; Davignon et al., 2018). Eligible individuals were members of KPNC for at least 9 months in each calendar year from 2013 to 2015 and were aged 14–25 as of 1 January 2014, the start of the study period. The ASD case group included all individuals with an ASD diagnosis documented in their medical record before 2014 (n = 4123). Of these cases, 34% were diagnosed following a comprehensive clinical assessment including the Autism Diagnostic Observation (ADOS) by specialists at a KPNC ASD assessment center. The remaining 66% had an ASD diagnosis recorded in their medical record on at least two separate occasions by non-specialist providers.

For comparison, we chose two groups with either behavioral or medical SHCN and one group from the general population (GP). These groups were sampled from individuals who did not have an ASD diagnosis by the end of 2015. For the behavioral comparison group, we randomly sampled individuals diagnosed with ADHD (n = 20,615). For the medical comparison group, we included all individuals with diabetes mellitus (DM; n = 2156) identified in the KPNC diabetes registry (Karter et al., 2002). Individuals sampled in the ADHD and DM groups were not mutually exclusive and approximately 75 individuals overlapped across these groups as described in Davignon et al. (2018). Individuals with either of these conditions and an ASD diagnosis were included in the ASD group only. A third comparison group from the GP of patients included a random sample of individuals without ASD, ADHD, or DM diagnoses by the end of 2015 (n = 20,516). The ADHD and GP groups were matched in a 5:1 ratio to ASD cases on sex and age group (ages 14–17, 18–21, and 22–25).

KPNC’s Institutional Review Board (IRB) approved all study procedures.

Healthcare utilization

Healthcare utilization data were obtained from KPNC’s inpatient, outpatient, laboratory, and pharmacy databases for the period of January 2014–December 2015. Following a similar procedure used in earlier work (Zerbo, Qian et al., 2018), we examined utilization in terms of visits to primary care (i.e. to the patient’s primary care provider or other provider in pediatric or internal/family medicine) and specialty care (to a provider in mental health, neurology, obstetrics/gynecology (OB/GYN), speech therapy, and physical/occupational therapy) that occurred at KPNC facilities and authorized providers outside of KPNC; visits to KPNC or non-KPNC EDs; hospitalizations including inpatient (psychiatric and non-psychiatric) and same-day hospitalizations at KPNC and non-KPNC facilities; and visits to radiology or laboratory departments. Utilization of preventive services included well visits; KPNC gender- and age-specific recommendations for vaccinations (i.e. flu and meningococcal vaccines); and metabolic screening including receipt of fasting lipid profile, fasting glucose, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) tests among youth with a body mass index (BMI) > 95%. In addition to visits for any reason to an OB/GYN provider, we also examined women’s health encounters specifically for pelvic exams and intrauterine device (IUD) insertion/removal. Utilization measures were coded as dichotomous (yes/no) and as mean and median counts of annual utilization within each category. We also examined utilization of prescribed medications defined as psychotherapeutics (antidepressants, antipsychotics, ADHD medications, other), anticonvulsants, analgesics/anti-rheumatic agents, respiratory/allergy medications, diabetes medications, cardiovascular medications, lipid/cholesterol medications, and hypertension medications.

Costs

We obtained the costs of healthcare provided directly by KPNC from the Cost Management Information System, an automated database of hospital, laboratory, radiology, outpatient, and home health utilization integrated with KPNC’s financial ledger. Costs, which include program and facility overhead, are generated for each health service defined by the system using standard accounting methods and program-specific relative value units for each service. The system also includes costs for KPNC-approved services provided by non-KPNC vendors (hospitals and individual providers). However, non-KPNC hospitalizations resulting from an ED visit could not be distinguished, so these visits were categorized as hospital rather than ED costs. Pharmacy costs were retrieved from the KPNC Pharmacy Information Management System, an automated clinical database that documents each prescription dispensed at any outpatient KPNC pharmacy across the region. Dental costs that are not provided by KPNC and patient out-of-pocket expenses that are not stored in these financial databases were excluded from the analyses.

Statistical analysis

Multivariable logistic regression was used to examine differences between dichotomous measures of healthcare utilization in youth with ASD and peers with ADHD, DM, and GP controls. For counts of the mean number of outpatient and inpatient visits, we used multivariable negative binomial regression models to estimate rate ratios accompanied by bootstrapping to construct the confidence intervals. All models were adjusted for covariates available in the KPNC electronic medical record (EMR). These included sex (male/female), age as of January 2014, race/ethnicity (White non-Hispanic, White Hispanic, Black, Asian, and other), membership length (total months from 2013 to 2015), and indicators (yes/no) of co-occurring medical conditions of allergies, asthma, autoimmune conditions, gastrointestinal conditions, mental health, injury, infection, metabolic conditions, musculoskeletal conditions, neurologic conditions, nutritional conditions, and BMI. Models comparing ASD with diabetes cases were further adjusted for indicators of co-occurring cardiovascular, genitourinary, and endocrine disorders.

For cost data, we calculated the mean and median total annual healthcare costs per member as well as costs by specific types of healthcare utilization. We estimated the mean ratio of costs between youth with ASD compared with youth with ADHD, youth with DM, and GP controls using generalized linear models specified with the log-link function and a gamma distribution.

We also analyzed results for utilization and costs stratified by age (14–17, 18–21, and 22–25) and sex.

Results

The demographic characteristics of the study population have been described previously (Davignon et al., 2018). Briefly, the mean (SD) age of youth with ASD was 18.4 (3.2) years and the male:female ratio was 4:1. Youth with ASD were more likely to be White, non-Hispanic (54%) than the DM (46%) and GP (36%) groups and the most likely of all three comparison groups to have Medicaid (20% vs 8%–13%).

Outpatient and inpatient healthcare service utilization

Compared with all three comparison groups, youth with ASD had the highest utilization of outpatient visits to mental health (37%) and neurology (3%) (Table 1). Youth with ASD’s utilization of outpatient visits to primary care (79.7%) and other specialties (35.9%) was similar to peers with ADHD but significantly greater than among GP and significantly lower than among peers with DM (Figure 1 and Table 1). Youth with ASD utilized OB/GYN services nearly half as frequently as peers in comparison groups. Youth with ASD were also significantly less likely than youth in all comparison groups to utilize urgent care and radiology services (Figure 1 and Table 1).

Table 1.

Healthcare utilization among transition-age youth (14–25 years old) with special healthcare needs and the general population, Kaiser Permanente Northern California, 2014–2015.

