The Economic Burden of Injuries in Children With ADHD in the U.S. From 2011 to 2020

Abstract

Objective:

Few studies have examined the economic burden of and sociodemographic disparities in injuries on families of children with ADHD on a national scale. The objective of this study was to address these literature gaps.

Methods:

Data from 7,102 children with ADHD aged 5 to 17 years in the Medical Expenditure Panel Survey 2011 to 2020 were analyzed for national trends, annualized average, and sociodemographic disparities in injury-related medical expenditures among children with ADHD.

Results:

The national economic burden of injuries for children with ADHD has nearly doubled over the10-year period. These costs were covered by private insurance (62%), public insurance (29%), and families (9%). Asian race was associated with higher total and family expenditures while having low income and public insurance were associated with lower family expenditures.

Conclusions:

Families and society carry significant and increasing economic burdens related to injuries in children with ADHD. Sociodemographic disparities are substantial and of policy relevance.

Keywords

ADHD injury burden expenditure disparity

“The persistent presence of symptoms involving inattention and impulsivity or hyperactivity” (American Psychological Association, 2023) puts millions (Fairman et al., 2020) of U.S. children with ADHD at disproportionally higher risk of unintentional injuries (Danielson et al., 2018; Stavrinos et al., 2011). Of all U.S. children with ADHD, 16.6% have had at least one injury-related condition every year compared to 12.6% of those without ADHD. Children with ADHD are also more likely to suffer from severe injuries (Shen et al., 2023), a top cause of childhood fatalities and disability in the U.S. (National Center for Injury Prevention and Control, 2023).

The heightened prevalence of injuries among children with ADHD calls for further research to examine the consequences of those injuries, one important aspect of which is the economic burden of injury on families of children with ADHD. According to CDC, $31.6 billion per year is spent treating and managing ADHD in the U.S. (CDC, 2022). Yet, it is still poorly understood how and to what extent heightened risk of injury has exerted additional medical expenditures for these families and for society. Most existing estimates on the injury-related expenditures for families of children with ADHD were based on administrative hospital records. Such an approach likely results in an underestimation of both insurance-paid and family/self-paid medical expenditures (Alghnam et al., 2016) because not all caregivers seek medical care at the participating hospitals or medical centers (Haagsma et al., 2020; Shen et al., 2021). Further, few existing studies have examined the longitudinal trend or sociodemographic disparities in injury-related medical expenditures among families of children with ADHD.

The present study advanced our understanding of the economic burden of injury at both family and societal levels among children with ADHD over the past decade. Using a nationally representative sample of noninstitutionalized civilian families of children with ADHD from the population-based Medical Expenditure Panel Survey (MEPS) datasets, this study examines (1) the longitudinal trend of the overall injury-related economic burden and (2) sociodemographic correlates of the economic burden of injuries at both family and societal levels among children with ADHD.

Methods

Ethical Review, Transparency, and Openness

This study was exempted from review by the University of Massachusetts Lowell Institutional Review Board (IRB) because this study only analyzed de-identified datasets that are publicly available. Datasets, codebooks, and documentations of the MEPS database are available at the MEPS website (https://meps.ahrq.gov/mepsweb/). SAS codes used for conducting all analyses in the present study are available upon reasonable request to the corresponding author.

Data Source

The 2011 to 2020 MEPS datasets, which consists of a nationally representative cohort of civilian, non-institutionalized households in the United States over a 10-year period was analyzed in the present study. MEPS is the only national survey project of American families regarding how they use and pay for medical care, health insurance, and out-of-pocket spending. MEPS also includes information relevant to these medical expenditures such as demographics, health conditions (i.e., diagnosis of ADHD), household income, employment/school status, and medical encounters (including injury events). Details regarding how MEPS surveys were implemented and how data were collected (including documentation, instruments, codebooks, etc.) have been documented widely in the existing literature (Chowdhury et al., 2019; J. Cohen, 1997; J. W. Cohen et al., 2009) and as well as the official website of Healthcare Research and Quality (ARHQ; https://meps.ahrq.gov/mepsweb/), which sponsored the MEPS project.

Study Sample and Characteristics

The present study consisted of a sample of 7,102 children with ADHD (29.88% female, age range: 5–17 years old) from the 2011 to 2020 MEPS datasets, representing 6,549,590 children (95% CI: 6,167,332–6,931,847) nationally. Detailed characteristics of the study sample and the national estimates by sociodemographic factors are reported in Table 1 Column “Sample Size” and “National Estimate”, respectively. Levels of these variables were based on the MEPS codebook, which is publicly available on the following page of the official MEPS website (https://meps.ahrq.gov/mepsweb/data_stats/download_data_files_codebook.jsp?PUFId=H147)

Table 1.

Descriptive Characteristics of Children With ADHD in MEPS (2011–2020).

