Emergency Department Visits for Nontraumatic Dental Problems: A Mixed-Methods Study

Abstract

Objectives. We documented emergency department (ED) visits for nontraumatic dental problems and identified strategies to reduce ED dental visits.

Methods. We used mixed methods to analyze claims in 2010 from a purposive sample of 25 Oregon hospitals and Oregon’s All Payer All Claims data set and interviewed 51 ED dental visitors and stakeholders from 6 communities.

Results. Dental visits accounted for 2.5% of ED visits and represented the second-most-common discharge diagnosis in adults aged 20 to 39 years, were associated with being uninsured (odds ratio [OR] = 5.2 [reference: commercial insurance]; 95% confidence interval [CI] = 4.8, 5.5) or having Medicaid insurance (OR = 4.0; 95% CI = 3.7, 4.2), resulted in opioid (56%) and antibiotic (56%) prescriptions, and generated $402 (95% CI = $396, $408) in hospital costs per visit. Interviews revealed health system, community, provider, and patient contributors to ED dental visits. Potential solutions provided by interviewees included Medicaid benefit expansion, care coordination, water fluoridation, and patient education.

Conclusions. Emergency department dental visits are a significant and costly public health problem for vulnerable individuals. Future efforts should focus on implementing multilevel interventions to reduce ED dental visits.

There are about 2 million annual emergency department (ED) visits in the United States for nontraumatic dental problems, representing 1.5% of all ED visits.¹ Dental visits to the ED have increased over the past decade.^2–6 Management of dental problems in the ED consists primarily of management of pain and infection through analgesics and antibiotics.⁷ Most EDs are not equipped to provide definitive dental care.^8–10

Previous studies^{1–7,9,11–16} have identified lack of insurance, Medicaid insurance, young adult age (18–44 years), and Black race as related to ED dental visits. Furthermore, elimination of state adult Medicaid dental benefits, for example in Oregon¹¹ and Maryland,¹² led to increases in ED dental visits, underscoring the importance of comprehensive dental coverage for Medicaid beneficiaries.

In Oregon, medical and dental benefits under the Medicaid program (Oregon Health Plan) were previously administered through separate fee-for-service programs. As part of an Affordable Care Act State Innovation Waiver, Oregon integrated medical, dental, and behavioral health care into regional Coordinated Care Organizations (CCOs) in 2014.¹³ A CCO performance goal is to reduce ED visits, including those involving nontraumatic dental problems. The Oregon Health Plan expanded coverage to an additional 220 000 individuals, primarily adults. Most enrollees receive services through managed care in which dental care organizations are paid on capitation. Adult coverage was limited to emergency care until 2014, when comprehensive dental benefits were reintroduced for all adults.

There remain important gaps in knowledge, such as procedures, prescriptions, repeat ED dental visit rates, and costs. More importantly, the determinants of and potential solutions to reduce ED dental visits have not been described from the perspectives of ED dental visitors, ED providers, dentists, and health system managers. As part of an effort by local foundations to reduce barriers to oral health, we investigated 2 hypotheses in this mixed-methods study. First, ED dental visits are common, are related to insurance status, result in palliative but not definitive care, and are costly. Second, ED dental visits are caused by modifiable health system, community, provider, and patient factors.¹⁷

METHODS

This mixed-methods study included both quantitative and qualitative components.

Quantitative Study

Oregon does not mandate reporting on ED visits. To address this gap, we obtained ED claims data from 2 complementary claims data sources: (1) a purposive sample (n = 45) of Oregon’s 60 hospitals (the strength of these data are the inclusion of all payer groups for the participating hospitals), and (2) the Oregon All Payer All Claims (APAC) file.¹⁸ The APAC file contains statewide information on ED discharges by patients insured by Medicaid and commercial payers. It includes data about procedures, prescriptions, repeat ED visits, and costs that are unavailable through hospital-supplied data.

The study included all ED visits in calendar year 2010, the most recent available data. We defined an ED visit for a nontraumatic dental problem (ED dental visit) by a primary hospital discharge diagnosis with the following International Classification of Diseases, Ninth Revision codes^{2,5,7,12,14,15,19-21}: 520.0–520.9, 521.0–521.9, 522.0–522.9, 523.0–523.9, and 525.0–525.9.

