Emergency Department Use for Nontraumatic Dental Conditions for U.S. Medicaid-Enrolled Children with Special Healthcare Needs: A Multistate Cross-Sectional Study

Abstract

Background.

Children with special healthcare needs (CSHCN) have greater unmet dental care needs than children without special healthcare needs (SHCN). This study evaluated whether (1) U.S. CSHCN have higher odds of nontraumatic dental conditions (NTDC)-related emergency department (ED) visits than children without SHCN, and (2) whether opioid prescribing following such visits differs by SHCN status.

Methods.

This cross-sectional study used 2019 U.S. Centers for Medicare & Medicaid Services (CMS) unredacted Transformed Medicaid Statistical Information System (T-MSIS) Analytic Files (TAF) Research Identifiable Files (RIF) for Medicaid enrollees up to age 20 years (N=24,959,716). To evaluate associations between SHCN, NTDC-related ED visits, and opioid prescriptions, clustered robust-standard error regression models estimated adjusted odds ratios (aOR) and 95% confidence intervals (CI).

Results.

Approximately 16.5% (n=4,117,103) had a SHCN, and 1.6% (n=398,262) visited the ED for NTDC. Of those with an NTDC-related ED visit, 3.6% (n=14,146) received an opioid prescription. CSHCN were at significantly higher odds of visiting the ED for NTDC (aOR, 1.28; 95% CI, 1.25 to 1.30; P<.001) and were 1.11 times more likely to receive an opioid prescription after an NTDC-related ED visit than children without SHCN (95% CI, 1.03 to 1.19; P=.006).

Conclusions.

Medicaid-enrolled CSHCN are more likely to visit the ED for NTDC and be prescribed opioids from such ED visits than children without SHCN. Future work should identify drivers of these disparities and inform preventive strategies.

Practical implications.

Improving access to preventive and routine dental care for CSHCN could reduce avoidable ED visits and reliance on opioid prescribing.

According to the American Academy of Pediatrics, children with special healthcare needs (CSHCN) “have an increased risk of chronic, physical, developmental, behavioral or emotional conditions and may require supports and services beyond that required by children generally”.¹ One-in-five U.S. children (14.1 million) have a special health care need (SHCN), with increased prevalence observed among males, children ages 12–17 years, non-Hispanic Black children, and those from economically disadvantaged backgrounds, including children enrolled in state Medicaid programs.^2–3 According to 2023 data, over one-third (36.6%) of CSHCN receive health insurance through Medicaid.⁴ CSHCN have a broad range of needs and present with varying degrees of condition severity.⁴ From 2019–2020 U.S. data, 27.2% of CSHCN required solely prescription medications, 26.7% required specialized health services and prescription medications, and 25.5% had a functional limitation.³ Compared to children without SHCN, CSHCN exhibit lower school engagement and report lower quality of life and well-being (e.g., inability to bounce back quickly after facing obstacles, lack of curiosity and interest in learning, and lower happiness levels).^5,6

Some CSHCN experience a higher prevalence of oral diseases, including tooth decay (dental caries), bleeding gums (gingivitis), and toothaches.^7–9 Risk factors associated with tooth decay among CSHCN include difficulty maintaining oral hygiene due to manual dexterity challenges, high consumption of sugar-sweetened foods and drinks often used for behavior management, and reduced salivary flow caused by medications. Irregular toothbrushing also contributes to gingivitis.^10–14 Pain expression difficulties may delay recognition of toothaches and care.¹⁵ CSHCN also exhibit greater barriers to accessing dental care and have higher unmet dental care needs than children without SHCN.^5,10,16 A study based on the 2006 National Survey of CSHCN data found that low-income children with intellectual and/or developmental disabilities (IDD) had 13.4 times the odds of unmet dental care needs than unaffected, higher-income children.¹⁰ Barriers include dental anxiety, lack of dentists trained to treat CSHCN, cost and transportation, and limited provider availability, factors heightened in Medicaid-eligible families.^10,16–18