	Sample size (n, %)		National estimate^a	95% CI
Total	7,102	100	6,549,590	[6,167,332–6,931,847]
Injury
Yes	1,020	14	1,023,827	[912,075–1,135,581]
No	6,082	86	5,525,761	[5,196,485–5,855,038]
Age (years)
5–9	2,191	31	1,862,575	[1,702,915–2,022,234]
10–14	3,122	44	2,875,261	[2,669,892–3,080,631]
15–17	1,789	25	1,811,754	[1,651,704–1,971,803]
Sex
Male	4,980	70	4,580,727	[4,275,375–4,886,079]
Female	2,122	30	1,968,862	[1,814,100–2,123,625]
Race/ethnicity
White/other race	3,450	48	4,392,579	[4,071,407–4,713,752]
Black	1,686	24	1,027,017	[899,130–1,154,904]
Hispanic	1,896	27	1,079,705	[957,775–1,201,635]
Asian	70	1	50,288	[32,815–67,762]
Family income level
High income (≥400% FPL)	1,202	17	1,759,387	[1,574,814–1,943,960]
Middle income (200–399% FPL)	1,632	23	1,844,719	[1,681,395–2,008,043]
Low income (<200% FPL)	4,268	60	2,945,484	[2,727,921–3,163,046]
Health insurance
Any private	2,735	39	3,442,477	[3,161,180–3,723,775]
Public only	4,203	59	2,953,027	[2,737,492–3,168,563]
Uninsured	164	2	154,085	[111,608–196,562]
Parental education
Less than high school	1,544	22	1,001,271	[874,453–1,128,089]
High school	1,902	27	1,429,044	[1,297,735–1,560,352]
College and higher	3,656	51	4,119,275	[3,814,433–4,424,118]
Parental marriage status
Both parents	3,283	46	3,586,584	[3,298,642–3,874,527]
Single parents	3,017	43	2,241,421	[2,070,521–2,412,321]
No parents	802	11	721,584	[620,965–822,204]

Note. MEPS = medical expenditure panel survey; FPL = federal poverty line.

Calculated as annual average during 2011 to 2020.

Variables of Interest

Medical Condition Variables

ADHD was defined using a MEPS variable called ADHDADDX, where 1 was coded for having an ADHD diagnosis) and 0 was coded for not having ADHD. As with many other MEPS variables, the ADHDADDX variable was obtained through participant interviews and supplemented/verified by medical records information when applicable and available. Injury was defined using a MEPS variable called INJURY, which was obtained during the MEPS interviews when asking about whether a reported medical condition was a result of “accident/injury” (1 for Yes and 0 for No). Information regarding causes/mechanisms of the injuries were not available.

Outcome Variables

Total medical expenditure was defined as all medical expenditures associated with injury for children with ADHD. Family-paid medical expenditure was defined as medical expenditures associated with injury that were paid by patient or patient families out-of-pocket. Insurance-paid medical expenditure was defined as medical expenditures associated with injury that were paid by either private insurers or public insurance programs such as Medicare, Medicaid, Department of Veterans Affairs, and any other federal, state, or local sources. This variable was calculated as the difference between total and family-paid medical expenditure. Note that all these outcome variables were calculated at the person (child)-level using both annualized national aggregates and per-child median expenditures.

Sociodemographic Variables

Age is a categorical variable with three levels—5 to 9 years (reference), 10 to 14 years, and 15 to 17 years; sex is a categorical variable with two levels—male (reference) and female; race/ethnicity is a categorical variable with three levels—White/Other (reference), Black, Hispanic, Asian; family income level is a categorical variable with three levels—high income (≥400% federal poverty level or FPL; reference), middle income (200%–399% FPL), and low income (<200% FPL); health insurance is a categorical variable with three levels—private (reference), public only, and unsured; parental education is a categorical variable with three levels—less than high school (reference), high school, and college or above; finally parental marital status is a categorical variable with three levels—living with both parents (reference), living with only one parent, or living without a parent.

Data Analysis Plan

A three-step analytic approach was implemented using SAS 9.4 that first analyzed the 10-year temporal trend of injury-related medical expenditures in children with ADHD, followed by calculating the annualized average medical expenditures across the past decade by type of medical service and source of payment as national aggregates. Finally, descriptive and inferential analyses were conducted to examine significant sociodemographic correlates of injury-related medical expenditures in children with ADHD. Details are as follows.

Step 1: 10-Year Trend Analysis

Survey data analysis methods (Heeringa et al., 2017; Lewis, 2016) were implemented to calculate (1a) the yearly national estimates and 95% confidence intervals (CI) for the total number of children with ADHD, (1b) total injury-related medical expenditures, (1c) family and insurance-paid medical expenditures associated with injuries, and finally (1d) median and interquartile range (IQR) for per child injury-related medical expenditures in each of the 10 years from 2011 to 2020.