	ASD	ADHD	DM	GP	ASD versus ADHD	ASD versus DM	ASD versus GP
	(N = 4123)	(N = 20,615)	(N = 2156)	(N = 20615)	ASD versus ADHD	ASD versus DM	ASD versus GP
	%	%	%	%	Adjusted OR (95% CI)^a
Any outpatient clinic visits	83.00	79.35	90.63	68.83	1.36 (1.24–1.49)	0.63 (0.53–0.76)	2.07 (1.89–2.26)
Primary care	69.51	68.19	85.30	60.21	1.09 (1.01–1.18)	0.42 (0.36–0.48)	1.38 (1.27–1.49)
Pediatric	33.83	34.49	31.45	30.06	0.90 (0.81–1.00)	0.33 (0.27–0.41)	1.21 (1.10–1.35)
General medicine	41.06	38.83	63.73	33.81	1.16 (1.07–1.26)	0.61 (0.53–0.70)	1.33 (1.23–1.44)
Urgent care	3.10	4.98	5.24	3.70	0.69 (0.57–0.83)	0.59 (0.44–0.79)	0.84 (0.69–1.03)
Mental health	37.45	27.16	14.29	5.36	1.73 (1.60–1.86)	4.65 (3.96–5.45)	10.87 (9.90–11.94)
Neurology	2.98	1.08	1.39	0.52	2.59 (2.03–3.32)	3.49 (2.16–5.65)	4.13 (3.10–5.50)
OB/GYN	19.45	34.35	34.44	26.47	0.43 (0.35–0.53)	0.66 (0.52–0.84)	0.59 (0.48–0.72)
Other specialties	35.87	37.32	54.22	30.91	1.05 (0.97–1.13)	0.58 (0.51–0.65)	1.23 (1.15–1.33)
Emergency department	14.04	22.39	30.61	14.20	0.66 (0.60–0.73)	0.50 (0.43–0.58)	0.91 (0.82–1.01)
Hospitalizations	4.63	3.48	11.13	1.86	1.81 (1.52–2.16)	0.66 (0.52–0.83)	2.45 (2.01–2.97)
Hospitalization in KPNC
Inpatient hospitalization	1.46	1.38	8.91	1.05	1.33 (0.99–1.78)	0.26 (0.19–0.36)	1.21 (0.89–1.64)
Outpatient hospitalizations (same day)	2.13	1.23	4.78	0.88	1.96 (1.51–2.54)	0.73 (0.52–1.03)	2.11 (1.60–2.78)
Hospitalization outside KPNC	1.72	1.74	1.86	0.60	1.57 (1.19–2.07)	1.46 (0.90–2.36)	3.15 (2.27–4.36)
Any inpatient visit	3.03	2.89	10.20	1.55	1.49 (1.21–1.83)	0.47 (0.36–0.61)	1.91 (1.52–2.40)
Psychiatric inpatient	1.26	1.32	0.88	0.36	1.66 (1.20–2.29)	2.28 (1.20–4.34)	4.64 (3.07–7.03)
Non-psychiatric inpatient	1.99	1.69	9.51	1.22	1.49 (1.16–1.93)	0.34 (0.25–0.45)	1.42 (1.09–1.86)
Physical/occupational therapy	2.57	3.63	3.53	3.29	0.89 (0.71–1.10)	0.91 (0.64–1.28)	0.77 (0.62–0.96)
Speech therapy	2.62	0.05	0	0.04	–^b	–^b	70.87 (34.29–146.46)
Radiology	14.21	21.25	23.79	16.66	0.70 (0.63–0.77)	0.68 (0.58–0.79)	0.79 (0.71–0.87)
Laboratory	39.32	38.54	82.33	29.77	1.10 (1.03–1.19)	0.18 (0.16–0.21)	1.45 (1.34–1.56)

ASD: autism spectrum disorder; ADHD: attention deficit and hyperactivity disorder; DM: diabetes mellitus; GP: general population; OR: odds ratio; CI: confidence interval; OB/GYN: obstetrics/gynecology; KPNC: Kaiser Permanente Northern California; BMI: body mass index.

Models were adjusted with sex, age, race, membership length, BMI, and medical and psychiatric co-occurring conditions (allergies, asthma, autoimmune conditions, gastrointestinal disorders, dementia, injury, infection, metabolic disorders, musculoskeletal disorders, neurological disorders, nutritional conditions) identified using the diagnostic codes assigned to primary and secondary reasons for visit. Models additionally adjusted for cardiovascular conditions, genitourinary conditions, and endocrine conditions in models with DM youth as the reference group.

Models that did not converge due to small cell sizes are denoted with “–..”

Figure 1.

Healthcare utilization in 14–25-year-old individuals with ASD, ADHD, diabetes mellitus, and general population controls, Kaiser Permanente Northern California, 2014–2015.

ED utilization was similar among youth with ASD and GP controls (14%) but significantly lower than among youth with ADHD (22%) and DM (31%) (Figure 1 and Table 1). Hospitalizations among youth with ASD (4.6%) were significantly more common than among youth with ADHD (3.5%) and GP controls (1.9%), particularly for same-day hospitalizations and hospitalizations outside of KPNC (Table 1). Youth with ASD were also 2 to 5 times more likely than all three comparison groups to have a psychiatric inpatient visit and more likely than youth with ADHD (adj-OR:1.49, 95% CI: 1.16–1.93) and GP youth (adj-OR: 1.42, 95% CI: 1.09–1.86) to have an inpatient visit for non-psychiatric reasons. Laboratory use was significantly lower in youth with ASD compared with DM (adj-OR: 0.18, 95% CI: 0.16–0.21) but higher than GP controls (adj-OR: 1.45, 95% CI: 1.34–1.56) (Table 1).

After adjustment for covariates, youth with ASD had a higher mean number of visits to primary care, mental health, and neurology and a lower mean number of visits to OB/GYN than all three comparison groups (Table 2). The mean number of inpatient psychiatric visits was also higher among youth with ASD relative to the DM and GP groups, but slightly lower than among youth with ADHD. Youth with ASD had a similar mean number of visits to urgent care and the ED as GP controls but lower numbers of visits to these services than both ADHD and DM youth.

Table 2.

Mean and median number of outpatient and inpatient visits per member among transition-age youth (14–25 years old) with special healthcare needs and the general population, Kaiser Permanente Northern California, 2014–2015.

