1 INTRODUCTION

The prevalence of dental disease and access to dental services among the US population remains a public health concern. In contrast with the situation for adults, efforts to improve paediatric dental coverage and affordability have been made over the last two decades. The United States has two major public insurance programmes targeted to families with income limitations: Medicaid and the Children's Health Insurance Program (CHIP). Children under 21 years of age are eligible for Medicaid and receive comprehensive dental benefits under the Medicaid Early and Periodic Screening Diagnostic and Treatment (EPSDT) programme, which is offered in each state.1 Children whose family income is above the Medicaid limit are eligible for CHIP, which covers children under age 19 years and provides dental coverage in each state.2, 3 While both the CHIP and Medicaid generally provide much needed child dental coverage, barriers to accessing dental services are multifaceted and have been widely discussed.4, 5 In 2011, the Centers for Medicaid and Medicare Services (CMS) identified barriers that impact dental services use among children, such as: the limited number of dental providers participating in Medicaid; low reimbursement rates (ie Medicaid pays dentists on average 49.4% of the usual fee for child services)6; the administrative burden on providers to complete claims; poor understanding of dental plans by beneficiaries; and transportation problems resulting in missed dental appointments.7 While low-income children and their families encounter environmental barriers to accessing dental services, parents’ understanding of the importance of oral health also remains a concern. An inadequate understanding of the child's oral health needs may delay the parents’ dental care seeking behaviours, which can lead to a dental emergency.8

Despite efforts to increase access to dental care, the use of Emergency Departments (EDs) for nontraumatic dental conditions (NTDCs) has been on the rise.9 The ED is designed to manage acute, traumatic and life-threatening injuries such as those to the jaw/face, or broken limbs. Nontraumatic dental conditions such as dental caries and its sequelae, periodontal disease, erosion, and impacted teeth should be addressed through regular and timely dental care that access to care initiatives aim to provide. That is, if addressed in the dental setting in a timely manner, NTDCs should not become urgent concerns that necessitate an ED visit for management of pain or infection. Still, in 2014, dental-related total ED charges in the United States were estimated at $2.4 billion dollars. The average hospital charge was $992 per NTDC-related dental visit.10 Several national studies have examined ED utilization in the paediatric population across sociodemographic, economic and clinical characteristics that typically influence the uptake of health services including utilization of ED services.4, 11-13 A few studies have examined NTDCs using the Nationwide Emergency Department Sample (NEDS) but differ from this present study based on the year(s) of dataset used4, 12 and dental disease classification.4 The last study to examine similar dental disease classifications and characteristics of patients using the ED for NTDCs as the present study was conducted over a decade ago, using the 2008 NEDS dataset.12 Since 2008, the national policy landscape has changed, specifically with the passage of the PPACA (Patient Protection and Affordable Care Act, 2010),14 which mandated that all health plans were to include dental coverage for children by 2014. Our study builds on the previous research by using comparable NTDC categories12 and the most recent available data at the time this study was conducted (2014-2015 NEDS data).

The objectives of our study were to the following: (i) describe the characteristics of children and adolescents who present to the ED for NTDCs and compare them to children and adolescents who present to the ED for any other reason; and (ii) identify the specific dental disease diagnosis and expected payer for ED Visits-NTDCs across all paediatric age groups. Through investigating these two objectives, we hope to update the current literature and inform policies that can help in reducing ED visits for NTDCs.

2 METHODS

The NEDS is the largest all-payer ED database in the United States, yielding national estimates of hospital-based ED visits.15-17 We used 2014-15 NEDS data on children and adolescents aged ≤ 21 for this analysis. The NEDS survey comprises of a stratified, single-stage cluster sample of US community hospital-based emergency department (ED) visits across 20% of the hospitals. The dataset includes patient-level and hospital-level characteristics, primary and secondary expected payers (reported by patients and hospital administrators) for the ED services rendered, presence of chronic or nonchronic diseases, principal diagnosis and up to 30 secondary diagnoses all reported as ICD-9-CM codes along with injury codes. In the United States, individuals do not have to pay for care at the time services are rendered. Therefore, the expected payer is requested so that claims can be submitted after the service have been rendered. If a patient is covered by an additional source of insurance, then he or she is considered to have a secondary expected payer. The study was approved by the Institutional Review Board under exempt category and appropriate data use protocols were completed by all researchers.