High levels of unmet dental needs coupled with barriers to dental care may predispose CSHCN to emergency department (ED) visits for managing bacterially-mediated dental diseases, collectively termed non-traumatic dental conditions (NTDC). In Oregon, Medicaid-enrolled CSHCN had 1.6 times the odds of NTDC-related ED use compared to children without SHCN.¹⁹ This study employed a broad definition of SHCN, including IDD, complex chronic conditions (CCC), and chronic medication use (CMU), unlike related literature focusing exclusively on children with cognitive impairments.¹⁹ Other literature indicates geographic heterogeneity in unmet dental care needs, with CSHCN in the West and South regions of the U.S. being approximately 1.5 times more likely to have unmet preventive dental care needs than CSHCN in the Northeast.²⁰ Considering the potential for regional variations in unmet dental care needs for CSHCN, Oregon findings may not generalize nationally.

An added public health concern associated with NTDC-related ED visits for CSHCN is the potential of being prescribed opioids. Although opioids provide short-term relief, they do not treat underlying disease and may cause side effects, misuse, or fatalities, particularly in children on chronic medications.^21,22 Opioid exposure is twice as high among CSHCN as among children without SHCN.²² Data from the 2015–2017 National Hospital Ambulatory Medical Care Survey show Medicaid-enrolled children and adults receiving dental ED care were five times more likely to receive opioids compared to non-dental ED patients.²³ In Oregon, CSHCN were 16% less likely to receive opioids after NTDC-related ED visits than peers without SHCN, though this difference was not significant.¹⁹ There are no national studies on opioid prescriptions from ED visits for NTDC among CSHCN.

The goals of this study were to test the hypotheses that U.S. CSHCN have higher odds of (1) ED use for NTDC and (2) opioid prescription from an ED visit for NTDC compared to children without SHCN. The findings of this study will shed light on the barriers to dental care for low-income CSHCN and offer insights on improving dental care quality for a medically vulnerable subgroup of children.

Materials and Methods

Data source.

The Centers for Medicare & Medicaid Services (CMS) unredacted Transformed Medicaid Statistical Information System (T-MSIS) Analytic Files (TAF) Research Identifiable Files (RIF) from calendar year 2019 were accessed through the CMS Chronic Conditions Data Warehouse (CCW).²⁴ Data extracted included all claims obtained through the inpatient, long-term care, other services, and pharmacy files.

Study participants.

This study included Medicaid and the Children’s Health Insurance Program (CHIP) beneficiaries aged 0 to 20 years who were non-dually eligible for Medicare. “Children” were defined as beneficiaries aged 0–20 years, consistent with Medicaid/CHIP eligibility for individuals under age 21, recognizing that minor variations in age cutoffs exist across states. The inclusion criteria for this study were based on state-level data quality. Beneficiaries included must have lived in a state that was not excluded for data quality concerns according to the CMS Data Quality (DQ) Atlas. States assigned as high concern, unclassified, or unusable based on claims volume, diagnosis codes, and procedure codes were excluded from the analyses (e.g., Illinois, New York, Rhode Island, South Carolina, Tennessee, Texas, Utah). Analyses involving race and ethnicity resulted in the exclusion of 17 additional states because of data quality concerns: Alabama, Arizona, Arkansas, Colorado, Connecticut, District of Columbia, Hawaii, Iowa, Kansas, Louisiana, Maryland, Massachusetts, Missouri, Montana, Oregon, West Virginia, and Wyoming. These exclusions resulted in 24,959,716 Medicaid or CHIP beneficiaries from 27 states included in the broader analyses that did not include race and ethnicity, and a subset of 18,496,189 beneficiaries included in analyses involving race and ethnicity.

Outcome variable.

This analysis included two main outcome measures: (1) experiencing a NTDC-related ED visit; and (2) receiving an opioid prescription following a NTDC-related ED visit. NTDC-related ED visits were defined using the definition developed by the Association of State & Territorial Dental Directors (ASTDD).²⁵ As previously described, a beneficiary was considered to have received an opioid after a NTDC-related ED visit if they were prescribed an opioid within seven days of the visit.¹⁹ Opioid prescriptions were examined as an indicator of management patterns following NTDC-related ED visits and not a direct measure of prescribing appropriateness.