	ASD	ADHD	DM	GP	ASD versus ADHD	ASD versus DM	ASD versus GP
	(N = 4123)	(N = 20,615)	(N = 2156)	(N = 20,615)	ASD versus ADHD	ASD versus DM	ASD versus GP
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Adj mean ratio (p)^a
Any outpatient clinic visits	10.90 (28.78)	5.04 (11.64)	7.57 (11.64)	2.82 (5.64)	2.16 (<0.0001)	1.44 (0.0036)	3.87 (<0.0001)
Primary care	1.80 (2.44)	1.79 (2.30)	3.82 (4.37)	1.37 (1.87)	1.01 (0.0984)	0.47 (<0.0001)	1.32 (<0.0001)
Pediatric	0.82 (1.80)	0.89 (1.79)	1.60 (3.42)	0.66 (1.44)	0.93 (0.0008)	0.51 (<0.0001)	1.24 (<0.0001)
General medicine	0.98 (1.94)	0.90 (1.79)	2.21 (3.39)	0.71 (1.43)	1.08 (<0.0001)	0.44 (<0.0001)	1.38 (<0.0001)
Urgent care	0.04 (0.27)	0.07 (0.35)	0.07 (0.35)	0.05 (0.27)	0.63 (0.0001)	0.58 (0.0007)	0.9 (0.1168)
Mental health	1.94 (5.14)	1.19 (4.18)	0.77 (3.68)	0.27 (2.12)	1.63 (<0.0001)	2.53 (<0.0001)	7.1 (<0.0001)
Neurology	0.04 (0.25)	0.01 (0.15)	0.02 (0.21)	0.01 (0.12)	2.78 (<0.0001)	1.79 (<0.0001)	5.44 (<0.0001)
OB/GYN	0.41 (1.24)	1.02 (2.63)	1.50 (4.72)	0.76 (2.33)	0.40 (<0.0001)	0.27 (0.0001)	0.54 (<0.0001)
Other specialties	1.08 (4.76)	1.73 (9.73)	1.96 (7.62)	0.97 (4.08)	0.62 (0.5470)	0.55 (<0.0001)	1.11 (<0.0001)
Emergency department	0.23 (0.82)	0.37 (1.22)	0.66 (1.64)	0.20 (0.62)	0.61 (<0.0001)	0.35 (<0.0001)	1.14 (0.7109)
Hospitalizations	0.07 (0.44)	0.07 (0.52)	0.26 (1.00)	0.04 (0.36)	1.09 (<0.0001)	0.28 (<0.0001)	2 (<0.0001)
Hospitalization in KPNC
Inpatient hospitalization	0.02 (0.15)	0.02 (0.16)	0.14 (0.55)	0.01 (0.13)	1.00 (0.0636)	0.12 (<0.0001)	1.4 (0.2372)
Outpatient hospitalizations (same-day)	0.03 (0.23)	0.02 (0.27)	0.10 (0.62)	0.02 (0.22)	1.25 (<0.0001)	0.28 (0.0545)	1.76 (<0.0001)
Hospitalization outside KPNC	0.03 (0.29)	0.03 (0.32)	0.03 (0.22)	0.01 (0.16)	1.02 (0.0010)	1.13 (0.1344)	3.22 (<0.0001)
Any inpatient visit	0.05 (0.34)	0.05 (0.38)	0.16 (0.61)	0.02 (0.22)	1.01 (0.0002)	0.29 (<0.0001)	2.18 (<0.0001)
Psychiatric inpatient	0.02 (0.20)	0.02 (0.23)	0.01 (0.15)	0.01 (0.10)	0.92 (0.0027)	1.51 (0.0096)	3.57 (<0.0001)
Non-psychiatric inpatient	0.03 (0.25)	0.03 (0.29)	0.15 (0.59)	0.02 (0.19)	1.08 (0.0024)	0.19 (<0.0001)	1.74 (0.0207)
Physical/Occupational therapy	0.15 (2.97)	0.11 (0.91)	0.10 (0.85)	0.10 (0.91)	1.42 (0.2641)	1.54 (0.5740)	1.47 (0.0161)
Speech therapy	0.77 (6.20)	0.01 (0.33)	0.00 (0.00)	0.00 (0.33)	147.87 (<0.0001)	–^b	221.81 (<0.0001)
Radiology	0.23 (0.71)	0.36 (0.96)	0.45 (1.07)	0.28 (0.87)	0.62 (<0.0001)	0.51 (<0.0001)	0.81 (<0.0001)
Laboratory	0.91 (2.42)	1.13 (3.62)	2.72 (3.85)	0.72 (2.16)	0.81 (0.4652)	0.33 (<0.0001)	1.27 (<0.0001)

Models that did not converge due to small cell sizes are denoted with “–..”

Key differences in utilization frequencies by age and sex are summarized in Figures 2 and 3. The frequency of any outpatient clinic visit tended to decrease after age 17 in all groups; however, while primary care utilization dropped by about 20% between the ages of 14–17 and 22–25 for ADHD, DM, and GP youth, this reduction was only 10% among youth with ASD (Figure 2). Pediatric visits predictably dropped off after age 18 but a larger proportion of youth with ASD (6.2%), ADHD (3.3%), and DM (10.1%) continued to see their pediatrician between the ages of 18 and 25 than among GP controls (2.3%). Neurology utilization increased with age among individuals with ASD while remaining relatively low across all ages in the comparison groups (Figure 2). Mental health utilization decreased at later ages among the comparison groups but remained high at all ages in youth with ASD (Figure 2). Girls with ASD utilized OB/GYN care at significantly lower rates than controls at all ages (Figure 2).

Figure 2.

Age-stratified healthcare utilization in 14–25-year-old individuals with ASD, ADHD, diabetes mellitus, and general population controls, Kaiser Permanente Northern California, 2014–2015.

Figure 3.

Sex-stratified healthcare utilization in 14–25-year-old individuals with ASD, ADHD, diabetes mellitus, and general population controls, Kaiser Permanente Northern California, 2014–2015.

Across all outcome groups, female youth were generally more likely than male youth to utilize outpatient and inpatient services, including primary care, mental health, the ED, and hospitalizations (Figure 3). In general, the relative frequencies of outpatient and inpatient visit utilization between ASD and comparison groups did not differ appreciably by sex.

Preventive healthcare services utilization

Transition-age youth with ASD were more likely to receive a routine adult medical exam, metabolic screening (i.e. liver, glucose, and lipids), and seasonal influenza vaccine relative to GP and ADHD youth, but less likely to receive well adolescent care than youth with ADHD or DM (Table 3). Transition-age girls with ASD had significantly lower rates of some women’s preventive health services such as pelvic exams and insertion of IUDs than all the three comparison groups (Table 3 and Figure 1). In addition, youth with ASD were the least likely to have received a recommended meningitis vaccine though vaccination rates were relatively low in all the outcome groups (18%–23%) (Table 3).

Table 3.

Preventive healthcare utilization among transition-age youth (14–25 years old) with special healthcare needs and the general population, Kaiser Permanente Northern California, 2014–2015.