An emergency department visit was the dependent variable. This was categorized as ED visits for nontraumatic dental conditions (ED visits-NTDCs) and ED visits for any other reason (ED visits-Any Other Reason). Emergency department visits for NTDCs are those which ideally should be addressed during regular routine dental care and should not result in palliative care at the emergency department. The NTDCs for our study were categorized based on a previous study by Allareddy et al12 as well as using the defined Association of State and Territorial Dental Directors (ASTDD)a criteria for nontraumatic dental conditions (NTDCs) and the caries, periodontal, and preventive procedures (CPP) diagnoses.18 Caries, periodontal and prevention diagnoses are a subset of NTDC diagnoses that are routinely provided in primary care general dental practices, or clinics, and excludes procedures that would more likely be addressed by specialists.18 All other dental disease categories that did not fit into the ASTDD classification of NTDCs and CPPs were included in the category ED visits-Any Other Reason. For example, dental conditions such as disorders of teeth and development, dentofacial anomalies, disorders of jaw, and the salivary glands. The category ED visits-Any Other Reason also includes all other paediatric visits for any other nondental concern.1

To address our primary objective, which was to describe the characteristics of children and adolescents who present to the ED for NTDCs and compare them with those children and adolescents who present to the ED for any other reason. Utilizing patient and socioeconomic characteristics (proxy for patient-level data), we included the following variables: age group (<1—Infants, 1-5 years—Preschool, 6-10 years —School age, 11-14 years—Early adolescents, 15-17 years—Middle adolescents, 18-21 years—Late adolescents); gender (female, male); median household income for patient's zip code ($1-$39,999, $40,000-$50,999, $51,000 - $65,999, $66,000 or more); expected primary payer (Medicaid, private, uninsured [includes self-pay, no pay/charity] and other [including Worker's Compensation, CHAMPUS, CHAMPVA, Title V, Medicare or other government programmes]); patient location (large metropolitan areas, small metropolitan areas, micropolitan areas, and not metropolitan or micropolitan areas); time of visit (weekend vs. weekday), disposition (treat and release, transfer to short-term hospital, admitted as an inpatient to hospital, and other (includes death in ED); and the Charlson Comorbidity Index, a weighted index that takes into account the seriousness of a set of 17 comorbid conditions which is then grouped into a 3 category (0, 1,2) Grouped Charlson Comorbidity Index (GRPCI).

To identify the specific dental disease diagnosis and expected payer for ED Visits-NTDCs across all paediatric age groups, we used the ICD 9 and ICD 10 codes and categorized those utilizing a similar protocol for analysis as Allareddy et al12 (see Appendix S1). Specifically, we included three of the four dental disease categories as defined in Allareddy et al12 (ie dental hard tissues and teeth, pulp and periapical tissue, gingival and periodontal disease). The fourth category in the Allareddy et al12 study, cellulitis and abscess of the mouth, is defined as a NTDC-only category in the ASTDD guide; thus, we did not include in the above categories. The other dental diseases and conditions category includes the remainder of the dental diagnoses listed as NTDC and CPP in the ASTDD classification guide.18 The types of dental diagnoses were examined by age and expected payer.

2.1 Statistical analyses

Due to the complex survey design of the dataset, weighted analyses were conducted to provide national estimates of the total number of ED visits for NTDCs and all other visits for children and adolescents aged 0-21 years in 2014-2015. Up to 1.3% of observations were missing from the data and were not included in the analysis. Stratified analyses were conducted to report the proportion of children in the ED visits-NTDCs and ED visits-Any Other Reason categories by patient and socioeconomic characteristics. For ED visits-NTDCs, analysis was performed across age groups and expected payer groups for each of the four types of dental diagnoses. Chi-squared tests were performed to test the statistical significance of differences between groups (α = 0.05).