Step 2: Annualized Average Analysis (national aggregate)

In this step, we combined all 10 years of MEPS data in order to calculate the annualized average national estimates of the economic burden of injuries in children with ADHD across the past decade. Similar to Step 1, survey data analysis methods were used to calculate (2a) the annualized average number of children with ADHD, (2b) the annualized average number of children with ADHD who had an injury, (2c) the annualized average medical expenditures associated with injuries by type of medical service and source of payment (family or society).

Step 3: Sociodemographic Disparity Analysis

Using the combined 10-year dataset, both descriptive and multivariate linear regression (using logarithm transformation) analyses were conducted to examine how sociodemographic factors were associated with injury-related total medical expenditures, injury-related family-paid medical expenditures, and injury-related insurance-paid medical expenditures. The outcome variables in these analyses were annualized averages of respective medical expenditures across the 10-year period from 2011 to 2020. Statistical doctrine dictates that 30 subjects are needed for each predictor/independent variable in the regression model (VanVoorhis & Morgan, 2007), so a sample size of 210 was required for the seven predictor variables (i.e., age, race/ethnicity, sex, parental marital status, parental educational levels, household income levels, and insurance status). Due to the initial logarithm transformation, the coefficient estimate for each sociodemographic factor from the regression models was expressed as the exponential of the beta coefficient (i.e., Exp^(β)) for easier interpretation—representing the ratio of respective medical expenditures between a particular subgroup of that sociodemographic factor and the reference group of that same factor. Accordingly, Exp^(β) that is less than 1 suggests a relatively lower level of medical expenditure of that subgroup compared to the reference group while Exp^(β) that is higher than 1 suggests a relatively higher level of medical expenditure of that subgroup compared to the reference group. All estimates mutually adjusted for all other sociodemographic factors in the regression models.

All analyses accounted for the complex sampling design, weighting structure, and currency inflation (all reported in 2020 U.S. dollars) of the MEPS data when generating respective national estimates reported in this study (Pfeffermann, 1996).

Results

Ten-Year Trend Analysis

As reported in Table 2, results of the 10-year trend of the national economic burden of injuries suggested that overall, the total injury-related medical expenditures have significantly increased (nearly doubled) over the past decade, increasing from $749.9 million in 2011 to $1,515.3 million in 2020 (inflation adjusted). Median expenditures increased from $257 (IQR = $873) per child in 2011 to $439 (IQR = 1,271) in 2020. Figure 1 visualizes this temporal trend in total expenditures with the proportion (area) attributed to family-paid medical expenditures and insurance-paid medical expenditures stacked on the plot. Injury-related insurance-paid medical expenditure reached their peak in 2019, followed by a substantial decline in 2020. Injury-related family-paid medical expenditures have been increasing steadily from a 10-year low in 2017 through 2020. Detailed statistics of the computed national estimates used for plotting Figure 1 was provided in Supplemental Materials.

Table 2.

10-Year Trend of Injury-Related Medical Expenditures Among Children With ADHD*.

Year	Children with ADHD		Total injury-related medical expenditures		Per child expenditures
Year	National estimate	95%CI	National estimate	95% CI	Median	IQR
2011	6,026,884	[5,297,677–6,756,091]	$749,936,833	[$363,329,423–$1,136,544,242]	$257	$873
2012	6,299,531	[5,561,677–7,037,384]	$566,823,062	[$319,682,725–$813,963,399]	$253	$511
2013	6,329,458	[5,547,085–7,111,832]	$694,421,391	[$395,830,926–$993,011,857]	$183	$510
2014	6,581,771	[5,865,376–7,298,166]	$867,523,686	[$533,701,977–$1,201,345,396]	$154	$1,066
2015	6,569,692	[5,855,968–7,283,417]	$725,104,821	[$247,903,271–$1,202,306,371]	$215	$563
2016	6,575,120	[5,754,499–7,395,741]	$1,861,027,459	[$461,753,115–$3,260,301,803]	$222	$685
2017	6,843,429	[6,014,773–7,672,085]	$1,972,047,673	[0–$4,363,210,624]⁺	$141	$574
2018	6,746,736	[5,843,385–7,650,087]	$1,516,669,729	[$283,694,984–$2,749,644,473]	$476	$1,480
2019	6,607,483	[5,838,019–7,376,946]	$3,293,637,985	[0–$6,673,666,051] ⁺	$707	$2,047
2020	6,915,791	[6,046,313–7,785,270]	$1,515,314,664	[0–$3,033,105,165] ⁺	$439	$1,271

Based on data from Medical Expenditure Panel Survey (MEPS) 2011 to 2020; All expenditures were adjusted for inflation and converted to constant 2020 dollars using the Personal Health Care Expenditure (PHCE) index for overall health care. Confidence intervals were calculated using bootstrap methods (the number of repeating bootstrap samples is 2000).

Lower limit of the confidence interval was negative due to large standard errors and was manually recoded to zero to reflect the fact that real-life medical expenditures are always non-negative.

Figure 1.