	ASD	ADHD	DM	GP	ASD versus ADHD	ASD versus DM	ASD versus GP
	(N = 4123)	(N = 20,615)	(N = 2156)	(N = 20,615)	ASD versus ADHD	ASD versus DM	ASD versus GP
	%	%	%	%	Adjusted OR (95% CI)^a
Any routine medical exam	43.56	43.56	36.46	35.69	1.27 (1.18–1.38)	0.99 (0.87–1.13)	1.38 (1.28–1.50)
Well adult care	15.40	7.99	11.46	8.23	2.11 (1.90–2.34)	2.09 (1.73–2.51)	2.07 (1.86–2.30)
Well adolescent care	27.21	29.96	25.09	26.31	0.75 (0.68–0.83)	0.38 (0.31–0.46)	0.95 (0.85–1.05)
Preventive women’s health
Pelvic exam	9.66	15.58	17.17	14.53	0.52 (0.40–0.69)	0.71 (0.52–0.97)	0.53 (0.40–0.70)
IUD	0.88	4.34	2.95	2.76	0.21 (0.10–0.46)	–^b	0.27 (0.13–0.60)
Metabolic screening^c
ALT (obesity only)	30.18	23.68	53.70	17.89	1.49 (1.26–1.75)	0.46 (0.36–0.59)	1.77 (1.49–2.11)
AST (obesity only)	17.36	13.15	28.55	8.28	1.53 (1.25–1.87)	0.72 (0.54–0.96)	1.93 (1.54–2.42)
Fasting glucose (obesity only)	25.46	15.07	27.37	12.26	1.99 (1.67–2.37)	1.14 (0.88–1.48)	2.37 (1.96–2.85)
Any lipid (obesity only)	30.38	18.07	66.72	15.33	2.06 (1.76–2.42)	0.22 (0.18–0.29)	2.45 (2.06–2.90)
Flu vaccine^c	40.14	29.40	38.03	25.23	1.54 (1.43–1.66)	1.15 (1.02–1.29)	1.82 (1.69–1.96)
Meningitis vaccine (Age 14–17 only)	18.07	21.64	22.57	20.55	0.82 (0.72–0.94)	0.80 (0.63–1.02)	0.86 (0.75–0.98)

ASD: autism spectrum disorder; ADHD: attention deficit and hyperactivity disorder; DM: diabetes mellitus; GP: general population; OR: odds ratio; IUD: intrauterine devices; BMI: body mass index; ALT: alanine aminotransferase; AST: aspartate aminotransferase.

Models that did not converge due to small cell sizes are denoted with “–.”

In addition, adjusted for number of outpatient visit counts.

Medication utilization

Youth with ASD had high use of psychotherapeutic medication including antidepressants (26%), antipsychotics (19%), and ADHD medications (13%) (Table 4). After adjustment for covariates, transition-age youth with ASD were more likely to have prescriptions for antidepressants, antipsychotics, and anticonvulsants than all the three comparison groups, and more likely to have prescriptions for cardiovascular, hypertensive, and metabolic conditions than the ADHD and GP groups. Medications for gastrointestinal and respiratory/allergy issues were also more highly prescribed among youth with ASD compared with GP, but similarly prescribed to youth with ADHD.

Table 4.

Prescribed medication utilization among transition-age youth (14–25 years old) with special healthcare needs and the general population, Kaiser Permanente Northern California, 2014–2015.

Medication group	ASD	ADHD	DM	GP	ASD versus ADHD	ASD versus DM	ASD versus GP
	(N = 4123)	(N = 20,615)	(N = 2156)	(N = 20,615)	ASD versus ADHD	ASD versus DM	ASD versus GP
	%	%	%	%	Adjusted OR (95% CI)^a
Any medications (overall)	70.19	69.19	93.14	48.33	1.10 (1.02–1.19)	0.21 (0.18–0.26)	2.37 (2.20–2.56)
Psychotherapeutic agents	44.29	39.15	17.76	5.40	1.28 (1.19–1.37)	5.85 (5.01–6.82)	14.51 (13.21–15.95)
Antidepressants	25.52	13.31	11.36	3.29	2.56 (2.35–2.79)	4.35 (3.64–5.20)	9.72 (8.69–10.88)
Antipsychotics	19.11	4.01	2.23	0.67	6.92 (6.18–7.75)	14.45 (10.43–20.02)	37.98 (31.10–46.37)
ADHD medications	12.59	27.51	2.50	0	0.36 (0.32–0.39)	4.82 (3.57–6.52)	–^b
Other	18.53	9.13	8.53	2.81	2.60 (2.36–2.87)	4.25 (3.46–5.22)	7.48 (6.61–8.46)
Anti-convulsants	5.29	1.28	2.37	0.55	5.00 (4.09–6.11)	4.12 (2.83–5.99)	9.04 (7.07–11.57)
Analgesics/anti-rheumatic agents	11.52	19.23	25.14	14.43	0.62 (0.55–0.69)	0.59 (0.50–0.69)	0.71 (0.64–0.79)
Anti-infective agents	27.21	30.94	41.37	24.58	0.87 (0.81–0.95)	0.72 (0.64–0.82)	1.02 (0.95–1.11)
Gastrointestinal medications	7.32	8.31	15.63	5.42	0.99 (0.86–1.13)	0.78 (0.64–0.95)	1.28 (1.11–1.48)
Respiratory/allergy medications	18.05	18.86	17.81	13.86	1.02 (0.93–1.13)	1.30 (1.10–1.53)	1.23 (1.11–1.35)
DM medications	1.38	0.76	84.93	0.03	1.69 (1.23–2.32)	<0.01 (<0.01–<0.01)	40.37 (17.17–94.92)
Cardiovascular medications	9.51	4.59	15.12	1.50	2.34 (2.06–2.66)	0.82 (0.68–0.99)	6.32 (5.38–7.42)
Lipid/cholesterol medications	0.46	0.14	6.45	0.04	2.47 (1.35–4.52)	0.09 (0.06–0.15)	–^b
Hypertension medications	8.73	4.20	11.27	1.43	2.38 (2.08–2.72)	1.07 (0.87–1.32)	5.93 (5.02–7.00)
Others	33.66	32.95	84.14	26.34	1.09 (1.01–1.17)	0.11 (0.10–0.13)	1.30 (1.20–1.40)

ASD: autism spectrum disorder; ADHD: attention deficit and hyperactivity disorder; DM: diabetes mellitus; GP: general population; OR: odds ratio; BMI: body mass index.

Models that did not converge due to small cell sizes are denoted with “–.”