3 RESULTS

Children aged ≤ 21 years made 70,616,194 ED visits in 2014-15; of those, 463,535 (0.7%) were for ED visits-NTDCs. For every 1000 ED visits among children and adolescents during 2014-15, there were 6.6 ED visits-NTDCs.

3.1 Objective 1: Differences in patient characteristics for ED Visits-NTDCs and ED visits-any other reason

Statistically significant differences were observed for all patient characteristics tested, except for gender when comparing children visiting the ED for NTDCs and children visiting for any other reason (Table 1). Late adolescents aged 18-21 accounted for 51.8% of ED Visits-NTDCs in contrast with 22.4% of ED Visits-Any Other Reason. Approximately, 70% of the ED Visits-NTDCs were made by children from regions with a median household income below $50,999. Medicaid was the major expected primary payer for both types of visits; accounting for 58.2% ED Visits-NTDCs and 57.1% of ED Visits-Any Other Reason. There were more children (19.4%) in the ED Visits-NTDCs group who were uninsured than children (8.8%) in the ED Visits-Any Other Reason group. A larger proportion of children who visited the ED for any other reason (29.6%) reported private insurance as their expected primary payer in contrast with children who visited the ED for NTDCs (18.8%). A higher share of children and adolescents in the ED Visits-NTDCs group were seen on a weekend (32.3%) and treated and released (96.3%) than children in the ED Visits-Any Other Reason group (29.7% and 89.7%, respectively). Lastly, among NTDC visits, fewer children had a GRPCI of one or more severe chronic conditions (4.5%) than children attending the ED any other reason (8.2%) suggesting that the children visiting the ED for NTDCs are healthier.

TABLE 1. ED visits-NTDCs and ED visits-any other reason by patient characteristics and children and adolescents, NEDS data 2014-15 ED visits-NTDCsa ED visits-any other reasonb Age group % [95% CI] % [95% CI] Less than 1 year 0.4 [0.4-0.5] 9.3 [9.3-9.4] 1-5 years 13.9 [13.7-14.1] 28.0 [28.0-28.0] 6-10 years 15.7 [15.4-15.9] 16.2 [16.1-16.2] 11-14 years 7.4 [7.2-7.6] 12.1 [12.1-12.1] 15-17 years 10.5 [10.3-10.7] 11.7 [11.7-11.7] 18-21 years 51.8 [51.5-52.2] 22.4 [22.4-22.4] Gender Female 51.1 [50.8-51.4] 51.2 [51.2-51.2] Male 48.8 [48.5-49.2] 48.7 [48.7-48.7] Median household income $1-$39,999 41.4 [41.1-41.7] 36.5 [36.5-36.5] $40,000-$50,999 29.4 [29.1-29.7] 26.8 [26.8-26.8] $51,000 - $65,999 19.0 [18.8-19.3] 21.2 [21.2-21.3] $66,000 or more 10.0 [9.8-10.2] 15.3 [15.2-15.3] Expected primary payerc Medicaid 58.2 [57.9-58.6] 57.1 [57.1-57.1] Private insurance 18.8 [18.6-19.1] 29.6 [29.6-29.6] Uninsured (Self Pay/No Pay/Charity) 19.4 [19.2-19.7] 8.8 [8.8-8.8] Other (including Medicare) 3.4 [3.2-3.5] 4.3 [4.3-4.3] Patient location Large metro 44.5 [44.2-44.9] 50.3 [50.3-50.3] Small metro 34.7 [34.4-35.0] 32.8 [32.8-32.8] Micropolitan 12.8 [12.6-13.0] 9.9 [9.9-9.9] Noncore 7.7 [7.6-7.9] 6.8 [6.7-6.8] Time of visit Weekend 32.3 [32.0-32.6] 29.7 [29.7-29.8] Weekday 67.6 [67.3-67.9] 70.2 [70.2-70.2] Patient disposition Treat and release 96.3 [96.2-96.4] 89.7 [89.1-89.7] Transfer 0.2 [0.1-0.2] 1.2 [1.2-1.2] Admitted 0.8 [0.7-0.9] 3.6 [3.6-3.6] Other includes death in ER 2.5 [2.4-2.6] 5.4 [5.3-5.4] Grouped Charlson Comorbidity Index 0 95.3 [95.1-95.4] 91.4 [91.4-91.4] 1 4.5 [4.4-4.6] 8.2 [8.1-8.2] 2 0.1 [0.1-0.1] 0.3 [0.3-0.3] Chi-square tests used for p-values. All p-values were significant at p < 0.001 except for gender (p = 0.373). a All children visiting ED for NTDCs. b All children visiting ED for all other reasons. c Expected primary payer refers to the insurance where claims will be submitted after services are rendered. A secondary payer is when an individual has an additional source of insurance for claims to be submitted. 3.2 Objective 2: Dental diagnosis and expected payer across all age groups for NTDCs

