Opioid Prescribing Following Non-Surgical Root Canal Therapy: Findings from the National Dental Practice-Based Research Network

Abstract

Background:

Despite guidelines discouraging routine opioid use for localized endodontic infections, opioids continue to be prescribed following non-surgical root canal therapy (RCT). This study evaluated opioid prescribing patterns across a broad range of practice settings in the United States by engaging the National Dental Practice-Based Research Network.

Methods:

Practitioners collected standardized clinical, radiographic, and demographic data before and after RCT. The primary outcome was whether opioids were prescribed after treatment. Generalized estimating equations (GEE) multivariable models were used to assess associations of dentist, patient, and clinical characteristics with opioid prescribing, accounting for clustering of patients within the dentist who performed the RCT.

Results:

A total of 153 dentists (104 general dentists, 49 endodontists) enrolled 1,723 patients requiring RCT. Opioids were prescribed to 263 patients (15%). In multivariable models, we found higher odds of opioid prescribing among patients who reported being somewhat to extremely afraid of the RCT (OR=1.41; 95% CI: 1.15, 1.74; p=.001), whose tooth was tender to palpation (OR=1.36; 95% CI: 1.05, 1.77; p=.03), or whose treatment was completed in one visit (OR=1.51; 95% CI: 1.13, 2.02; p=.008). Lower odds of opioid prescribing when a radiolucency was present (OR=0.77; 95% CI: 0.62, 0.96; p=.04).

Conclusions:

Opioid prescribing after RCT was associated with higher odds among patients with dental fear or tenderness to palpation, single-visit treatment, and with lower odds when radiolucency was present.

Practical Implications:

Findings suggest opportunities to reduce opioid use in endodontics through better anxiety management, use of non-opioid analgesics, and targeted efforts in high-prescribing settings.

INTRODUCTION

Although dental opioid prescribing has declined, dentists still account for approximately 8–9% of all opioid prescriptions in the U.S., particularly among adolescents and young adults (1, 2). Despite increased awareness, evolving guidelines, and policy interventions, continued vigilance is needed because non-opioid analgesics such as NSAIDs alone or combined with acetaminophen provide superior pain relief compared with opioids (1, 3, 4). Clinical guidelines discourage routine opioid use for dental pain, recommending NSAIDs as first-line therapy and limiting opioids to short courses for severe cases (1, 5). Nevertheless, opioids remain commonly prescribed, and many prescriptions go unused, leaving excess pills that can be diverted or misused (6), underscoring the public health importance of reducing unnecessary dental opioid use.

Despite guidelines recommending non-opioids as first-line therapy, opioids are still commonly prescribed for RCT, particularly when NSAIDs or acetaminophen are ineffective or contraindicated (7, 8). Surveys show that more than half of dentists would prescribe an opioid for RCT and over 90% for extractions, even though most cases can be managed with definitive treatment and non-opioid analgesics (9). Although overall prescribing has declined, substantial regional and practice-level variation persists (1, 2, 10), highlighting gaps between guidelines and practice and the need for real-world data on factors driving opioid use during RCT.

The present study addresses this need by engaging the National Dental Practice-Based Research Network (“Network”), a nationwide consortium of general and specialty dental practices that conducts research in “real-world” clinical settings (11). Funded by the National Institute of Dental and Craniofacial Research, the PBRN spans all US regions and includes private practices, large group practices, public clinics, and academic centers, enabling a comprehensive assessment of practice variation (11). This infrastructure captures the complexity and range of everyday clinical decision-making, which is often missed in single-site or controlled trial settings (11, 12). The rationale for this study is to address a critical gap in knowledge by identifying how patient-, provider-, and procedural-level factors influence opioid prescribing for non-surgical RCT, an area where detailed clinical data are currently limited.

Within the Network’s PREDICT (Practitioners Engaged in Investigating Clinical Treatment) project, we prospectively evaluated opioid prescribing during RCT among general dentists and endodontists to identify clinical and provider factors that can inform targeted opioid-stewardship interventions. The objectives of this report are twofold: 1) ascertain the frequency of prescribing opioids after RCT; and 2) identify independent predictors, characteristics associated with opioid prescription.