Psychotherapeutic medication prescriptions among individuals with ASD increased with age from approximately 42% in 14–17 to 51% in 22–25-year-olds (Figure 2). A rise in the frequency of prescriptions for antidepressants occurred by age in all groups, though the growth was largest among individuals with ASD (Figure 2). Use of antipsychotics was also notably elevated and increased with age among youth with ASD (Figure 2). In sex-stratified analyses, high use of psychotherapeutic medication, particularly antidepressants and antipsychotics, was observed in both males and females with ASD relative to same-sex peers; females with ASD had the highest use of antidepressants of both sexes in all the outcome groups though their overall rate of psychotherapeutic medication use was similar to females with ADHD (Figure 3).

Costs of healthcare utilization

The crude mean annual cost of healthcare utilization during the study period was US$5474.16 for transition-age youth with ASD, US$3669.53 for youth with ADHD, US$9725.80 for youth with DM, and US$2016.37 for GP youth (Table 5). After adjustment for covariates, the overall costs of healthcare for transition-age youth with ASD was 2 to 3 times higher than the costs for ADHD and GP youth but less than youth with DM. The largest group differences in outpatient costs were observed for mental health and neurology visits within KP and any health visit outside of KP (Table 5). Medication use also accounted for high costs among youth with ASD who spent significantly more on antipsychotic and anticonvulsive medications compared with peers in other groups (Figure 4 and Table 5).

Table 5.

Healthcare costs^a for transition-age youth (14–25 years old) with special healthcare needs and the general population, Kaiser Permanente Northern California, 2014–2015.

	ASD	ADHD	DM	GP	ASD versus ADHD	ASD versus DM	ASD versus GP
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Adj mean ratio (95% CI)^b
Any outpatient clinic visits	3165.70 (6933.34)	1873.82 (4566.52)	3471.18 (6559.55)	1013.48 (2845.37)	2.02 (1.91–2.13)	1.04 (0.95–1.14)	3.20 (3.02–3.39)
Primary care	627.00 (884.42)	628.82 (813.70)	1491.38 (1831.54)	461.86 (635.15)	1.05 (0.99–1.11)	0.50 (0.46–0.54)	1.27 (1.20–1.35)
Pediatric	370.49 (748.66)	363.67 (717.42)	1107.55 (1671.58)	269.23 (549.34)	1.03 (0.96–1.11)	0.44 (0.40–0.49)	1.21 (1.13–1.30)
General medicine	256.51 (585.83)	265.15 (539.22)	383.83 (986.14)	192.63 (419.46)	1.61 (1.52–1.71)	0.46 (0.41–0.51)	1.87 (1.77–1.99)
Mental health	799.75 (1857.48)	479.70 (1457.66)	286.18 (1242.31)	95.57 (703.78)	1.97 (1.82–2.13)	4.33 (3.76–4.98)	10.25 (9.48–11.08)
Neurology	26.31 (181.87)	8.86 (119.53)	14.22 (139.21)	4.34 (87.45)	2.70 (2.55–2.86)	3.40 (3.07–3.77)	3.82 (3.63–4.02)
OB/GYN	157.09 (451.40)	416.31 (995.99)	586.88 (1697.69)	315.46 (906.10)	0.43 (0.36–0.51)	0.54 (0.44–0.68)	0.57 (0.48–0.68)
Other specialties	254.64 (741.93)	350.80 (1147.10)	385.52 (1047.87)	242.85 (738.65)	0.94 (0.88–1.01)	0.87 (0.77–0.98)	1.06 (0.98–1.14)
Non KP visit	1427.08 (6197.67)	324.37 (3674.47)	989.31 (5404.81)	147.60 (2370.80)	9.08 (8.26–9.97)	0.97 (0.82–1.16)	18.41 (16.69–20.30)
Emergency department
Any KP emergency visit	190.59 (727.84)	305.77 (950.07)	584.39 (1513.89)	165.40 (552.05)	0.68 (0.63–0.74)	0.49 (0.43–0.57)	0.95 (0.88–1.03)
Hospitalizations	752.25 (9278.71)	734.12 (10259.93)	3550.03 (34567.64)	545.36 (13711.01)	1.44 (1.31–1.58)	0.34 (0.28–0.42)	1.82 (1.66–2.00)
Hospitalization in KPNC
Inpatient hospitalization	339.19 (4442.01)	245.45 (3337.13)	1895.04 (12157.85)	203.90 (3761.78)	1.67 (1.52–1.83)	0.15 (0.12–0.18)	1.33 (1.22–1.46)
Outpatient hospitalizations (same day)	205.47 (1389.38)	196.01 (1224.47)	212.67 (1237.43)	156.32 (1093.40)	1.11 (1.02–1.22)	1.35 (1.17–1.56)	1.21 (1.11–1.32)
Hospitalization outside KPNC	413.05 (7949.82)	488.68 (8622.76)	1654.99 (32358.30)	341.47 (12133.04)	1.35 (1.23–1.48)	1.30 (1.07–1.59)	3.24 (2.95–3.55)
Medication
Antipsychotics	755.30 (2753.09)	243.70 (980.77)	78.97 (755.37)	9.84 (204.59)	3.53 (3.26–3.82)	17.44 (15.10–20.13)	83.08 (78.05–88.42)
Anti-convulsives	140.65 (1486.53)	11.46 (259.81)	15.15 (283.03)	9.37 (411.25)	12.17 (11.39–13.00)	18.21 (16.10–20.60)	23.80 (22.34–25.36)
Others	264.19 (1470.11)	304.64 (1886.15)	1813.40 (3361.19)	116.60 (1236.04)	0.89 (0.83–0.95)	0.13 (0.12–0.15)	2.34 (2.19–2.49)
All medications	1160.14 (3570.77)	559.80 (2167.45)	1907.52 (3464.97)	135.81 (1320.69)	2.09 (1.96–2.23)	0.59 (0.53–0.65)	9.01 (8.43–9.63)
Total cost	5474.16 (13405.03)	3669.53 (13120.57)	9725.80 (37226.94)	2016.37 (14679.88)	1.84 (1.74–1.94)	0.80 (0.73–0.87)	3.06 (2.88–3.25)

ASD: autism spectrum disorder; ADHD: attention deficit and hyperactivity disorder; DM: diabetes mellitus; GP: general population; CI: confidence interval; OB/GYN: obstetrics/gynecology; KP: Kaiser Permanente; KPNC: Kaiser Permanente Northern California; BMI: body mass index.

All costs are in U.S. dollars

Figure 4.

Mean annual costs of healthcare in 14–25-year-old individuals with ASD, ADHD, diabetes mellitus, and typical controls, Kaiser Permanente Northern California, 2014–2015.