Stacked plot illustrating the 10-year trend of economic burden of injuries on children with ADHD (2011–2020). The shaded area reflects the burden of insurance-paid medical expenditures while the dashed area represents the additional costs borne by families. The proportion of medical expenditures paid by families has increased steadily since 2017 (difference between the total medical expenditure and insurance-paid medical expenditure).

Annualized Average Analysis

As shown in the top part of Table 3, among the 7,102 children with ADHD sampled from MEPS 2011 to 2020 [representing 6.5 million children (95% CI: 6.1 million–6.9 million) nationwide], 1,020 children (14.4%) had suffered an injury during the MEPS survey period [representing approximately 1 million children with ADHD nationwide (95% CI: 0.9 million–1.1 million].

Table 3.

Annualized Injury-Related Medical Expenditures Among Injured Children With ADHD: by Service Type and Payment Source.^a

	Sample size	National estimate^b	95% CI
Total Number of children with ADHD	7,102	6,549,590	[6,167,332–6,931,847]
Total Number of children with ADHD who had an injury	1,020	1,023,828	[912,075–1,135,581]
Total injury-related expenditures per year in children with ADHD^c		$1,376,250,730	[$887,700,000–$1,864,800,000]
By type of service
Ambulatory		$430,724,694	[$346,066,618–$520,399,242]
Emergency department		$373,869,565	[$270,289,252–$504,664,809]
Hospital inpatient		$508,464,053	[$202,912,311–$924,997,223]
Home health		$25,324,391	NA^d
Prescribed medicines		$37,868,027	[$14,237,108–$70,805,390]
By source of payment
Private insurance		$846,411,390	[$474,049,624–$1,325,452,311]
Family (out of pocket)		$121,733,352	[$79,922,306–$181,660,702]
Medicare		$3,559,103	NA^d
Medicaid		$298,650,718	[$218,439,303–$396,026,964]
Other^e		$109,455,270	[$46,955,866–$218,826,680]

Based on data from Medical Expenditure Panel Survey (MEPS) 2011 to 2020.

Calculated as annual average during 2011 to 2020.

All expenditures were adjusted for inflation and converted to constant 2020 dollars using the Personal Health Care Expenditure (PHCE) index for overall health care. Confidence intervals were calculated using bootstrap methods (the number of repeating bootstrap samples is 2000).

Unavailable because the point estimation was based on small number of patients (n < 10).

Other sources include payments from the Department of Veterans Affairs (except TRICARE); other Federal sources; various State and local sources; various unclassified sources; Medicaid payments reported for persons who were not reported as enrolled in the Medicaid program at any time during the year; work’s compensation, and private insurance payments reported for persons without any reported private health insurance coverage during the year.

As shown in the bottom shaded part of Table 3, these injuries contributed to an annualized average total of $1,376.3 million (95%CI: $887.7 million–$1,864.8 million) in injury-related medical expenditures per year. The top healthcare services that contributed to the economic burden of injuries on families of children with ADHD was inpatient hospitalization (37%), followed by ambulatory services (31%) and emergency department services (27%). Nine percent of the economic burden of injuries was shouldered by families, 62% by private insurance, and the rest by public programs such as Medicare, Medicaid, and other federal, state, and local sources.

Sociodemographic Disparity Analysis

Table 4 shows how economic burden of injuries were distributed across different sociodemographic subgroups. The descriptive analyses of median expenditure per child suggested that older children, females, Asian Americans, those with average household incomes, those with private insurance, those with parental education level above high school, and children who lived with both parents incurred the highest total expenditures associated with injuries. The same pattern was observed for the insurance-paid medical expenditures.

Table 4.

Annualized Injury-Related Medical Expenditures Among Injured Children With ADHD: by Socio-Demographic Factors.^a

	Total medical expenditure ($)	Family-paid medical expenditure ($)	Insurance-paid medical expenditure ($)
	Median per injured child (first–third quartile)	Median per injured child (first–third quartile)	Median per injured child (first–third quartile)
Age (years)
5–9	228 (40–609)	0 (0–36)	147 (0–544)
10–14	247 (13–925)	0 (0–38)	191 (0–755)
15–17	336 (0–815)	0 (0–67)	258 (0–664)
Sex
Male	251 (0–692)	0 (0–38)	183 (0–570)
Female	328 (48–1,185)	0 (0–45)	287 (10–1,052)
Race/ethnicity
White/other race	291 (15–849)	0 (0–55)	213 (0–689)
Black	184 (0–672)	0 (0–0)	155 (0–608)
Hispanic	231 (38–809)	0 (0–22)	153 (0–686)
Asian	444 (193–473)	93 (47–286)	290 (56–314)
Family income level
High income (≥400% FPL)	299 (48–790)	22 (0–123)	184 (3–571)
Middle income (200–399% FPL)	312 (5–1,075)	0 (0–74)	232 (0–853)
Low income (<200% FPL)	209 (0–680)	0 (0–0)	179 (0–632)
Health insurance
Any private	333 (49–922)	11 (0–113)	233 (0–740)
Public only	203 (0–637)	0 (0–0)	172 (0–585)
Uninsured	157 (1–2,250)	86 (0–148)	5 (0–1,165)
Parental education
Less than high school	155 (0–631)	0 (0–0)	122 (0–552)
High school	298 (0–824)	0 (0–0)	217 (0–690)
College and higher	288 (26–858)	0 (0–71)	205 (0–698)
Parental marriage status
Both parents	321 (8–875)	0 (0–79)	235 (0–730)
Single parents	250 (13–756)	0 (0–2)	182 (0–664)
No parents	127 (0–548)	0 (0–0)	97 (0–482)