METHODS

This study was conducted in the Network (11, 13, 14). The University of Alabama at Birmingham (Protocol X161121003), the University of Minnesota (Protocol RNI00002290), and the University of Toronto (Protocol 34105) granted human research ethics approval. Network enrollment questionnaires (EQ) were completed by Network dentists, with those indicating performance of RCT eligible for further screening and recruitment; performance of RCT is routine among endodontists and common among general dentists in the Network, as documented in prior National Dental PBRN studies (13, 15). Patient clinical and radiographic data were collected by participating Network dentists. During the study, 153 dentists (104 general dentists and 49 endodontists) used a consecutive enrollment strategy over a 6-month study enrollment period (April to September 2017). Participating dentists were informed that the study evaluated RCT outcomes, including pain, but were not told that opioid prescribing would be an outcome. Each general dentist was asked to enroll a target of 7 patients, and each endodontist to enroll a target of 15 patients over 10 to 14 weeks. Different enrollment targets for general dentists and endodontists reflected differences in RCT volume, and limiting enrollment to a short period was intended to minimize practice burden and improve participation (16). Sample size was based on detecting differences in 6- and 12-month pain outcomes. Study data were collected from patients and practitioners before the RCT and at completion. All data were collected electronically; patients completed forms on a tablet in a private area. The full study protocol, practitioner recruitment flyer, and all data forms used in this study are publicly available (17, 18). This study follows the guidelines for STrengthening the Reporting of OBservational studies in Epidemiology (STROBE), specifically, for cross-sectional studies (see figure and supplemental table).

Figure 1:

STROBE flow diagram of included patients and based on whether dentist prescribed an opioid after completing a non-surgical root canal treatment (RCT)

Patient Eligibility.

Patients were screened for the following eligibility criteria: 18 or 19 years of age or older (depending on the state); anticipate being able to return to the clinic in 12 months (for a long-term follow-up objective); able and willing to provide contact information for one other person with a different telephone number, have internet access and willing to complete questionnaires electronically, willing to provide consent, able to read in English or Spanish, and having a permanent tooth in need of a RCT. In their clinical examination, dentists ascertained the final eligibility requirements for completing RCT: that the tooth had no prior RCT and that only one of the patient’s teeth required RCT. The patient completed a brief form after completing the RCT that queried demographics.

Information from the Dental Practitioner.

Dentist demographic information, whether a specialist or in general practice, practice setting (private practice including whether owner and number of dentists; member of a large group organization (Health Partners [HP] and Permanente Dental Associates [PDA] were specifically queried, and an option for “other” managed care organization not specified); public health or Veterans Affairs clinic, or in an academic setting) were obtained from the EQ. As was which of the six (Western, Midwest, Southwest, South Central, South Atlantic, Northeast) Network administrative/geographic regions the practice was located.

Clinical findings reported by the Dental Practitioner.

Dentists completed a questionnaire prior to treatment that included tooth number, pulpal and periapical diagnostic findings (sensitivity to cold, percussion, biting pressure, and palpation), probing depths, abutments to partial (fixed or removable) denture, proximal contacts, swelling, presence of a draining sinus tract, mobility, and periapical radiolucency. All data were collected electronically. After completing the RCT, practitioners completed a questionnaire on procedures used and findings, which included medications recommended or prescribed after the RCT was completed, and whether it took one or multiple visits to complete the RCT.

Statistical Analysis.

Most variables of interest were categorical, specifically, dichotomous, so proportions were calculated. The “outcome” variable was whether opioids were prescribed after the RCT was completed. The associations of each characteristic with whether opioids were prescribed were first assessed in bivariate analysis (univariable) that adjusted for clustering of patients within a dentist using a generalized estimating equation (GEE). This was implemented using PROC GENMOD in SAS with CORR=EXCH option. All p-values reported, whether from a dichotomous, ordinal, or categorical variable, are based on a score statistic, which approximates a chi-square distribution. Odds ratios (OR) were calculated and used as a measure of the magnitude of association.

Variables with associations with a p<0.1 in bivariate analysis (after adjustment for clustering with GEE) were entered into a domain model (namely, dentist characteristics, patient demographics, psycho-social, clinical domains). The domain models were primarily groups of characteristics known or suspected to be correlated with each other. The best examples are the patient psychosocial measures. They are known to be correlated with each other and associated with pre-RCT pain (19), therefore, these variables were evaluated for their association with higher odds of opioid prescribing. We wanted to identify which psycho-social measures had independent associations with opioid prescription. Dentist characteristics, specifically practice characteristics, differ by Network region. The Network’s two large group member organizations are located in the Midwest (HP) and Western (PDA) regions. These large member practices and academic practices have been found to differ from other practice settings, e.g., private practices. These member groups and academic practices may be higher odds to have current guidelines to follow for a variety of conditions/procedures. Patient demographic characteristics were included as covariates to account for demographic variation, rather than as causal predictors of opioid prescribing. We had no strong a priori expectations regarding clinical findings prior to RCT completion and opioid prescription with one exception. The presence of a draining sinus tract is typically associated with reduced pressure and pain in the affected area and was therefore hypothesized to be associated with lower odds of opioid prescribing.