Claims records had primary diagnosis and up to 5 additional diagnosis codes per visit. We used the primary discharge diagnosis to generate specific but conservative estimates of ED dental visit rate (number of ED dental visitors divided by number of all ED visitors) and associated hospitalizations. To generate upper-bound estimates of ED dental visit rate and potentially associated hospitalizations, we described an additional cohort with a nondental primary diagnosis but with a nontraumatic dental problem in any of the other discharge diagnoses.

All claims data included visit-level information on the patient’s age, gender, race/ethnicity, residential zip code, insurance type, and diagnosis codes. Race/ethnicity data were self-reported. Missing data were flagged with a binary indicator. We did not impute race/ethnicity data because creating a valid algorithm was beyond the scope of this study. We linked residential zip codes to data from the 2010 US Census to create the following zip code–level variables: federal poverty level, high-school graduation rates, and unemployment.²²

We used hospital data to describe visit-level characteristics of ED visitors. With these data, we created a multivariable logistic regression model to identify visit-level predictors of an ED dental visit among all ED visitors.²³ The binary outcome was whether the ED visit had a primary discharge diagnosis for a nontraumatic dental problem. Predictor variables included age, gender, race/ethnicity, insurance type, and residential zip code–level measures of poverty, education, and unemployment. A hospital-level fixed-effects model controlled for clustering by ED.

We used APAC claims data to describe procedures, prescriptions, and repeat ED dental visits. Dental procedures were identified by Current Procedural Terminology codes. We described new prescriptions filled at any pharmacy within 3 days after an ED dental visit; we excluded refills of existing prescriptions. The APAC file includes an encrypted patient identifier, which allows the identification of repeat visits.

Finally, we estimated hospital costs to provide ED dental services. Using APAC data, we applied 2010 Center for Medicare and Medicaid Services national payment tables to all Current Procedural Terminology codes associated with an ED dental visit. This approach is commonly used to estimate hospital costs.²⁴

Qualitative Study

We conducted the study in 6 counties associated with the Oregon Rural Practice–Based Research Network.²⁵ We recruited a purposive sample of past ED dental visitors and community stakeholders. We recruited past ED dental visitors from hospitals and safety net dental clinics. The stakeholders included ED providers (including physicians and nurses), hospital administrators, dental society leaders and dentists, and nonprofit health program executives. We used snowball techniques to identify additional community stakeholders.^26,27

We generated preliminary 12-item interview scripts, which included questions about the interviewee’s beliefs regarding the determinants, delivery, and outcomes of dental care in the ED as well as beliefs regarding future solutions to reduce ED dental visits. We used cognitive interviewing methods to pretest and refine the scripts. The final 12-item scripts were used to train 3 interviewers. Interviews were conducted in person or by phone and recorded. A professional medical transcription service transcribed the data.

We created a codebook to identify (1) delivery and outcomes associated with ED dental visits, (2) determinants of ED dental visits, and (3) interviewee-derived solutions to reduce ED dental visits. Three research assistants coded 3 randomly selected transcripts to establish intercoder agreement through subjective assessment. Discrepancies between the coders were discussed with a fourth coder and resolved. The remaining transcripts were divided among the 3 coders. Each transcript was coded by 2 coders using NVivo 8 software (QSR International Pty Ltd, Victoria, Australia) and the versions merged.

RESULTS

We present findings for the quantitative and qualitative components of this study.

Quantitative Study

We obtained 2010 data on all ED visits from 25 hospitals. Compared with nonparticipating hospitals, participating hospitals were less likely to be located in rural areas and had a greater number of ED visits and inpatient beds (Table A, available as a supplement to the online version of this article at http://www.ajph.org). There were 748 502 ED visits in the study sample, including 15 081 (2.5%) for nontraumatic dental problems. Characteristics of ED visits are described in Table 1. The majority of ED dental visits were by uninsured (49.3%) and Medicaid (32.8%) patients.