IQR = interquartile range; FPL = federal poverty line.

Based on data from Medical Expenditure Panel Survey (MEPS) 2011 to 2020.

As shown in Table 5, three multivariate loglinear regression analyses were conducted on how total expenditure, family-paid medical expenditure, and insurance-paid medical expenditure related to injuries were correlated respectively to each sociodemographic variable. For total medical expenditures related to injuries, race/ethnicty emerged as the only significant predictor. Specifically, using White/other as the reference group, children who were Asian were found to incur more injury-related medical costs (Exp^(β) = 2.75, p < .05). For family-paid medical expenditures related to injuries, both race/ethnicity, family income level, and health insurance status were found to be signficiant correlates. Specifically, similar to total expenditure, Asian American children with ADHD were found to incur more injury-related medical costs paid by their families out of pocket than their counterparts who identified themselves as White/Other race (Exp^(β) = 10.29, p < .0001). Furthermore, families with low level of income (<200% federal poverty line) paid less out of pocket for injury-related medical expenses for their children with ADHD than those with high income levels (>400% federal poverty line; Exp^(β) = .49, p < .05). Finally, children with ADHD who were covered by public insurance only were found to incur lower economic burden on their families for injury-related medical costs than those covered by private insurance (Exp^(β) = .20, p < .0001). No significant predictors were identified for injury-related insurance-paid medical expenditures.

Table 5.

Correlates of Injury-Related Medical Expenditures for Injured Children With ADHD.^a

	Model A: total medical expenditure*		Model B: family-paid medical expenditure*		Model C: insurance-paid medical expenditure*
	Exp^(β) **	p	Exp^(β)	p	Exp^(β)	p
Age (years)
5–9 (ref.)
10–14	.9675	.9173	.9357	.8004	1.0179	.9578
15–17	.9043	.7560	1.3831	.2662	.9837	.9612
Sex
Male (ref.)
Female	1.3650	.3241	.8354	.3486	1.4391	.2682
Race/ethnicity
White/other race (ref.)
Black	.7362	.3355	.8171	.3082	.6892	.2538
Hispanic	.9732	.9352	.9256	.7274	.9580	.8986
Asian	2.7471	.0338	10.2883	<.0001	1.9453	.2816
Family income level
High income (≥400% FPL)
Middle income (200–399% FPL)	1.2680	.5128	.8663	.6447	1.7683	.5663
Low income (<200% FPL)	.9585	.9339	.4938	.0441	1.0408	.9669
Health insurance
Any private (ref.)
Public only	.5913	.2388	.1996	<.0001	.6637	.6812
Uninsured	1.1669	.9053	2.5972	.1335	.5488	.5493
Parental education
Less than high school (ref.)
High school	.9105	.8668	.9273	.6795	.9048	.9181
College and higher	.9566	.9398	1.1669	.3205	.7945	.8196
Parental marriage status
Both parents (ref.)
Single parents	1.1859	.5871	.9925	.9745	1.3100	.7857
No parents	.8869	.7993	1.1444	.6538	.5379	.5329

Note. FPL = federal poverty line. Bolded = statistically significant at 0.05.

Based on data from Medical Expenditure Panel Survey (MEPS) 2011 to 2020.

Model A (loglinear)—DV: natural logarithm of injury-related total medical expenditures for a given child with ADHD; Model B (loglinear)—DV: natural logarithm of injury-related family-paid medical expenditures for a given child with ADHD; Model C (loglinear)—DV: natural logarithm of injury-related insurance-paid medical expenditures for a given child with ADHD; IVs for all models: age, sex, race/ethnicity, family income level, health insurance, parental education, parental marriage status; reported regression coefficient β adjusted after accounting for all other IVs in respective regression models.

This column was calculated as the exponential of the beta coefficients generated from the loglinear models for ease of interpretation. Values of this column that correspond to each subgroup can be interpreted as the ratios of that subgroup’s injury-related medical expenditures in comparison to the respective reference subgroup.