Independent variable and covariates.

The independent variable was SHCN status, defined in three ways as previously described:¹⁹ 1) a medical diagnosis code for IDD; 2) a medical diagnosis code for a CCC; or 3) CMU. To identify children with a medical diagnosis code for an IDD, a set of ICD-10-CM codes was used (Supplemental data). To measure CCC and define CMU, we also followed the methods as previously described.¹⁹

The study also included beneficiary characteristics as model covariates, including age, sex, race, ethnicity, and rural/urban designation. Race and ethnicity were treated as separate variables and self-reported on the application for Medicaid or CHIP. The Economic Research Service Rural-Urban Commuting Area Codes were used to determine the rural/urban status associated with each Medicaid beneficiary’s residence. A state-level indicator for Medicaid adult dental coverage status was also included (categorized as none/emergency only coverage, limited coverage, or comprehensive coverage) according to the classification developed by the Center for Health Care Strategies.²⁶

Statistical analyses.

The NTDC-related ED visit proportions per 1,000 Medicaid/CHIP beneficiaries aged 0 to 20 years, as well as the proportions of beneficiaries receiving an opioid prescription following a NTDC-related ED visit per 1,000 Medicaid/CHIP beneficiaries in the same age group, were calculated for the overall population and by SHCN status. Proportions were reported overall and by age group, sex, race, ethnicity, and residence designation. Chi-square tests were used to test for significant differences in the proportions across categories within each group and whether the category-specific proportions significantly differ across SHCN status. To evaluate the associations between SHCN, NTDC-related ED visits, and opioid prescriptions following a NTDC-related ED visit among all beneficiaries included in the study, clustered regression models with robust standard errors were used to generate odds ratios (OR) and 95% confidence intervals (CI). The models were adjusted for covariates and state-level clustering to generate adjusted odds ratios (aOR). Statistical significance was set at α=0.05. Analyses were conducted with SAS Enterprise Guide 7.1 and Stata version 18.0.

Results

Beneficiary characteristics.

Table 1 displays the counts and proportions per 1,000 Medicaid/CHIP beneficiaries aged 0 to 20 years included in this study overall and by SHCN status. Approximately 16.5% of Medicaid enrollees (n = 4,117,103) had an SHCN. From the broader study population, 4,117,103 children (164.95 per 1,000) had SHCN. The proportion of CSHCN increased across age groups, with the highest proportion at 195.36 per 1,000 for children aged 12 to 20. Males were significantly more likely to have SHCN than females (185.06 vs. 144.12 per 1,000; P<.001). There were significant differences in SHCN prevalence by race and ethnicity, with non-Hispanic white children having the highest proportions of SHCN at 199.74 per 1,000, followed by multiracial, other race, or unknown race children at 188.17 per 1,000 (P<.001). Children living in rural areas were significantly more likely to have SHCN than children living in urban areas (181.23 vs. 161.12 per 1,000; P<.001).

Table 1.

Beneficiary characteristic counts and proportions per 1,000 Medicaid/CHIP beneficiaries ages 0 to 20, overall and by special healthcare needs (SHCN) status, 2019^* (N=24,959,716).

	Beneficiaries Overall No. (proportion)	Beneficiaries with SHCN No. (proportion)	Beneficiaries without SHCN No. (proportion)	P Value‡
Overall	24,959,716 (1000.00)	4,117,103 (164.95)	20,842,613 (835.05)
Age Group (years)				<0.001¶
Age 0 to 6	8,031,135 (321.76)	924,159 (115.07)	7,106,976 (884.93)
Age 7 to 11	6,796,559 (272.30)	1,213,514 (178.55)	5,583,045 (821.45)
Age 12 to 20	10,132,022 (405.93)	1,979,430 (195.36)	8,152,592 (804.64)
Sex				<0.001¶
Female	12,261,452 (491.25)	1,767,099 (144.12)	10,494,353 (855.88)
Male	12,698,264 (508.75)	2,350,004 (185.06)	10,348,260 (814.94)
Race and Ethnicity †				<0.001¶
American Indian/Alaskan Native	311,794 (12.49)	42,860 (137.46)	268,934 (862.54)
Asian/Pacific Islander	700,925 (28.08)	91,703 (130.83)	609,222 (869.17)
Non-Hispanic Black	3,948,286 (158.19)	565,845 (143.31)	3,382,441 (856.69)
Hispanic	5,289,563 (211.92)	750,655 (141.91)	4,538,908 (858.09)
Non-Hispanic White	6,565,298 (263.04)	1,311,369 (199.74)	5,253,929 (800.26)
Multiracial/Other Race/Unknown	1,680,323 (67.32)	316,182 (188.17)	1,364,141 (811.83)
Residence Designation				<0.001¶
Rural	4,751,675 (190.37)	861,158 (181.23)	3,890,517 (818.77)
Urban	20,208,041 (809.63)	3,255,945 (161.12)	16,952,096 (838.88)