TABLE 1—

Characteristics of Oregon Emergency Department (ED) Visitors (Hospital Data): 25 Oregon Hospitals, 2010

Variable	ED Dental Visits (n = 15 081), Mean ±SD or No. (Column %)	All Other ED Visits (n = 733 421), Mean ±SD or No. (Column %)	P
Age, y	32.13 ±12.4	40.46 ±24.2	< .001
Age categories, y
0–14	562 (3.7)	84 759 (14.6)
15–19	783 (5.2)	35 355 (6.1)
20–39	9 951 (66.1)	182 656 (31.4)	< .001
40–64	3 567 (23.7)	174 734 (30.0)
≥ 65	192 (1.3)	105 009 (18.0)
Male gender	7 499 (49.7)	331 294 (45.0)	< .001
Race/ethnicity
White	10 072 (66.8)	474 119 (64.6)
Black	747 (5.0)	28 925 (3.9)
Native American	232 (1.5)	7 206 (1.0)
Hispanic	93 (0.6)	4 357 (0.6)	< .001
Asian	45 (0.3)	7 917 (1.1)
Other	609 (4.0)	47 589 (6.5)
Missing	3 283 (21.8)	163 308 (22.3)
Payer
Uninsured	7 432 (49.3)	129 873 (17.7)
Medicaid - Oregon	4 943 (32.8)	174 148 (23.7)
Medicaid - other states	109 (0.7)	3 428 (0.5)
Commercial	1 436 (9.5)	211 522 (28.8)
Medicare	882 (5.9)	168 033 (22.9)	< .001
Other	187 (1.2)	44 426 (6.1)
Missing	92 (0.6)	1 991 (0.3)
Residential zip code measures, %
Below poverty level	17.9 ±20.3	10.9 ±9.2	.07
Completed high school	86.4 ±9.4	88.5 ±7.2	.4
Unemployed	10.8 ±6.0	10.8 ±8.6	.9

Note. Percentages may not add to 100 because of rounding.

Dental problems were the 12th-most-common ED diagnosis category overall and the second-most-common ED diagnosis for adults aged 20 to 39 years (Tables B1 and B2, available as a supplement to the online version of this article at http://www.ajph.org). The most common dental discharge diagnosis, representing 41% of dental visits, was “unspecified disorder of the teeth” (Table C, available as a supplement to the online version of this article at http://www.ajph.org). Fifty-nine ED dental visits (0.4%) resulted in hospital admission, primarily for infectious complications (Table D, available as a supplement to the online version of this article at http://www.ajph.org).

We identified an additional 2982 ED visits and 266 associated hospitalizations with a secondary diagnosis of a nontraumatic dental problem. (Tables E and F, available as a supplement to the online version of this article at http://www.ajph.org) These suggest an upper-bound ED dental visit frequency of 2.9% and potentially associated hospitalization rate of 1.8%.

Emergency department visitors with no insurance, Medicaid, and Medicare had, respectively, 5.2 (95% confidence interval [CI] = 4.8, 5.5), 4.0 (95% CI = 3.7, 4.2), and 1.9 (95% CI = 1.7, 2.1) times increased odds that their ED visit was associated with a nontraumatic dental problem, compared with those who were commercially insured (Table 2). Young adults (aged 20–39 years) were most likely to experience ED visits, with an odds ratio of 8.2 (95% CI = 7.4, 9.1) compared with the youngest patients (ages 0–14 years; Table 2).

TABLE 2—

Multivariable Predictors of Primary Dental Diagnosis Among Emergency Department Visitors in 25 Oregon Hospitals in 2010

Variable	OR (95% CI)
Age categories, y
0–14 (Ref)	1.0
15–19	3.6 (3.2, 4.1)
20–39	8.2 (7.4, 9.1)
40–64	3.8 (3.4, 4.2)
≥ 65	0.4 (0.3, 0.5)
Gender
Female (Ref)	1.0
Male	1.3 (1.2, 1.3)
Race/ethnicity
White (Ref)	1.0
Black	1.0 (0.9, 1.1)
Native American	1.1 (1.0, 1.3)
Hispanic	0.7 (0.6, 0.9)
Asian	0.5 (0.4, 0.7)
Other	0.6 (0.6, 0.7)
Missing	1.4 (1.2, 1.5)
Payer
Uninsured	5.2 (4.8, 5.5)
Medicaid - Oregon	4.0 (3.7, 4.2)
Medicaid - other states	3.6 (2.5, 5.0)
Commercial insurance (Ref)	1.0
Medicare	1.9 (1.7, 2.1)
Other	0.4 (0.4, 0.5)
Residential zip code measures
Below poverty level (per % Increase)	1.005 (1.00, 1.01)
Completed high school (per % Increase)	0.99 (0.98, 0.99)
Unemployed (per % Increase)	1.00 (0.99, 1.01)

Note. CI = confidence interval; OR = odds ratio. C-statistic = 0.8. Hospital fixed-effect model controlled for clustering of emergency department visits by hospital. Because of > 20% missing data rate, missing race/ethnicity data were coded with a binary indicator.