Discussion

The present study analyzed a sample of 7,102 children with ADHD aged 5 to 17 years old from the 2011 to 2020 MEPS database to estimate the 10-year trend, annualized average, as well as sociodemographic disparities of the national economic burden of injuries on families of children with ADHD. Injury-related medical expenditures nearly doubled over the past decade for children with ADHD after accounting for inflation. This 10-year trend translated to an annualized average of $1,376,250,730 in injury-related medical expenditures per year between 2011 and 2020, to which the top contributing healthcare service was inpatient hospitalization, ambulatory services, and ED services, respectively. Furthermore, this national economic burden of injuries to children with ADHD was distributed to both families (9%; in the form of out-of-pocket expenditures) and the society (62% by private insurance programs and 29% by public programs). Finally, Asian Americans with ADHD were found to incur more injury-related total and family-paid medical expenditures. Families of children with ADHD with low-income levels and those covered only by public programs were found to incur less family-paid medical expenditures related to injuries.

One possible contributor to such a rise in injury-related medical expenditures may be the increasing awareness of pediatric injuries by parents/guardians for their child with ADHD given the accumulating evidence that children with ADHD are at higher risk of injuries than their counterparts without ADHD (Shen et al., 2023). Combined with the understanding of both short-term and long-term consequences of pediatric injuries (Ruiz-Goikoetxea et al., 2018), families might be more likely to seek medical treatments for their injured child compared to 10 years ago. Additionally, it is also possible that the improved access to healthcare services has availed families of children with ADHD with more options (e.g., inpatient, ambulatory, and ED services) and affordability of the treatment of their child’s injuries (Aggarwal et al., 2022; Luxon, 2015). Further, caution is needed when interpreting the drop of injury-related expenditures in 2020. As much as it is possible that this drop reflected a genuine decrease in the economic burden of injuries for children with ADHD in 2020, we cannot rule out the possibility that this deviation from the general trend (of increasing expenditures in the past 9 years) may be artifacts arising from more biased sampling during the first year of the COVID-19 pandemic (response rate to MEPS questionnaires was 27.6% in 2020 vs. 39.5% in 2019) and the decreased availability of in-person care.

The present study also identified three sociodemographic predictors of injury-related medical expenditures for children with ADHD: household income, insurance status, and Asian American race. Specifically, children with ADHD who live in household with income less than 200% federal poverty line and those who had access to only public insurance programs were likely to incur fewer out-of-pocket (family) expenditures associated with injuries. This finding may not be surprising as most public insurance programs (e.g., Medicaid) have eligibility criteria related to household income levels (Bhanja et al., 2021; Buchmueller et al., 2015) and for those who do qualify, public programs generally require fewer out-of-pocket expenses (e.g., monthly premium, deductible, copayments/coinsurance, and out-of-pocket maximum; Machlin & Carper, 2018; Wray et al., 2021).

Those identifying as Asian Americans—one of the fastest growing population groups in the United States—had higher medical expenditures than all other racial/ethnic groups. One possible explanation might be found in the literature regarding the impact of patient-provider race/ethnicity concordance on healthcare service visits related to pediatric injuries. For example, recent research had revealed that race/ethnicity concordance may significantly increase the likelihood of seeking healthcare services for patients from minority backgrounds (Ma et al., 2019). And among all active physicians with minority backgrounds, Asian Americans accounted for 17.1% of all active physicians compared to Black or African Americans (5%) and Hispanic (5.8%), according to the Diversity in Medicine 2019 Report by the Association of American Medical Colleges. It is possible, therefore, that the higher likelihood for Asian American patients to find race/ethnicity concordant physicians might have contributed to the increased level of healthcare service utilizations (Ma et al., 2019), hence incurring higher levels of injury-related medical expenditures. However, although the race/ethnicity concordance hypothesis might explain why Asian Americans incur higher medical costs than other patients from minority backgrounds, it does not adequately address why injury-related medical expenditures for Asian American patients were found higher than White patients. Other mechanisms such as access to services and even cultural values may be at play here. For example, research in other patient populations (e.g., Medicare fee-for-service enrollees) has found that among all racial/ethnic groups (including White populations), Asians Americans had the most favorable characteristics related to access to healthcare services, such as living in a Metropolitan area, living less than 2 miles from a hospital, and living with adequate physician availability (Kim et al., 2019). It is therefore possible that the ease of access to healthcare services made it more likely for parents of Asian American children with ADHD to take their injured child to a provider for both severe and minor injuries. But readers should be cautious for these interpretations because the relationship between access to and utilization of healthcare services among different racial/ethnic patient populations still require further research (Jetty et al., 2022).