^*Beneficiaries from 7 states were excluded due to data quality concerns.

^†Analyses by race and ethnicity resulted in the exclusion of 17 additional states because of data quality.

^‡Calculated using the χ² test.

^¶Statistically significant, P<.05.

NTDC-related ED use.

Table 2 displays the proportion of beneficiaries aged 0 to 20 who experienced a NTDC-related ED visit per 1,000 Medicaid/CHIP beneficiaries overall and by SHCN status. Approximately 1.6% of Medicaid enrollees (n = 398,262) used the ED for NTDC-related needs. Overall, the proportion of experiencing a NTDC-related ED visit was 15.96 per 1,000. The proportion was significantly higher for CSHCN (18.16) than for those without (15.52), which was consistent across all demographic subgroups (P<.001). Children in the youngest age group (ages 0 to 6 years) were significantly more likely to experience a NTDC-related ED visit than older children (P<.001). The proportion for children aged 0 to 6 years (26.86 per 1,000) was more than double the proportion for older children. The proportion of experiencing a NTDC-related ED visit was significantly higher for males (18.15) than for females (13.68) (P<.001). Overall, non-Hispanic Black children had the highest proportions of NTDC-related ED visits (18.37), followed by multiracial, other race, and unknown race children (17.98) and American Indian/Alaskan Native (AIAN) children (17.25). However, among CSHCN, multiracial, other race, and unknown race children had the highest proportions of NTDC-related ED visits (21.05), followed by non-Hispanic black children (20.48) and AIAN children (19.27). Children living in rural areas also had significantly higher proportions of experiencing a NTDC-related ED visit (16.85) compared to urban children (15.75) (P<.001).

Table 2.

Proportion of beneficiaries who experienced a non-traumatic dental condition (NTDC) emergency department (ED) visit per 1,000 Medicaid/CHIP beneficiaries ages 0 to 20, and the proportions of beneficiaries who received an opioid prescription following an NTDC-related ED visit per 1,000 beneficiaries who experienced an NTDC related ED visit, ages 0 to 20, overall and by special healthcare needs (SHCN) status, 2019^* (N=24,959,716).

	Beneficiaries Who Experienced a NTDC-related ED Visit (N=398,262)					Beneficiaries Who Received an Opioid Prescription Following a NTDC-related ED Visit (N=14,146)
	Beneficiaries Overall		Beneficiaries with SHCN	Beneficiaries without SHCN		Beneficiaries Overall		Beneficiaries with SHCN	Beneficiaries without SHCN
	Proportion	P Value‡	Proportion	Proportion	P Value‡	Proportion	P Value‡	Proportion	Proportion	P Value‡
Overall	15.96		18.16	15.52	<0.001¶	35.52		50.50	32.06	<0.001¶
Age Group		<0.001¶					<0.001¶
Age 0 to 6	26.86		31.25	26.29	<0.001¶	5.98		7.83	5.70	<0.001¶
Age 7 to 11	11.72		15.48	10.90	<0.001¶	16.22		19.85	15.10	<0.001¶
Age 12 to 20	10.15		13.70	9.30	<0.001¶	112.38		117.18	110.66	0.004¶
Sex		<0.001¶			<0.001¶		<0.001¶
Female	13.68		15.83	13.32	<0.001¶	43.71		70.23	38.40	<0.001¶
Male	18.15		19.92	17.75	<0.001¶	29.56		38.71	27.23	<0.001¶
Race and Ethnicity †		<0.001¶					<0.001¶
American Indian/Alaskan Native	17.25		19.27	16.93	<0.001¶	43.31		66.59	39.09	0.000¶
Asian/Pacific Islander	8.94		10.83	8.66	<0.001¶	21.38		29.20	19.91	0.063
Non-Hispanic Black	18.37		20.48	18.01	<0.001¶	34.49		44.71	32.55	<0.001¶
Hispanic	13.59		15.86	13.21	<0.001¶	26.07		34.78	24.35	<0.001¶
Non-Hispanic White	16.68		18.02	16.35	<0.001¶	38.85		55.14	34.36	<0.001¶
Multiracial/Other Race/Unknown	17.98		21.05	17.26	<0.001¶	22.01		39.81	16.99	<0.001¶
Residence Designation		<0.001¶					<0.001¶
Rural	16.85		19.02	16.37	<0.001¶	42.77		59.58	38.45	<0.001¶
Urban	15.75		17.94	15.33	<0.001¶	33.69		47.95	30.49	<0.001¶