Within the APAC data set, there were 1 587 649 ED visits resulting in a discharge. There were 14 948 unique patients with commercial or Medicaid insurance who accounted for 25 683 ED dental visits (1.6% of APAC visits to the ED). Dental procedures were seldom performed (Table G, available as a supplement to the online version of this article at http://www.ajph.org): 7% of ED dental visits were associated with a facial nerve block, and 2% resulted in drainage of a dental abscess. Tooth extractions occurred in 0.04% of ED dental visits. The most frequent dispensed medications within 3 days after an ED dental visit were opioid analgesics (associated with 56% of visits), penicillin antibiotics (36%), clindamycin antibiotics (16%), and anti-inflammatory analgesics (9%; Table H, available as a supplement to the online version of this article at http://www.ajph.org).

More than 25% of patients with an ED dental visit had more than 1 ED dental visit in 2010; 21% had 2 to 4 ED dental visits (Figure A, available as a supplement to the online version of this article at http://www.ajph.org). Emergency department dental visits generated mean hospital costs of $402 (95% CI = $396, $408).

Qualitative Findings

We interviewed 17 ED dental visitors and 34 stakeholders. Table I (available as a supplement to the online version of this article at http://www.ajph.org) describes the qualitative study cohort. The qualitative findings were organized into 3 domains: (1) delivery and outcomes associated with ED dental visits, (2) determinants of ED dental visits, and (3) solutions to reduce ED dental visits proposed by interviewees.

Delivery and outcomes associated with emergency department dental visits.

Emergency department dental visitors recognized that oral health was important but were uniformly frustrated at being caught in an endless cycle:

[I’ve been] have told that [tooth decay] can fester into a bad infection and then it can actually kill you. If you are that bad you need to go to the [ED]. But, then you get to the [ED] and they don’t know what to do. “Okay, we’ll give her something for pain. That’s all we can do.” They don’t even refer you somewhere. “Just go to the dentist.” That’s all they say. That’s pretty much going in circles, going around and around.

Most ED dental visitors stated that subsequent visits to the dentist failed to solve their problems, mainly because of their inability to pay for definitive treatment (e.g., root canals, crowns, partial dentures). Over time, many ED dental visitors would skip the dentist and instead rely on the ED for management of acute pain.

Emergency department dental visitors also described alternative ways of managing dental pain:

Basically, if I can’t make it through the pain until the nerve burns out, then I will try to acquire antibiotics on the street and deal with it that way. . . . Another trick is I can go to the [agricultural] coop and buy Terramycin for animals and you can take that for yourself. . . . If you live on a farm and you raise animals, you know it.

Emergency department providers and community dentists agreed that care provided in the ED was palliative. A hospital administrator observed, “One person coming back [to the ED] 3 and 4 times . . . we just can’t keep prescribing them pain medication. Something needs to be done.”

Emergency department dental visitors perceived oral health–related knowledge gaps among ED providers: “[ED providers] don’t really know what they’re doing with teeth . . . or I don’t think they really want to deal with it.”

Determinants of emergency department dental visits.

Table 3 summarizes interviewees’ perceptions about the health system-, community-, provider-, and patient-level factors that contribute to ED dental visits. Health system and community determinants include poor integration of medical and dental services, limited benefits under Medicaid, and lack of water fluoridation. Provider-level factors include a limited supply of dentists willing to accept Medicaid beneficiaries, insufficient dissemination of information about community dental resources, and a lack of alternatives to EDs for management of acute dental problems, particularly on nights and weekends when dental offices are closed.

TABLE 3—

Determinants of Oregon Emergency Department Dental Visits and Potential Solutions, From Interviews With Dental Emergency Department Visitors and Stakeholders: 2010