After all domain models were run, characteristics within each domain that were associated with opioid prescription were identified using a combination of clinical, a priori reasons, and statistics. Then a full model was run, including all characteristics that retained a p<0.1 in domain models. Stepwise procedures in general have severe limitations in identifying and evaluating statistical significance of associations (20, 21), especially when starting from a long list of variables. We limited their use to within domains, guided by a priori knowledge, described above. All p-values presented below and in the tables for associations with prescriptions of opioids have been adjusted for the clustering of patients within the practitioner. Observations with missing values were excluded using complete-case analysis; no imputation was performed. All analyses were conducted using the SASv9.4 statistical package.

RESULTS

A total of 153 dentists initially enrolled 1,860 patients in this study, of whom 137 were excluded due to the following: 68 patients were ineligible based on clinical examination, 6 patients did not provide an email address, 3 patients lost contact before baseline forms were completed, and 60 patients changed their minds and decided to withdraw, leaving 1,723 patients. The following medications were grouped together as having been prescribed opioids: tramadol, hydrocodone, codeine, and oxycodone. The data form requested medications the dentist recommended or prescribed after completing the RCT. Exact dosage was not ascertained. Overall, 263 (15%) of patients were prescribed opioids.

Dentist characteristics.

Details of dentist characteristics are available in a recent publication (22). The mean age of the dentists was 52 years (SD=12), and the median was 55 years (IQR: 42-63). The majority were males (71%, N=68), non-Hispanic (NH) white (70%, N=57), and owners of a private practice (73%, N=69). Fewer were associates/employees of a private practice (11%, N=10), members of a large group practice organization (10%, N=9), in academic (3%, N=3), or public or Federal practice (3%, N=3); practice setting was missing for 2 dentists. All Network regions were represented.

Dentist characteristics and prescription of opioids (Table 1).

Table 1.

Dentist demographics and practice characteristics overall and by whether the dentist prescribed an opioid after completing a non-surgical root canal treatment (RCT)

Dentist characteristics1			Prescribed an opioid				p2	OR3	95% Confidence Interval
	All (N=1,723)		No (N=1,460)		Yes (N=263)
	N	%	N	%	N	%
Endodontist	935	54%	801	55%	134	51%	.4	0.71	0.31, 1.64
Male gender	1,231	71%	1,033	71%	198	75%	.5	0.77	0.36, 1.65
Age (years)							.2	0.80	0.14, 1.13
< 45 years	644	38%	518	36%	126	48%		per 10 years
45 to 54	319	19%	247	17%	72	27%
55 to 64	565	34%	510	36%	55	21%
65 and older	156	9%	146	10%	10	4%
missing n=39
Race-ethnicity (NH: Non-Hispanic)							.6	NA4
NH White	1,181	70%	1,006	70%	175	70%
NH African-American/Black	118	7%	77	5%	41	16%
NH Asian	178	11%	160	11%	18	7%
NH Multi-race/other	149	9%	140	10%	9	4%
Hispanic	54	3%	48	3%	6	2%
missing n=43
Practice setting							NE5
Private practice-owner	1,295	77%	1,067	74%	228	92%
Private practice-associate	161	10%	150	10%	11	4%
Member group	88	5%	79	6%	9	4%
Public health/Government	28	2%	27	2%	1	0%
Academic	112	7%	112	8%	0	0%
missing n=39
Member group or Academic	200	12%	191	13%	9	4%	.003	0.22	0.07, 0.69
Region							.01	NA
Western	167	10%	142	10%	25	10%
Midwest	398	23%	332	23%	66	25%
Southwest	423	25%	398	27%	25	10%
South Central	303	18%	198	14%	105	40%
South Atlantic	248	14%	215	15%	33	13%
Northeast	184	11%	175	12%	9	3%
missing n=0

¹Characteristics from Network enrollment questionnaire. Missing: age n=39; race-ethnic n=43; practice type n=39

²p:p-value from general estimating equations adjusted for clustering of patients within dentist

³OR: Odds ratio for bivariate and ordinal; for bivariate referent is group not listed, e.g., female, practice settings other than group member or Academic