Age-related differences in healthcare costs varied by group: total costs among youth with ASD decreased after age 17, primarily driven by decreased expenditures on services delivered outside of KPNC (e.g. name types of services). In contrast, among youth in the comparison groups, costs tended to stay the same or increase after age 17 (Figure 4). In general, healthcare costs were higher for transition-age girls than boys in all the outcome groups. However, both girls and boys with ASD had higher total mean costs of healthcare compared with their same-sex peers in the ADHD and GP groups (Figure 4).

Discussion

The unique healthcare utilization profile of youth with ASD over the transition period was characterized by relatively high and stable levels of mental healthcare across the ages of 14–25 and increased frequency of neurology care and psychotherapeutic medication use at older ages. We also observed the highest utilization of outpatient primary care and the lowest utilization of OB/GYN and urgent care among youth with ASD compared with the other groups. Both youth with ASD and GP used the ED at similarly low rates relative to youth with ADHD and DM. Healthcare utilization patterns varied over age with a notable decline in frequency of primary care visits after age 18 for all groups. However, age-related decreases in primary care were smaller among youth with ASD compared to GP youth, suggesting that lack of access to adult providers accepting patients with SHCN is not a large contributor to this drop-off in care. Healthcare costs were generally higher for individuals with ASD compared with ADHD and GP but less than those incurred by individuals with DM.

High healthcare utilization, particularly mental health and neurology, among transition-age youth with ASD is consistent with the high rate of co-occurring medical and psychiatric diagnoses previously reported in this study sample (Davignon et al., 2018). Over one-third of youth with ASD were diagnosed with a co-occurring psychiatric condition—most commonly ADHD, anxiety, depression, and bipolar disorder (Davignon et al., 2018). Utilization of mental health services by youth with ASD was high and relatively stable between the ages of 14 and 25, a finding in line with other reports (Shea et al., 2018; Shimabukuro et al., 2008; Weiss et al., 2018). In contrast, utilization of mental health services decreased after age 17 in youth with ADHD and DM, despite elevated rates of psychiatric conditions such as depression and anxiety in these two groups. Youth in the ASD group commonly received infectious disease, neurologic, allergic/immunologic, and gastrointestinal diagnoses. Neurology visits were one of the only outpatient services to increase over age in youth with ASD and coincided with higher anticonvulsant prescriptions after age 18, likely reflecting the adolescent onset of epilepsy observed in some individuals with ASD (Bolton et al., 2011; El Achkar & Spence, 2015).

While there is disagreement about which transition outcome measures are most informative, increases in ED visits and hospitalizations in young adulthood are commonly used indicators of inadequate primary care (Coyne et al., 2017). We found that ED use among youth with ASD did not differ with age and was lower or comparable with ED use in the other comparison groups. This contrasts with other national studies that have observed higher ED use in young adulthood among individuals with ASD compared with non-ASD peers (Deavenport-Saman et al., 2016; Liu et al., 2017) but is consistent with an earlier report using data from several other large healthcare systems in the United States including KPNC (Cummings et al., 2016). Multiple competing factors could contribute to lower ED use among patients with ASD, such as positive experiences with the child’s primary care providers or, by contrast, a history of negative experiences with the ED (Brousseau et al., 2009; Lunsky et al., 2015). Hospitalization rates and expenditures, particularly same-day and non-KPNC hospitalizations, were higher among ASD youth compared with ADHD and GP, perhaps reflecting emergency care for suicide attempts or other acute psychiatric conditions (Kato et al., 2013; Lokhandwala et al., 2012). Youth with ASD also had high use of inpatient psychiatric visits and psychotherapeutic medications, further underscoring the substantial mental healthcare needs of this population. An age-related shift of ASD healthcare costs toward greater expenditures on inpatient psychiatry, long-term psychiatric care, and psychiatric medication has been noted in other studies (Shea et al., 2018; Shimabukuro et al., 2008).

A study of multiple U.S. health systems reported lower utilization of preventive healthcare including flu and meningitis vaccination and well-child visits among 3–17-year-old children with ASD compared with unaffected peers (Cummings et al., 2016). Some of these patterns are consistent with the findings in our study, which included adolescents and young adults. Individuals with ASD were more likely than ADHD and GP peers to receive some preventive healthcare services like well visits, flu vaccines, and metabolic screenings, and less likely to receive others such as pelvic exams and meningitis vaccines. Flu vaccination rates were similar between youth with ASD and DM, likely attributable to their frequent contact with the healthcare system. Meningitis vaccinations were 14%–20% lower among ASD youth aged 14–17 years compared with the other groups; this could reflect lower college matriculation rates (Wei et al., 2013) and associated immunizations among youth with ASD, patient hypersensitivity to touch and pain, or possibly parents’ enduring concerns about the purported link between vaccines and autism (Zerbo, Modaressi et al., 2018). Girls with ASD also accessed OB/GYN and women’s preventive health services such as pelvic exams and IUD fittings at lower rates than peers in the comparison groups. Low utilization of OB/GYN and cervical screenings has also been noted in other studies of autistic adults at KPNC and elsewhere (Nicolaidis et al., 2013; Zerbo, Qian et al., 2018). Lower rates of IUD fittings may reflect a lower level of engagement in sexual activity, or poorer access to developmentally appropriate sex education, or both. As the unique experiences of autistic women come to light (Tint & Weiss, 2018), further research is needed to understand the origins of this OB/GYN healthcare gap in adolescence and its persistence through adulthood.

Our study adds corroborating evidence from a large integrated health system that healthcare costs are relatively high for transition-age youth with ASD (Shea et al., 2018; Shimabukuro et al., 2008) and that the composition of services may change and costs increase as the individual ages (Cidav et al., 2013). Notably, we show that the healthcare costs for ASD are generally higher than those for youth with ADHD and GP controls. This difference may be partly attributable to California’s insurance mandate for ASD-related health services; research has shown that state insurance mandates, implemented in 46 U.S. states and the Washington D.C., increase child and adolescent healthcare spending, including for services not directly related to ASD (Barry et al., 2017). A larger proportion of youth with ASD had public insurance compared with GP controls in our sample (20% vs 8%) and potentially had lower co-pays which could contribute to increased service use (Davignon et al., 2018). Weiss et al. (2018) conducted a similar analysis using administrative health data in Ontario Canada and found that healthcare expenditures among young adults aged 18–24 with ASD were higher than those of peers with and without other developmental disabilities. However, in comparison to peers with DM, transition-age youth with ASD utilized most healthcare services, with the exception of mental health and neurology services, at similar or lower rates and incurred lower costs.