^*Beneficiaries from 7 states were excluded due to data quality concerns.

^†Analyses by race and ethnicity resulted in the exclusion of 17 additional states because of data quality.

^‡Calculated using the χ² test.

^¶Statistically significant, P<.05.

Opioid prescription following NTDC-related ED visit.

Table 2 displays the proportion of beneficiaries who received an opioid prescription following a NTDC-related ED visit per 1,000 beneficiaries who experienced a NTDC-related ED visit (hereafter referred to as receiving an opioid prescription) overall and by SHCN status. Overall, the proportion of receiving an opioid prescription was 35.52 per 1,000. The proportion was significantly greater for CSHCN (50.50) than for those without SHCN (32.06), which was consistent across demographic variable subgroups (P<.001). The proportion of receiving an opioid prescription was significantly greater among children aged 12 to 20 years (112.38) than it was for younger children. Females were significantly more likely to receive an opioid prescription (43.71) than male children (29.56) (P<.001). The proportion of receiving an opioid prescription was highest among AIAN children (43.31), followed by non-Hispanic white children (38.85) and non-Hispanic Black children (34.49). This ordering was consistent for CSHCN, but the proportions were much higher. For example, for AIAN CSHCN, the proportion was 66.59 per 1,000. Children living in rural areas were significantly more likely to receive an opioid prescription (42.77 per 1,000) than children living in urban areas (33.69 per 1,000) (P<.001).

Clustered robust standard error models.

Table 3 displays the results of the clustered robust standard error models predicting the odds of experiencing a NTDC-related ED visit. CSHCN were 27.7% (aOR, 1.28; 95% CI, 1.25 to 1.30; P<.001) more likely to experience a NTDC-related ED visit than children without SHCN. Compared to children aged 0 to 6 years, the odds of experiencing a NTDC-related ED visit were 57.1% (aOR, 0.43; 95% CI, 0.41 to 0.45; P<.001) lower for children aged 7 to 11 years and 63.1% (aOR, 0.37; 95% CI, 0.33 to 0.41; P<.001) lower for children aged 12 to 20 years. The odds of experiencing a NTDC-related ED visit were 23.7% (aOR, 0.76; CI, 0.75 to 0.78; P<.001) lower for female children than male children. Compared to non-Hispanic white children, non-Hispanic Black children were 12.3% (aOR, 1.12; CI, 1.06 to 1.19; P<.001) more likely to experience a NTDC-related ED visit. Asian/Pacific Islander (API), Hispanic, and multiracial, other race, and unknown race children were less likely to experience a NTDC-related ED visit. There were no significant differences in odds by residence designation (P=.061).

Table 3.

Clustered robust-standard error models predicting the odds of experiencing a non-traumatic dental condition (NTDC) emergency department (ED) visit (N=18,496,189) and the odds of receiving an opioid prescription following an NTDC-related ED visit by special healthcare needs (SHCN) status (N=295,749).