⁴NA: Not applicable – categorical

⁵NE: Not estimable, has a zero cell

Over half of the patients were treated by an endodontist (N=935; 54%), by a male dentist (N=1,231; 71%), by a dentist 45 years or older (N=1,040; 62%), of NH white race-ethnicity (N=1,181; 70%), by an owner of a private practice (N=1,295; 77%), and more were in the Midwest (N=398; 23%) or Southwest (N=423; 25%) regions than the other four regions. The only dentist characteristic associated with prescribing opioids was practice setting and region. Practitioners in large group practices (e.g., HP) or academic practices had lower odds of opioid prescribing (OR=0.22, 95% CI: 0.07, 0.69, p=.003). Opioid prescriptions differed across region (p=.01); higher in the South Central region and lower in the Northeast region. When both practice setting and Network region were analyzed in the same model, both characteristics were significantly related to opioid prescriptions (p=.02; Supplemental Table 1).

Patient demographics and opioid prescription (Table 2).

Table 2.

Patient demographics overall and by whether the dentist prescribed an opioid after completing a non-surgical root canal treatment (RCT)

Pateint characteristics1			Prescribed an opioid				p2	OR3	95% Confidence Interval
	All (N=1,723)		No (N=1,460)		Yes (N=263)
	N	%	N	%	N	%
Male gender	701	41%	609	42%	92	35%	.4	0.93	0.78, 1.11
missing n=8
Age (years)							.07	0.94	0.89, 1.00
<35 years	399	23%	306	21%	93	36%		per 10 years
35 to 44	293	17%	254	18%	39	15%
45 to 54	347	20%	299	21%	48	19%
55 to 64	367	22%	317	22%	50	19%
65 and older	293	17%	265	18%	28	11%
missing n=24
Race-ethnicity (NH: Non-Hispanic)							.08	NA4	Categorical
NH White	1,165	69%	1,002	70%	163	64%
NH African-American/Black	205	12%	154	11%	51	20%
NH Asian	57	3%	51	4%	6	2%
NH Multi-race/other	69	4%	59	4%	10	4%
Hispanic	188	11%	162	11%	26	10%
missing n=39
Education							.8	NA	Categorical
<= High School/GED	313	19%	254	18%	59	23%
Some college/Associate’s Degree	596	35%	506	36%	90	35%
Bachelor's degree	445	26%	379	27%	66	26%
Graduate degree	328	20%	286	20%	42	16%
missing n = 41
Patient has any dental Insurance	1,302	76%	1,092	75%	210	80%	.4	0.93	0.78, 1.10

¹Missing: gender n=8; age n=24; race-ethnic n=39; education n=41

²p:p-value from general estimating equations adjusted for clustering of patients within dentist

³OR: Odds ratio for bivariate and ordinal; for bivariate referent is group not listed, e.g., female

⁴NA: Not applicable – categorical

The only patient demographics associated with the prescription of opioids were age and race-ethnicity. Prescriptions odds decreased slightly with age, 6% for every 10 years (OR=0.94; p=.07). Prescription of opioids varied with patients’ race-ethnicity (p=.08). Within-group proportions of patients prescribed opioids were higher among NH African American patients (25%) and lower among NH Asian patients (11%). When analyzed jointly, neither retained p<.1 (Supplemental Table 1).

Patient collected data prior to RCT (Table 3).

Table 3.

Patient pre-Root Canal Treatment (RCT) pain and psychosocial characteristics overall and by whether patient was prescribed an opioid after RCT

Characteristics1			Prescribed an opioid				p2	Odds Ratio	95% Confidence Interval
	All (N=1,723)		No (N=1,460)		Yes (N=263)
	N	%	N	%	N	%
Pain measures
Severe tooth pain at time of RCT	367	21%	283	19%	84	32%	.02	1.29	1.01, 1.65
Screen positive for TMD	709	41%	586	40%	123	47%	.13	1.15	0.95, 1.38
Psycho-social measures
Expect very good Tx outcome	1,290	75%	1,098	75%	192	73%	.3	0.91	0.75, 1.10
Some to extremely afraid	576	33%	461	32%	115	44%	<.001	1.44	1.17, 1.76
Pain catastrophizing score of 5 or higher (on 2-10 scale)	567	34%	458	32%	109	42%	.03	1.25	1.00, 1.57
Anxiety score of 3 or higher (on 2-8 scale)	645	38%	530	37%	115	44%	.5	1.06	0.87, 1.30
Depression score of 3 or higher (on 2-8 scale)	476	28%	381	26%	95	36%	.08	1.21	0.97, 1.51
Stress score of 8 or higher (on 4-20 scale)	922	55%	760	53%	162	63%	.5	1.05	0.90,1.23