In general, among the four types of dental disease categories, ‘other dental diseases and conditions’ accounted for the highest proportion of ED visits (47.6%). These were followed by ‘diseases of pulp and periapical tissues’ (25.3%), ‘diseases of dental and hard tissues’ (19.1%), and the remaining were dental visits related to ‘gingival and periodontal disease’ (7.9%) (Table 2). More than 60% of children < 1 year received a dental diagnosis included in the category ‘gingival and periodontal disease’. In general, the first primary tooth erupts around age 6 months and the clinical appearance of the gingiva or gum tissue during this time is red and inflamed, which may elicit a diagnosis of gingival and periodontal disease. Among all age groups, children age 6-10 years most frequently received a diagnosis in the category ‘diseases of the pulp and periapical tissues’ (36.3%). Diseases of the pulp and periapical tissues are the most common consequence of untreated dental caries. Diseases of the pulp and periapical also featured prominently for those aged 15-17 and 18-21 who received this diagnosis (45.9% combined) and dental hard tissues diagnosis (37.4% combined). The ED diagnosis of ‘other dental diseases and conditions’ (ie jaw pain, unspecified disorders of the teeth and supporting structures, endodontic over/underfill, orthodontic-related dental examination) accounted for (53.6%-54.4%) of diagnosis among adolescents aged 15-21 years. Specifically, unspecified disorders of the teeth and supporting structures made up a large proportion (76%) of the ‘other dental diseases and conditions’ category among all children, suggesting that ED clinicians were unable to categorize the dental concern with the exact diagnostic code.

TABLE 2. ED visits for nontraumatic dental concerns (ED Visits-NTDC) by NTDC categories and age group, NEDS data 2014-15 Total Pop Less than 1 1-5 years 6-10 years 11-14 years 15-17 years 18-21 years % [95% CI] % [95% CI] % [95% CI] % [95% CI] % [95% CI] % [95% CI] % [95% CI] Disorder of dental hard tissues and teeth 19.1 [18.9-19.4] 2.7 [1.5-4.7] 19.2 [18.6-19.9] 17.9 [17.3-18.5] 16.3 [15.5-17.2] 16.9 [16.1-17.6] 20.5 [20.2-20.9] Diseases of pulp and periapical tissue 25.3 [25.1-25.6] 3.5 [2.2-5.6] 28.1 [27.4-28.9] 36.3 [35.5-37.1] 26.1 [25.0-27.1] 24.4 [23.6-25.2] 21.5 [21.2-21.9] Gingival and periodontal disease 7.9 [7.7-8.1] 62.2 [57.6-66.6] 2.4 [23.7-25.2] 8.4 [7.9-8.9] 6.9 [6.3-7.5] 5.1 [4.7-5.5] 3.6 [3.4-3.7] Other dental diseases and conditions 47.6 [47.3-47.9] 31.5 [27.4-36.1] 28.2 [27.4-28.9] 37.4 [36.6-38.1] 50.7 [49.6-51.9] 53.6 [52.6-54.6] 54.4 [53.9-54.8]

Medicaid was the most frequently reported ‘expected primary payer’ among all patients who visited the ED for nontraumatic dental conditions (Table 3). Among late adolescents aged 18-21 years, nearly half (44.9%-49.1%) reported Medicaid and a disproportionately large share (26.3%-32.5%) were uninsured across all four NTDC categories.