Domain and Examples	Quote	Solutions Suggested by Interviewees
Health system and community
Disjointed or poor integration of medical and dental care services	“It is not an integrated health solution. [Oral health] is a separate health issue. It is like your mouth is . . . different . . . than the rest of your body so I see it is as being treated separately. . . . Dentistry [is] an afterthought.” (Community stakeholder)	Integrate medical and dental benefits through Coordinated Care Organizations.
	“With the health department [as a place to go for dental care], I mean it helps, it’s cheap, but you don’t have an actual doctor. He’s not your doctor . . . and I just wanted to get everything done from a real dentist [versus a different dentist each time].” (ED visitor)
Problems with Oregon’s Medicaid program (e.g., confusion and misperceptions about which dental services are covered under Medicaid, incomplete or limited coverage of dental services)	“[Most of] our patients . . . are on Medicaid. I don’t think they understand that they have dental coverage . . . where they can go and get cleanings and things like this. I think they just are undereducated on what kind of coverage they have and who to contact.” (ED clinic manager)	Offer all Medicaid enrollees, including adults, comprehensive dental benefits, including regular preventive and restorative care.
	“[Medicaid is] a mirage benefit.” (Community stakeholder)	Assign primary dental care providers to Medicaid enrollees.
	“You just get emergency [treatment]. That’s it. They will not cover fillings. They only cover emergencies.” (Dental society leader)	Use dental case workers to help Medicaid enrollees seek dental care.
	“As soon as that emergency is treated, they won’t cover other things which can help prevent the problems from happening in the first place.” (Dental society leader)	Enact community-level campaign to educate Medicaid enrollees about benefits
		Educate medical and dental providers on coverage.
Lack of community water fluoridation	“[Most ED patients] do not live in areas with fluoridated water. If you grow up without fluoridated water, you are much more vulnerable to dental disease which put[s] you at risk of having problems that would lead you to [the] emergency department.” (ED physician and hospital medical director)	Support efforts to maintain and promote community water fluoridation.
Providers
Insufficient dissemination of information about available dental resources in the community	“We see patients all the time at our clinic that don’t know of our program. We have tried to get the word out in all the emergency rooms, public health clinics, homeless shelters but still we hear of patients that don’t know about [our clinic] until they go to the emergency room.” (Dental society leader)	Provide information to vulnerable individuals about community dental resources available (e.g., free clinics).
No urgent care clinics	“[A] lot that could be treated at an urgent care [clinic] . . . end up coming to the emergency room. . . . Frequently these patients have no other alternative.” (ED charge nurse)	Open additional urgent care clinics.
Inflexible dental office policies and dentists unavailable to provide after-hours treatment (e.g., evenings, weekends, holidays)	“[The dental office] said they couldn’t get me in . . . if the pain is that bad and I [couldn’t] wait [for] an appointment to go ahead [and] go to the hospital.” (ED visitor)	Increase the number of dental clinics. Train more dentists. Hire dentists to staff hospital EDs. Provide continuing medical education courses for ED providers.
Referrals to the ED by dental offices
Overburdened dentists and dissatisfaction; limited supply of dentists who accept Medicaid	“There is a lack of OHP [Oregon Health Plan—the state Medicaid program] providers in these communities. . . . There are a few providers taking the lion’s share of [patients with] OHP, which overburdens them . . . doesn’t allow for care to be delivered in a timely fashion and . . . leads to patient dissatisfaction [and] provider dissatisfaction.” (General dentist in private practice)
ED dental visitors
Social and economic disadvantage	“The number one issue by a long shot was not having enough money to pay for a dentist.” (Nonprofit health organization executive)	Expand dental benefits for Medicaid-eligible patients. Modify oral health behaviors to ensure regular use of preventive and restorative dental care, consistent oral hygiene, and heathy eating.
	“If the [dental] pain is so bad that I have to come up with it, then I will come up with it. . . . Maybe one month, I might not pay the power bill . . . so I can go to the dentist, but the next month I have to pay it off or go on a payment plan.” (ED visitor)
Health-related culture and values	“People who have been on Medicaid for years and years and years, they just use the [ED]. It is the easiest to thing to do. I can go anytime I want. I don’t have to have an appointment.” (ED nurse manager)
	“[ED visitors] are not forward thinking. They don’t try and make appointments. They are not engaged in their own care and so they come to the [ED] because it is an easy, fast thing to do. It doesn’t require any accountability, making an appointment, follow-up, anything.” (ED nurse manager)
	“The problem with the emergency room is that I know that if I have an emergency . . . they have to give me care. They can’t deny me. There is a law . . . whether I have money or not.” (ED visitor)
Oral health behaviors (e.g., symptom-driven dental care use, inconsistent visits to the dentist, dental fears, unhealthy dietary habits, poor hygiene)	“I don’t really feel the need [to seek regular dental care] . . . although I think if I was flush with cash, I [still] wouldn’t be going to the dentist necessarily. It’s not one of my habits.” (ED visitor)
	“A lot of people have a really substantial amount of fear. . . . Fear of pain.” (General dentist)

Note. ED = emergency department. The stakeholders included ED providers (including physicians and nurses), hospital administrators, dental society leaders and dentists, and nonprofit health program executives.