	Odds of Experiencing a NTDC-related ED Visit N=18,496,189		Odds of Receiving an Opioid Prescription Following a NTDC-related ED Visit¶ N=295,749
	aOR* (95% CI†)	P Value‡	aOR† (95% CI‡)	P Value¶
SHCN (Ref. No)
Yes	1.28 (1.25–1.30)	<0.001#	1.11 (1.03–1.19)	0.006#
Age Group (Ref. Age 0 to 6)
Age 7 to 11	0.43 (0.41–0.45)	<0.001#	2.69 (2.21–3.28)	<0.001#
Age 12 to 20	0.37 (0.33–0.41)	<0.001#	21.51 (14.74–31.40)	<0.001#
Sex (Ref. Male)
Female	0.76 (0.75–0.78)	<0.001#	1.32 (1.17–1.49)	<0.001#
Race and Ethnicity (Ref. Non-Hispanic White)
American Indian/Alaskan Native	1.02 (0.88–1.18)	0.79	1.11 (0.91–1.34)	0.30
Asian/Pacific Islander	0.57 (0.50–0.65)	<0.001#	0.66 (0.53–0.82)	<0.001#
Non-Hispanic Black	1.12 (1.06–1.19)	<0.001#	0.82 (0.73–0.92)	0.001#
Hispanic	0.85 (0.76–0.94)	0.002#	0.76 (0.65–0.89)	0.001#
Multiracial/Other Race/Unknown	0.94 (0.88–0.99)	0.048#	0.88 (0.76–1.01)	0.073
Residence Designation (Ref. Urban)
Rural	1.07 (0.99–1.15)	0.061	1.12 (1.00–1.26)	0.050

^*aOR: adjusted odds ratio.

^†95% CI: 95% confidence interval.

^‡The model is adjusted for all variables in the table, and the model is clustered at the state level. The model includes the beneficiaries that live in the 27 states that can be reported because of no claims data or race and ethnicity data quality concerns.

^¶The model is adjusted for all variables in the table and clustered at the state level. It includes the beneficiaries who live in the 27 states that can be reported because of no claims data or race and ethnicity data quality concerns and experienced an emergency department visit for a non-traumatic dental condition.

^#Statistically significant, P<.05.

Table 3 also displays the results of the clustered robust standard error models predicting the odds of receiving an opioid prescription among beneficiaries who experienced a NTDC-related ED visit. CSHCN were 11.0 % (aOR, 1.11; CI, 1.03 to 1.19; P=.006) more likely to receive an opioid prescription than children without SHCN. Compared to children aged 0 to 6 years, the odds of receiving an opioid prescription were 2.7 times (aOR, 2.69; CI, 2.21 to 3.28; P<.001) as high for children aged 7 to 11 years and 21.5 times (aOR, 21.51; CI, 14.74 to 31.40; P<.001) as high for children aged 12 to 20 years. The odds of receiving an opioid prescription were 32.3% (aOR, 1.32; CI, 1.17 to 1.49; P<.001) higher for female children than male children. Compared to non-Hispanic white children, children of all other races and ethnicities were significantly less likely to receive an opioid prescription (P<.001), apart from AIAN (P=.30) and multiracial, other race, and unknown race children (P=.073), whose odds were not significantly different from non-Hispanic white children. There were no significant differences in the odds of receiving an opioid prescription between children living in rural areas and their urban counterparts (P=.50).

Discussion

Our study is the first to use multi-state data to assess NTDC-related ED visits for Medicaid-enrolled children by SHCN status and to examine whether CSHCN were more likely to receive an opioid prescription following an NTDC-related ED visit compared to children without SHCN. Our two main findings were consistent with our hypotheses. CSHCN were nearly 28% more likely to visit the ED for NTDC and about 11% more likely to receive an opioid prescription following their visit compared to children without SHCN.