TABLE 3. Dental diagnoses by age groups and expected payer, NEDS Data 2014-15 Disorder of dental hard tissues and teeth Diseases of pulp and periapical tissue Gingival and periodontal disease Other dental diseases and conditions % [95% CI] % [95% CI] % [95% CI] % [95% CI] Less than 1 years Medicaid 92.3 [60.8-98.9] 82.2 [56.6-94.2] 82.2 [77.2-86.3] 69.8 [61.7-76.9] Private insurance 7.6 [1.0-39.2] 5.2 [0.7-29.3] 11.4 [8.1-15.7] 18.5 [12.8-25.8] Uninsured 0 7.3 [1.0-37.2] 4.2 [2.3-7.3] 8.3 [4.9-14.0] Othera 0 5.3 [0.7-29.8] 2.2 [1.0-4.7] 3.3 [1.4-7.8] 1-5 years Medicaid 72.7 [70.9-74.4] 72.4 [70.9-73.8] 75.8 [74.4-77.3] 71.9 [70.4-73.3] Private insurance 13.6 [12.3-14.9] 16.4 [15.2-17.6] 14.1 [13.0-15.4] 17.6 [16.4-18.9] Uninsured 11.1 [9.9-12.4] 8.2 [7.4-9.2] 7.5 [6.6-8.4] 7.9 [7.2-8.6] Other 2.6 [2.1-3.3] 3.0 [2.5-3.6] 2.5 [2.0-3.1] 2.5 [2.1-3.1] 6-10 years Medicaid 70.6 [68.8-72.3] 68.3 [67.1-69.6] 71.9 [69.3-74.3] 70.7 [69.5-71.9] Private insurance 15.4 [14.1-16.9] 19.7 [18.6-20.7] 16.5 [14.5-18.6] 18.7 [17.7-19.8] Uninsured 10.9 [9.8-12.2] 8.7 [7.9-9.4] 7.4 [6.0-8.9] 7.9 [7.2-8.6] Other 3.0 [2.4-3.7] 3.3 [2.9-3.8] 4.3 [3.3-5.5] 2.7 [2.3-3.2] 11-14 years Medicaid 69.1 [66.3-71.7] 68.5 [66.3-70.6] 69.2 [65.0-73.2] 67.2 [65.6-68.7] Private insurance 15.6 [13.6-17.8] 19.1 [17.4-21.0] 19.0 [15.8-22.8] 20.5 [19.2-21.9] Uninsured 12.0 [10.2-13.9] 9.6 [8.3-10.9] 8.2 [6.1-10.8] 9.1 [8.2-10.1] Other 3.4 [2.4-4.6] 2.8 [2.1-3.7] 3.6 [2.2-5.7] 3.1 [2.6-3.8] 15-17 years Medicaid 67.7 [65.4-69.8] 64.8 [62.9-66.7] 64.7 [60.5-68.7] 65.5 [64.2-66.7] Private insurance 15.2 [13.5-16.9] 20.7 [19.1-22.4] 21.9 [18.6-25.7] 20.6 [19.5-21.7] Uninsured 14.5 [12.9-16.3] 11.3 [10.1-12.6] 9.7 [7.5-12.6] 10.6 [9.8-11.4] Other 2.6 [2.0-3.5] 3.2 [2.5-3.9] 3.6 [2.3-5.6] 3.4 [2.9-3.9] 18-21 years Medicaid 47.2 [46.2-48.2] 44.9 [43.9-45.8] 46.5 [44.2-48.8] 49.1 [48.5-49.7] Private insurance 17.0 [16.3-17.8] 20.4 [19.6-21.2] 23.0 [21.1-25.1] 20.2 [19.7-20.7] Uninsured 32.5 [31.6-33.4] 30.6 [29.8-31.5] 26.3 [24.4-28.4] 26.9 [26.4-27.4] Other 3.3 [2.9-3.6] 4.1 [3.7-4.5] 4.1 [3.2-5.1] 3.8 [3.6-4.0] a Other category—Medicare, Worker's Compensation, CHAMPUS, CHAMPVA, Title V and other government programmes. 4 DISCUSSION