Limitations of the Present Study

There are several important limitations of the present study, mostly due to various constraints of the dataset being analyzed. For example, although this study utilized a nationally representative population-based survey dataset (i.e., MEPS), the sample was nonetheless limited to non-institutionalized civilian families who agreed to participate in the MEPS project from 2011 to 2020. However, using MEPS is still superior compared to other datasets such as hospital administrative records for its (1) comprehensive data collection in family and insurance-paid medical expenditures and (2) its nationally representative sampling framework that adjusts for household nonresponse and survey attrition when determining the required number of initial sample units each year (Chowdhury et al., 2019). A second limitation is the lack of detailed information regarding the injury variable in MEPS such as the type of injury (e.g., unintentional vs. intentional) or the mechanism of injury (e.g., falls, transportation injuries, and animal bites). Such in-depth injury data would have allowed examination of how various injury types/mechanism might have exerted different economic burden on families of children with ADHD. To address these two limitations, future research should try linking MEPS data with other national surveys or databases such as the National Health Interview Survey (NHIS) or National Emergency Department Sample (NEDS) data to broaden the availability of sample and injury data. Third, the examination of sociodemographic disparities of injury-related medical expenditures was limited to the available sociodemographic variables in the MEPS dataset and selected by the study team as conceptually important based on previous literature. But such an approach failed to consider (1) the social determinants of health (SoDH) factors that were not included in MEPS or (2) any non-traditional covariates that were out of the knowledge boundaries of the research team even though they were in fact collected by MEPS. To address the latter limitation, future researchers may consider modern data science techniques such as machine learning algorithms to pull in significantly more (or even all) variables from MEPS to automatically search for significant predictors of injury-related medical expenditures in children with ADHD.

Conclusions

Injuries were found to exert an increasingly significant economic burden on both families of children with ADHD as well as to society. Such burdens were also found to be unevenly distributed among various sociodemographic subgroups. Findings of the present study highlight the importance for the scientists, healthcare providers, and policy makers to consider the burden of injuries when developing behavioral and medical resources for the pediatric ADHD community.

Supplemental Material

sj-docx-1-jad-10.1177_10870547231196328 – Supplemental material for The Economic Burden of Injuries in Children With ADHD in the U.S. From 2011 to 2020

Supplemental material, sj-docx-1-jad-10.1177_10870547231196328 for The Economic Burden of Injuries in Children With ADHD in the U.S. From 2011 to 2020 by Jiabin Shen, Junxin Shi, Lynne Gauthier and Wenjun Li in Journal of Attention Disorders

Footnotes

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: This study was supported by the University of Massachusetts Lowell Seed Grant Program (PI: Jiabin Shen).

Ethical Approval

This is a secondary data analysis of deidentified and publicly available datasets. The UMass Lowell IRB has confirmed that no ethical approval is required.

Consent to Participate

Not applicable as this is a secondary data analysis study.

ORCID iD

Jiabin Shen

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Jiabin Shen (UML Assistant Professor of Psychology): Dr. Shen’s research has focused on the development and evaluation of child-centered injury prevention programs for children with special needs, particularly for children from diverse cultural and socio-economic backgrounds. Dr. Shen is also experienced in conducting population-based health disparity research using MEPS datasets funded by the Health Resources and Services Administration.

Junxin Shi (RINCH Senior Biostatistician) is specialized in analyzing large national databases, including NHIS (National Health Interview Survey), MEPS (Medical Expenditure Panel Survey), NTDB (National Trauma Data Bank), NEDS (Nationwide Emergency Department Sample), NIS (National Inpatient Sample), KIDS (Kid’s Inpatient Database), NRD (Nationwide Readmissions Database), and NBR (National Burn Repository).

Lynne Gauthier (UML Associate Professor of Physical Therapy & Kinesiology) is a mid-career investigator with experience in advanced computer and data analytic techniques to provide neurorehabilitation interventions to underserved populations. Dr. Gauthier is a recipient of continuous grant funding from competitive institutes such as PCORI and American Heart Association.

Wenjun Li (UML Professor of Public Health) is a senior investigator in biostatistics and epidemiology with a focus on applying advanced mathematical and statistical methodologies to investigate critical medical and public health programs. Dr. Li has over 15 years of experience as PI for multiple federal grants by NIH, CDC, AHRQ, and PCORI in both clinical trials and observational studies.

References

Aggarwal

Yeh

R. W.

Dahabreh

I. J.

Robertson

S. E.

Wadhera

R. K.

(2022). Medicare eligibility and healthcare access, affordability, and financial strain for low-and higher-income adults in the United States: A regression discontinuity analysis. PLoS Medicine, 19(10), e1004083.

Alghnam

Vanness

D. J.

Gaskin

D. J.

Thorpe

R. J.

Castillo

(2016). Estimating annual medical and out-of-pocket expenditures associated with traumatic injuries in the United States. Journal of Trauma and Acute Care Surgery, 80(2), 258–264.

American Psychological Association. (2023). APA dictionary of psychology. https://dictionary.apa.org/attention-deficithyperactivity-disorder

Bhanja

Lee

Gordon

S. H.

Allen

Sommers

B. D.

(2021). Comparison of income eligibility for Medicaid vs Marketplace coverage for insurance enrollment among low-income US Adults. JAMA Health Forum, 2, e210771.