This study utilized a cross-sectional design which raises the possibility that changes in ASD prevalence, changes in insurance coverage after age 18, and/or provider awareness of ASD over time could induce cohort effects. Furthermore, other studies have noted tendencies of healthcare “drop out” among young adults with ASD (Cidav et al., 2013). Future work could reduce some of these cohort effects and more directly assess change in services with increasing age by examining utilization patterns over time in a cohort of adolescents followed through the transition into young adulthood. A longitudinal analysis may also limit any influence on results arising from geographic variation in the availability of providers across KPNC’s service area. Some inaccuracies in the EMR are also possible; for example, specialty visits may be subject to misclassification if those encounters were categorized as primary care. We also lacked complete data on independent living status and intellectual disability for individuals with ASD, which could be important modifiers of healthcare utilization. We adjusted analyses for common co-occurring psychiatric and medical conditions, but future work may also be able to consider the influence of condition severity. While our GP controls did not have diagnoses of ASD, ADHD, or DM, a small proportion may have had special healthcare utilization patterns due to other chronic health conditions or disabilities. Finally, given that our study was conducted in an integrated healthcare system, our findings are not necessarily generalizable to other healthcare settings or insurance models dissimilar to KPNC. Our findings of increased utilization and costs in individuals with ASD compared to peers without ASD in a mostly commercially insured population were similar to findings in children and adults with public insurance (Cidav et al., 2013; Vohra et al., 2017). However, whether type of insurance coverage, which may change for an individual during the transition to adulthood, could impact utilization and costs warrants closer study.

Study strengths include the large sample size and comprehensive data on diagnostic and utilization variables in transition-age youth receiving most if not all their health services through KPNC’s integrated health system. This allowed us to avoid recall bias and adjust for important confounders influencing utilization including co-occurring conditions. Several studies have compared healthcare needs between youth with ASD and peers with and without other neurodevelopmental disabilities, but we additionally included a medical comparison group of peers growing up with DM. This study also benefits from having a sizable proportion of ASD youth receiving a rigorous diagnostic assessment, including the ADOS through the KPNC ASD clinics.

Conclusion

The healthcare utilization burden and expenditures of youth with ASD generally fell above those of peers with ADHD and at or below those of peers with DM, illustrating the complex combination of psychiatric and medical healthcare needs of individuals with ASD. Psychiatric expenditures including mental health visits, psychiatric inpatient visits, and prescribed psychotherapeutic medications were typically high across the transition period and distinguished youth with ASD from the comparison groups. OB/GYN utilization was notably the lowest among adolescent girls and young women with ASD, a finding that warrants closer examination in studies focusing on the unique healthcare needs of the female autistic population. Utilization of outpatient services, including primary care, was lower at older ages among all outcome groups, suggesting that the transition to adulthood could lead to discontinuities in healthcare for youth with and without SHCN.

Footnotes

Acknowledgements

The authors acknowledge G. Thomas Ray, MBA, at the Division of Research, KPNC, for his contributions to the cost analyses.

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: Funded by Autism Speaks and the Working for Inclusive and Transformative Healthcare (WITH) Foundation.

ORCID iD

Jennifer L Ames

References

Baio

Wiggins

Christensen

D. L.

Maenner

M. J.

Daniels

Warren

. . . Dowling

N. F.

(2018). Prevalence of autism spectrum disorder among children aged 8 years: Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2014. Morbidity and Mortality Weekly Report Surveillance Summaries, 67(6), 1–23. https://doi.org/10.15585/mmwr.ss6706a1

Barry

C. L.

Epstein

A. J.

Marcus

S. C.

Kennedy-Hendricks

Candon

M. K.

Xie

Mandell

D. S.

(2017). Effects of state insurance mandates on health care use and spending for autism spectrum disorder. Health Affairs, 36(10), 1754–1761. https://doi.org/10.1377/hlthaff.2017.0515

Bauman

M. L.

(2010). Medical comorbidities in autism: Challenges to diagnosis and treatment. Neurotherapeutics, 7(3 Special Issue: Physical Health Across the Lifespan), 320–327. https://doi.org/10.1016/j.nurt.2010.06.001

Bolton

P. F.

Carcani-Rathwell

Hutton

Goode

Howlin

Rutter

(2011). Epilepsy in autism: Features and correlates. British Journal of Psychiatry, 198(4), 289–294. https://doi.org/10.1192/bjp.bp.109.076877

Brousseau

D. C.

Gorelick

M. H.

Hoffmann

R. G.

Flores

Nattinger

A. B.

(2009). Primary care quality and subsequent emergency department utilization for children in Wisconsin Medicaid. Academic Pediatrics, 9(1), 33–39. https://doi.org/10.1016/j.acap.2008.11.004

Cadman

Eklund

Howley

Hayward

Clarke

Findon

. . . Glaser

(2012). Caregiver burden as people with autism spectrum disorder and attention-deficit/hyperactivity disorder transition into adolescence and adulthood in the United Kingdom. Journal of the American Academy of Child and Adolescent Psychiatry, 51(9), 879–888. https://doi.org/10.1016/j.jaac.2012.06.017

Cheak-Zamora

N. C.

Yang

Farmer

J. E.

Clark

(2013). Disparities in transition planning for youth with autism spectrum disorder. Pediatrics, 131(3 Special Issue: Physical Health Across the Lifespan), 447–454. https://doi.org/10.1542/peds.2012-1572

Cidav

Lawer

Marcus

S. C.

Mandell

D. S.

(2013). Age-related variation in health service use and associated expenditures among children with autism. Journal of Autism and Developmental Disorders, 43(4), 924–931. https://doi.org/10.1007/s10803-012-1637-2

Coyne

Hallowell

S. C.

Thompson

(2017). Measurable outcomes after transfer from pediatric to adult providers in youth with chronic illness. Journal of Adolescent Health, 60(1), 3–16. https://doi.org/10.1016/j.jadohealth.2016.07.006

10.

Croen

L. A.

Najjar

D. V.

Ray

G. T.

Lotspeich

Bernal

(2006). A comparison of health care utilization and costs of children with and without autism spectrum disorders in a large group-model health plan. Pediatrics, 118(4), e1203–1211. https://doi.org/10.1542/peds.2006-0127

11.

Croen

L. A.

Zerbo

Qian

Massolo

M. L.

Rich

Sidney

Kripke

(2015). The health status of adults on the autism spectrum. Autism, 19(7), 814–823. https://doi.org/10.1177/1362361315577517

12.

Crowley

Wolfe

Lock

McKee

(2011). Improving the transition between paediatric and adult healthcare: A systematic review. Archives of Disease in Childhood, 96(6), 548–553. https://doi.org/10.1136/adc.2010.202473

13.