A major theme among patient-level factors was limited financial resources for definitive care. Other themes included suboptimal patient behavior such as reliance on EDs for routine care, poor dietary and hygiene behaviors, and fear of dentists and dental treatment.

In regard to knowledge about Medicaid, there were differences among ED dental visitors and health providers. All interviewed ED dental visitors knew Medicaid offered limited dental coverage. Among providers, dentists were also aware of the coverage limitations. On the other hand, some ED providers believed Medicaid provided all adult enrollees with comprehensive dental care. As one physician observed,

[A] large population of our patients have state-funded insurance. They are on Medicaid and I don’t think they understand that they have dental coverage. . . . I think they just are undereducated on what kind of coverage they have.

Solutions proposed by interviewees.

Interviewees proposed a number of solutions (Table 3). The most common was increasing the dental workforce, which was mentioned by ED dental visitors, ED providers, dentists, and other community stakeholders. Some ED providers believed educating Medicaid enrollees about their dental benefits and opening more urgent care clinics would reduce ED dental visits. Dental visitors to the ED believed the most important solution was expanding Medicaid dental benefits. Other proposed solutions included improving dental care use for Medicaid enrollees through integrated medical–dental benefits within CCOs and use of caseworkers, promoting water fluoridation, and patient education to improve oral health behaviors.

DISCUSSION

We investigated the scope of ED dental visits in Oregon, and our combination of quantitative and qualitative findings yielded insights into determinants and potential solutions. Overall, our data revealed a cohesive story. Many socioeconomically vulnerable individuals undergo repeat ED dental visits because of the inability to access definitive dental treatment. Most ED dental visitors left the ED with only palliative care in the form of pain medications. Emergency department dental visitors and community stakeholders provided a number of solutions to address these problems. However, no single interviewee articulated a comprehensive, integrated solution aimed at preventing ED dental visits and addressing the immediate treatment needs of ED dental visitors.

Our estimate of ED dental visit rates in Oregon (2.5%) is higher than the nationally reported rate (1.5%).¹ Oregonians visit EDs more for nontraumatic dental problems than for heart disease, pneumonia, or asthma (Table B1, available as a supplement to the online version of this article at http://www.ajph.org). Extrapolation to all Oregon EDs and all payer groups suggests that there are more than 28 000 annual ED dental visits. There are several potential explanations for the greater ED dental visit rate in Oregon. First, the Oregon Medicaid program eliminated adult dental benefits in 2003, which resulted in increased ED dental visits.¹¹ Second, many rural communities in Oregon have limited dental care options, and vulnerable individuals in these communities are at greater risk for having unmet dental needs.^28,29 Third, there is local resistance to water fluoridation, which contributes to greater prevalence of dental caries.¹⁶

Consistent with previous literature,^{1–7,9,11–15} we found that uninsured and Medicaid beneficiaries represented a high proportion (82%) of ED dental visitors. This is likely explained by limited resources to access definitive dental care in non-ED settings. The interviews revealed that the ED is perceived as the only available setting for management of acute dental problems.

Vulnerable adults will be affected by ongoing efforts to expand and restructure the Oregon Medicaid program. At the time of the interviews, most adult enrollees did not have dental benefits. Although dentists and ED dental visitors were aware of this, most ED providers continued to believe that enrollees had benefits. Medicaid expansion has now reinstated adult dental benefits. Conflicting perceptions about Medicaid highlight the importance of concerted efforts to educate Medicaid providers and enrollees about benefits.

Most ED dental visits resulted in palliative rather than definitive care (e.g., tooth extraction). The high frequency of nonspecific dental discharge diagnosis suggests ED provider uncertainty in managing dental problems. Although it is unreasonable to view the ED as a setting where dental problems are managed routinely, the observation that up to 1.8% of dental visits presented with complications requiring hospitalization highlights the need to preserve access to EDs for dental care unless emergent dental care is available elsewhere—both to start antibiotics before infectious complications develop and to ensure emergent care for those dental problems requiring it.

We generated novel estimates of hospital costs for dental visits. If extrapolated to all Oregon ED dental visits, our data suggest that there is more than $11 million in annual hospital costs, which is likely an underestimation because provider fees are excluded. These findings underscore the importance of strategies that redirect scarce resources currently spent on dental care in the ED to efforts leading to definitive care.