Buchmueller

Ham

J. C.

Shore-Sheppard

L. D.

(2015). The medicaid program. Economics of Means-Tested Transfer Programs in the United States, 1, 21–136.

CDC. (2022). Research on ADHD. https://www.cdc.gov/ncbddd/adhd/research.html

Chowdhury

S. R.

Machlin

S. R.

Gwet

K. L.

(2019). Methodology report# 33: Sample designs of the medical expenditure panel survey household component, 1996–2006 and 2007–2016. Agency for Healthcare Research and Quality.

Cohen

(1997). Design and methods of the Medical Expenditure Panel Survey, household component. US Department of Health and Human Services, Public Health Service, Agency.

Cohen

J. W.

Cohen

S. B.

Banthin

J. S.

(2009). The medical expenditure panel survey: A national information resource to support healthcare cost research and inform policy and practice. Medical Care, 47, S44–S50.

10.

Danielson

M. L.

Bitsko

R. H.

Ghandour

R. M.

Holbrook

J. R.

Kogan

M. D.

Blumberg

S. J.

(2018). Prevalence of parent-reported ADHD diagnosis and associated treatment among US children and adolescents, 2016. Journal of Clinical Child & Adolescent Psychology, 47(2), 199–212.

11.

Fairman

K. A.

Peckham

A. M.

Sclar

D. A.

(2020). Diagnosis and treatment of ADHD in the United States: Update by gender and race. Journal of Attention Disorders, 24(1), 10–19.

12.

Haagsma

J. A.

James

S. L.

Castle

C. D.

Dingels

Z. V.

Fox

J. T.

Hamilton

E. B.

Liu

Lucchesi

L. R.

Roberts

N. L.

Sylte

D. O.

(2020). Burden of injury along the development spectrum: Associations between the Socio-demographic Index and disability-adjusted life year estimates from the Global Burden of Disease Study 2017. Injury Prevention, 26(2), i12–i26.

13.

Heeringa

S. G.

West

B. T.

Berglund

P. A.

(2017). Applied survey data analysis. CRC Press.

14.

Jetty

Jabbarpour

Pollack

Huerto

Woo

Petterson

(2022). Patient-physician racial concordance associated with improved healthcare use and lower healthcare expenditures in minority populations. Journal of Racial and Ethnic Health Disparities, 9, 68–71.

15.

Kim

E. J.

Parker

V. A.

Liebschutz

J. M.

Conigliaro

DeGeorge

Hanchate

A. D.

(2019). Racial and ethnic differences in healthcare utilization among Medicare fee-for-service enrollees. Journal of General Internal Medicine, 34, 2697–2699.

16.

Lewis

T. H.

(2016). Complex survey data analysis with SAS. CRC Press.

17.

Luxon

(2015). Infrastructure–the key to healthcare improvement. Future Hospital Journal, 2(1), 4.

18.

Sanchez

(2019). The impact of patient-provider race/ethnicity concordance on provider visits: Updated evidence from the medical expenditure panel survey. Journal of Racial and Ethnic Health Disparities, 6, 1011–1020.

19.

Machlin

S. R.

Carper

(2018). Out-of-pocket health care expenses by age and insurance coverage, 2011. Agency for Healthcare Research and Quality.

20.

Pfeffermann

(1996). The use of sampling weights for survey data analysis. Statistical Methods in Medical Research, 5(3), 239–261.

21.

Ruiz-Goikoetxea

Cortese

Aznarez-Sanado

Magallon

Zallo

N. A.

Luis

E. O.

de Castro-Manglano

Soutullo

Arrondo

(2018). Risk of unintentional injuries in children and adolescents with ADHD and the impact of ADHD medications: A systematic review and meta-analysis. Neuroscience & Biobehavioral Reviews, 84, 63–71.

22.

Shen

Shi

Cooper

Chen

Taylor

H. G.

Xiang

(2021). A population-based study of the incidence, medical care, and medical expenditures for pediatric traumatic brain injury. Journal of Surgical Research, 268, 87–96.

23.

Shen

Shi

Gauthier

(2023). A population-based study on injury risk and disparities among children with ADHD. Journal of Attention Disorders, 27(8), 859–866. https://doi.org/10.1177/10870547231158386

24.

Stavrinos

Biasini

F. J.

Fine

P. R.

Hodgens

J. B.

Khatri

Mrug

Schwebel

D. C.

(2011). Mediating factors associated with pedestrian injury in children with attention-deficit/hyperactivity disorder. Pediatrics, 128(2), 296–302.

25.

VanVoorhis

C. W.

Morgan

B. L.

(2007). Understanding power and rules of thumb for determining sample sizes. Tutorials in Quantitative Methods for Psychology, 3(2), 43–50.

26.

Wray

C. M.

Khare

Keyhani

(2021). Access to care, cost of care, and satisfaction with care among adults with private and public health insurance in the US. JAMA Network Open, 4(6), e2110275.

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

0.01 MB