The first finding was that CSHCN are more likely to have NTDC-related ED visits than children without SHCN. This aligns with an Oregon study showing Medicaid-enrolled CSHCN were 60% more likely to use the ED for NTDC.¹⁹ In contrast, two other studies reported no significant differences for children with IDD or autism spectrum disorder.^27,28 Discrepancies may reflect differences in definitions and populations. Our study and the Oregon study used a broad SHCN definition and focused on Medicaid-enrolled children, while the others examined specific conditions using nationally diverse samples. Notably, both studies still found Medicaid-enrolled children were about 50% more likely to use the ED for NTDC than those with private insurance, highlighting disparities as low-income CSHCN face barriers such as limited dental homes and challenges finding Medicaid-accepting providers.^{10,17,29–31}

The second finding was that CSHCN are more likely to receive an opioid prescription after an NTDC-related ED visit, differing from the Oregon study, which found no significant difference and a non-significant trend toward lower odds for CSHCN.¹⁹ This discrepancy may stem from state-level prescribing differences. Our multi-state data capture variability in opioid policies, dosage limits, and restrictions, while Oregon reflects one state.^32–34 Such heterogeneity underscores the need for both national and state-level studies to guide policy.

This study has clinical and policy implications. Most NTDC-related ED visits are preventable. Our study findings reinforce the importance of developing strategies that reduce NTDC-related ED use and opioid prescriptions among children with SHCN. Further research should examine factors driving NTDC-related ED visits. ED diversion initiatives, first conceptualized in 1990 to enhance access to primary care and outpatient services, may offer an additional approach to addressing these issues.^35,36 Such initiatives have been trialed in select states, including Oregon, and within Medicaid programs in Washington state, demonstrating success in reducing ED utilization.^37,38Clinically, ensuring children with SHCN have access to age-appropriate, office-based dental care provided through a dental home can help prevent reliance on the ED. Two strategies may reduce ED use. First is the need to strengthen medical-dental coordination by encouraging early preventive dental care during wellness visits and sharing management strategies that address the sensory and dexterity challenges common in CSHCN. Second is to support caregivers through programs that highlight oral health risks, promote healthy diets, and reinforce daily brushing with fluoridated toothpaste. Additionally, it is crucial for caregivers to recognize when the appropriate time is to seek dental care. For example, our study found that 7.5% of NTDC-ED use among children ages 0–6 was related to teething, a concern more appropriately managed in a dental office. Among older age groups (7–11 years and 12–20 years), periapical abscesses and dental caries were the most common diagnoses for NTDC-related ED visits. Supporting caregivers in distinguishing between normal and emergent issues could reduce unnecessary visits, financial strain, and inadequate care while promoting appropriate, preventive dental use. Additionally, future research should further examine the common diagnoses associated with NTDC-related ED visits and assess these patterns by age group to better understand age-specific drivers of ED utilization.

This study had six main limitations. First, its cross-sectional design precludes causal inferences. Second, the study population was limited to Medicaid or CHIP enrollees, reducing generalizability to privately insured or uninsured children. Third, reliance on administrative claims may have missed some NTDC-related ED visits or opioid prescriptions, especially out-of-pocket payments. Fourth, states with race and ethnicity data quality concerns were excluded from the analyses, thus, results may not represent the entire Medicaid population. Policies are needed to ensure that race and ethnicity data are mandatory to report at the state level and that federal agencies continue to monitor the quality of such variables. Fifth, while we adjusted for state-level clustering, the regression models did not explicitly account for unobserved between-state differences. Accordingly, residual state-level heterogeneity may have influenced the estimated associations. Future research incorporating state-stratified analyses will be important to directly assess potential heterogeneity across states and to further evaluate the robustness of the cross-state consistency assumption underlying pooled designs. Finally, given the large sample size, small differences may have reached statistical significance despite limited clinical or practical importance. Thus, findings should be interpreted with attention to the magnitude of observed effects.

Conclusions

In conclusion, U.S. CSHCN are more likely to visit the ED for NTDC and to receive opioid prescriptions following these visits than children without SHCN. Future research should explore the reasons behind the higher ED visits for NTDC among CSHCN and the factors contributing to the increased prescription of opioids among this group. Identifying clinical and policy strategies is crucial to address the needs of children and adolescents seeking dental care in the ED. The ultimate goal is to minimize ED visits for NTDC and opioid prescriptions to improve oral health and reduce oral health disparities among CSHCN.