Cummings

J. R.

Lynch

F. L.

Rust

K. C.

Coleman

K. J.

Madden

J. M.

Owen-Smith

A. A.

. . . Croen

L. A.

(2016). Health services utilization among children with and without autism spectrum disorders. Journal of Autism and Developmental Disorders, 46(3 Special Issue: Physical Health Across the Lifespan), 910–920. https://doi.org/10.1007/s10803-015-2634-z

14.

Davignon

M. N.

Qian

Massolo

Croen

L. A.

(2018). Psychiatric and medical conditions in transition-aged individuals with ASD. Pediatrics, 141(Suppl. 4), S335–S345. https://doi.org/10.1542/peds.2016-4300K

15.

Deavenport-Saman

Smith

Yin

(2016). Do children with autism overutilize the emergency department? Examining visit urgency and subsequent hospital admissions. Maternal and Child Health Journal, 20(2), 306–314. https://doi.org/10.1007/s10995-015-1830-y

16.

El Achkar

C. M.

Spence

S. J

. (2015). Clinical characteristics of children and young adults with co-occurring autism spectrum disorder and epilepsy. Epilepsy & Behavior, 47, 183–190. https://doi.org/10.1016/j.yebeh.2014.12.022

17.

Gordon

(2015). Similarity of the adult Kaiser Permanente membership in Northern California to the insured and general population in Northern California: Statistics from the 2011-12 California Health Interview Survey. https://divisionofresearch.kaiserpermanente.org/projects/memberhealthsurvey/SiteCollectionDocuments/chis_non_kp_2011.pdf

18.

Karter

A. J.

Ferrara

Liu

J. Y.

Moffet

H. H.

Ackerson

L. M.

Selby

J. V.

(2002). Ethnic disparities in diabetic complications in an insured population. Journal of the American Medical Association, 287(19), 2519–2527. https://www.ncbi.nlm.nih.gov/pubmed/12020332

19.

Kato

Mikami

Akama

Yamada

Maehara

Kimoto

. . . Matsumoto

(2013). Clinical features of suicide attempts in adults with autism spectrum disorders. General Hospital Psychiatry, 35(1), 50–53. https://doi.org/10.1016/j.genhosppsych.2012.09.006

20.

Liu

Pearl

A. M.

Kong

Leslie

D. L.

Murray

M. J.

(2017). A profile on emergency department utilization in adolescents and young adults with autism spectrum disorders. Journal of Autism and Developmental Disorders, 47(2), 347–358. https://doi.org/10.1007/s10803-016-2953-8

21.

Lokhandwala

Khanna

West-Strum

(2012). Hospitalization burden among individuals with autism. Journal of Autism and Developmental Disorders, 42(1), 95–104. https://doi.org/10.1007/s10803-011-1217-x

22.

Lounds

Seltzer

M. M.

Greenberg

J. S.

Shattuck

P. T.

(2007). Transition and change in adolescents and young adults with autism: Longitudinal effects on maternal well-being. American Journal of Mental Retardation, 112(6), 401–417. https://doi.org/10.1352/0895-8017(2007)112[401:TACIAA]2.0.CO;2

23.

Lunsky

Paquette-Smith

Weiss

J. A.

Lee

(2015). Predictors of emergency service use in adolescents and adults with autism spectrum disorder living with family. Emergency Medicine Journal, 32(10), 787–792. https://doi.org/10.1136/emermed-2014-204015

24.

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

25.

Nicolaidis

Raymaker

McDonald

Dern

Boisclair

W. C.

Ashkenazy

Baggs

(2013). Comparison of healthcare experiences in autistic and non-autistic adults: A cross-sectional online survey facilitated by an academic-community partnership. Journal of General Internal Medicine, 28(6), 761–769. https://doi.org/10.1007/s11606-012-2262-7

26.

Okumura

M. J.

McPheeters

M. L.

Davis

M. M.

(2007). State and national estimates of insurance coverage and health care utilization for adolescents with chronic conditions from the National Survey of Children’s Health, 2003. Journal of Adolescent Health, 41(4), 343–349. https://doi.org/10.1016/j.jadohealth.2007.04.008

27.

Roux

A. M.

Shattuck Paul

Rast Jessica

Rava Julianna

Anderson Kristy

(2015). National autism indicators report: Transition into young adulthood. https://drexel.edu/autismoutcomes/publications-and-reports/publications/National-Autism-Indicators-Report-Transition-to-Adulthood/

28.

Shea

L. L.

Xie

Turcotte

Marcus

Field

Newschaffer

Mandell

(2018). Brief report: Service use and associated expenditures among adolescents with autism spectrum disorder transitioning to adulthood. Journal of Autism and Developmental Disorders, 48(9), 3223–3227. https://doi.org/10.1007/s10803-018-3563-4

29.

Shimabukuro

T. T.

Grosse

S. D.

Rice

(2008). Medical expenditures for children with an autism spectrum disorder in a privately insured population. Journal of Autism and Developmental Disorders, 38(3 Special Issue: Physical Health Across the Lifespan), 546–552. https://doi.org/10.1007/s10803-007-0424-y

30.

Tint

Weiss

J. A.

(2018). A qualitative study of the service experiences of women with autism spectrum disorder. Autism, 22(8), 928–937. https://doi.org/10.1177/1362361317702561

31.

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

32.

Wei

J. W.

Shattuck

McCracken

Blackorby

(2013). Science, technology, engineering, and mathematics (STEM) participation among college students with an autism spectrum disorder. Journal of Autism and Developmental Disorders, 43(7), 1539–1546. https://doi.org/10.1007/s10803-012-1700-z

33.

Weiss

J. A.

Isaacs

Diepstra

Wilton

A. S.

Brown

H. K.

McGarry

Lunsky

(2018). Health Concerns and Health Service Utilization in a Population Cohort of Young Adults with Autism Spectrum Disorder. Journal of Autism and Developmental Disorders, 48(1), 36–44. https://doi.org/10.1007/s10803-017-3292-0

34.

Zerbo

Modaressi

Goddard

Lewis

Fireman

B. H.

Daley

M. F.

. . . Getahun

(2018). Vaccination patterns in children after autism spectrum disorder diagnosis and in their younger siblings. JAMA Pediatrics, 172(5), 469–475. https://doi.org/10.1001/jamapediatrics.2018.0082

35.

Zerbo

Qian

Ray

Sidney

Rich

Massolo

Croen

(2018). Health care service utilization and cost among adults with autism spectrum disorders in a U.S. integrated health care system. Autism in Adulthood, 1(1), 18–27. https://doi.org/10.1089/aut.2018.0004