In this study, we used the NEDS dataset from 2014 to 2015 to compare characteristics of children and adolescents who had an ED visit-NTDCs or an ED visit-Any Other Reason. Among the ED Visits-NTDCs group, we also examined the types of dental diagnoses they received at the ED by expected payer across all age groups. We highlight three key findings. First, late adolescents (18-21 years) made up half of all ED Visits-NTDCs more than any other age group. Second, Medicaid beneficiaries and the uninsured children made up a disproportionately large share of the ED Visits-NTDCs group, whereas Medicaid and private insurance was the most reported expected payer among the ED Visits-Any Other Reason group. Third, ED Visits-NTDCs related to disorders of the hard tissues and diseases of pulp and periapical tissues represents unaddressed dental decay and were common across all age groups, while a diagnosis of unspecified disorders was observed across all age groups, predominantly among adolescents.

Our first and second key findings suggest that late adolescents make up a larger portion of ED Visits-NTDCs, and that the majority of children in the ED Visits-NTDCs group are uninsured or covered by Medicaid, which is consistent with other literature.4, 12 The higher use of EDs among late adolescents for NTDCs may partly be explained by the differences in state policy on Medicaid and CHIP coverage. For example, the Medicaid and CHIP programmes provide coverage to children under 21 years and 19 years respectively; as a result, some of the late adolescents > 19 years lose coverage from CHIP.1-3 Previous researchers have found that the ED appears to be one of the common choices for dental and nondental concerns among children of low-income households and those with Medicaid as the expected primary payer across all age groups4, 11-13, 15, 19, 20 Previous studies also support that ED visits for NTDCs results in palliative nondefinitive care and are more costly than routine treatment visits with a trained dental provider.13, 21 Medicaid and uninsured children still continue to make up almost three-fourths of the NTDCs ED visits as they did in 2008.12 Despite various federal insurance reform policies (ie CHIP, Medicaid) to improve children's access to dental care in the community, ED visits for NTDCs do not show any change in expected payer distribution. This trend may be partially explained by provider shortages for Medicaid-enrolled individuals. As of 2016, only 39% of US dentists participated in Medicaid or CHIP programmes.22 As an attempt to address the dental provider shortage, some states have expanded their dental provider workforce by adopting mid-level oral health practitioners (dental therapists).23

In addition to improving access through dental provider participation, a focus on influencing preventive health behaviours and enhancing community resources is needed to ensure individuals have the means and knowhow to utilize those services.24 For instance, families of children who are Medicaid insured might not be fully aware of the dental benefits, or perceive oral health as a low priority.25 To address these challenges, some communities have used innovative approaches to prioritize oral health by promoting culturally relevant education and outreach. Other communities and programmes have focused on increasing access by streamlining the health care system and dental services, helping families set dental appointments, and arranging transportation.24, 26 In addition, innovative technologies, such as virtual dental homes and teledentistry, have been used to promote oral health while meeting the acute needs of those who cannot travel to dental clinics.24 Perhaps a more widespread adoption of such programmes might help in diverting the ED volume to the community clinics at the national level.

Our third key finding was also supported and consistent with previous research. We found a disproportionately large number of ED Visits-NTDCs to be categorized as unspecified dental disorders, irrespective of age group followed by dental caries-related diagnosis.4, 11-13 This suggests that EDs are not adequately equipped or staffed to diagnose and provide accurate dental diagnosis or definitive dental care.27

Lastly, findings from our study reveal that dental caries and its sequelae continue to be a problem, with almost half of the NTDCs seen in the ED during 2014-2015 being related to untreated dental caries and its consequences. In 2011-2014, 12.4 million of US children had untreated dental caries and nearly 5.4 million were among children aged 12-18 years compared to 4.6 million (aged 6-11 years) and 2.2 million (aged 0-5 years).28 These findings further suggest that untreated dental caries becomes a problem as age increases. When the consequences of dental caries become symptomatic by way of pain or infection, individuals have no other choice but to seek immediate dental care.