The qualitative data were consistent with our hypothesis that determinants of ED dental visits are present at multiple levels. The solutions offered by interviewees focused on single-level solutions, which revealed the limitation of each individual stakeholder in fully understanding the complexity of the problem. The most common solution proposed—to increase the dental workforce—is impractical because of cost implications and is unlikely to address the multilevel determinants of ED dental visits.

On the basis of our qualitative findings, we generated a conceptual model on the ED dental visit process (Figure 1). There are modifiable determinants of ED dental visits organized into 4 domains: health system, community, providers, and patients (Table 3). Reducing ED dental visits will involve interventions at each of these levels. Initially (at time 0), a vulnerable individual exhibits symptomatic dental pain. Some individuals never make it to a dentist because of financial and structural barriers to care, at which point the individual may seek care in the ED. If the individual is able to make it to the dentist, the recommended treatment is too costly, at which point the individual is directed to the ED. In either case, the patient receives palliative care in the ED and is referred to a dentist. The individual is shuffled back and forth from dental office to the ED. Over time, the individual stops going to the dentist and relies on the ED (time 1).

FIGURE 1—

Conceptual model of emergency department (ED) dental visits.

There is a 2-part strategy based on our model that could be adopted. The first is a primary prevention approach to reduce the peak prevalence of ED dental visits during young adulthood. Our data indicate that adults aged 20 to 39 years have the greatest odds of an ED dental visit, which is consistent with national data.³⁰ These findings suggest that most ED-related problems start during adolescence, when dental utilization rates drop³¹ and oral health behaviors (e.g., tooth brushing, frequency of carbohydrate consumption) worsen. Thus, interventions should address the problem by targeting Medicaid-enrolled adolescents. Such interventions could be implemented in schools and involve school nurses or hygienists to screen and refer adolescents to dental homes. A case manager could work with families and dental offices to ensure appropriate disease prevention and management. In addition, school- and community-based prevention programs could work to improve the quality of school lunches and eliminate sugar-sweetened beverages from vending machines. Home-based interventions could focus on improving oral hygiene behaviors, particularly twice-daily toothbrushing with fluoride toothpaste. Innovative strategies will likely be needed to motivate adolescents to engage in regular preventive behaviors.

The second is a secondary or tertiary prevention approach to address the dental care needs of patients with existing disease. The scope of care must include viable treatment options. Extraction of badly decayed front teeth without options to replace the missing space is not a cosmetically viable option, especially for individuals with jobs. This could explain why ED dental visitors refuse definitive treatment (e.g., extractions) when offered and instead opt for palliative care. Dental crowns and implants are costly and unlikely to be covered. Medicaid could provide definitive treatment options by intensifying involvement by denturists, trained providers specializing in removable dentures.³² Adults can also be managed with approaches from pediatric dentistry such as application of diamine silver fluoride, interim therapeutic restorations, and permanent molar stainless steel crowns.³³ These less-invasive and less-costly approaches are humane for adult patients, many of whom may have dental fears, and feasible for capitated CCOs. Dental disease could be managed in nondental office settings such as virtual dental homes, in which behavior change interventions could also be implemented to address dietary and hygiene-related risk factors.³⁴

There are 4 main study limitations. First, the scope of the problem appears to be greater in Oregon than nationally and our findings may not be generalizable to other states. Second, our hospital sample was biased toward urban and higher-volume EDs. However, APAC data show similar patterns of ED dental visit frequencies across the entire state (data not shown). Third, our APAC estimates may not generalize to other populations including Medicare beneficiaries and the uninsured. We believe it is unlikely that ED providers vary their treatment by payer type: ED providers have a federal mandate to provide a medical screening examination to all patients regardless of payer status, and providers are often unaware of payer status at the time of treatment. Fourth, the qualitative findings report the knowledge, attitudes, and beliefs of a convenience sample of interviewees. Solutions proposed by interviewees may not be feasible or effective. We addressed this limitation by amalgamating the qualitative study findings into a single multilevel model.

Low-income, socioeconomically vulnerable individuals are at greatest risk for ED dental visits. Our findings underscore the importance of implementing multilevel policies and interventions aimed at addressing this public health problem. Such interventions could involve a 2-pronged strategy that prevents dental disease in high-risk adolescents and provides definitive dental treatment to individuals with dental treatment needs. Future research efforts should be devoted to testing such multilevel interventions.