¹Missing: Tooth pain at time of RCT n=1, treatment expectations n=2, treatment fear n=1, pain catastrophizing n=33, anxiety n=27, depression n=17, stress n=40

²p:p-value from general estimating equations adjusted for clustering of patients within dentist

Only 21% patients reported severe tooth pain immediately prior to RCT, and 41% screened positive for TMD. Severe tooth pain at time of RCT, but not screening positive for TMD, was associated with prescription of opioids (OR=1.29; p=.02). Of the 6 psycho-social characteristics screened for (means given in Mungia et al., 2025) (22), 3 had an association with prescription of opioids: dental fear (p<.001), pain catastrophizing (p=.03), and depression (p=.08). When assessed jointly, only dental fear retained p<.1 (was p=.002). (Supplemental Table 1).

Clinical characteristics, overall and opioid prescription (Table 4).

Table 4.

Dentist pre-Root Canal Treatment (RCT) assessment overall and by whether patient was prescribed an opioid after RCT

Characteristics1			Prescribed an opioid				p2	Odds Ratio	95% Confidence Interval
	All (N=1,723)		No (N=1,460)		Yes (N=263)
	N	%	N	%	N	%
RCT Completed in one visit	1,407	82%	1,160	79%	247	94%	.002	1.56	1.21, 2.01
Maxillary	911	53%	772	53%	139	53%	.2	0.89	0.74, 1.08
Molar	1,037	60%	870	60%	167	64%	.14	1.17	0.95, 1.45
Diagnostic findings (Responds/tender to:)
percussion	1,187	69%	998	69%	189	72%	.4	1.08	0.89, 1.31
biting pressure	1,108	65%	924	64%	184	70%	.3	1.13	0.90, 1.42
cold	917	53%	754	52%	163	62%	.4	1.09	0.91, 1.32
palpation	504	29%	416	29%	88	34%	.04	1.28	1.01, 1.62
Other clinical characteristics
Any proximal contact	1,654	96%	1,400	96%	254	97%	.9	0.96	0.58, 1.59
Exhibits radiolucency	768	45%	668	46%	100	38%	.06	0.84	0.71, 1.00
Any movement	405	24%	374	26%	31	12%	.2	0.81	0.61, 1.06
Probing depth>= 5mm	181	11%	155	11%	26	10%	.8	1.04	0.80, 1.36
Swelling present	167	10%	143	10%	24	9%	.7	0.94	0.68, 1.31
Draining sinus tract	121	7%	106	7%	15	6%	.08	0.75	0.51, 1..11
Tooth abutment to partial	110	6%	100	7%	10	4%	.3	0.80	0.56, 1.17

¹Missing: Sensitivity to cold n=4, percussion n=10, to biting pressure n=8, to palpation n=10, probing depth n=23, abutment n=14, proximal contact n=3, swelling n=10, draining sinus tract n=11, any mobility n=13, radiolucency n=13

²p:p-value from general estimating equations adjusted for clustering of patients within dentist

The RCT was completed in one visit for 82% of the patients. These patients were higher odds to have been prescribed an opioid (OR=1.56; p=.002) than patients whose RCT took multiple visits to complete. Over half of the teeth were maxillary (N=911, 53%) and molars (N=1,037; 60%); neither characteristic was associated with prescription of opioids. Most treated teeth were tender to percussion (N=1,187; 69%), to biting pressure (N=1,108; 65%), and sensitive to cold (N=917; 53%), while fewer were tender to palpation (N=504; 29%). Only tenderness to palpation was associated with prescription of opioids (OR=1.28; p=.04). Virtually all treated teeth had a proximal contact (N=1,654; 96%), slightly less than half exhibited radiolucency (N=768; 45%), about a quarter had some type of horizontal movement (N=405; 24%), fewer had probing depths of 5 mm or more (N=181; 11%), swelling present (N=167; 10%), a draining sinus tract (N=121; 7%), or on a tooth that was an abutment to a partial (N=110; 6%). Radiolucency (OR=0.84; p=.06) and presence of a draining sinus tract (OR=0.75; p=.08) both had an inverse association with prescription of opioids. When all 4 clinical characteristics (completed the RCT in one visit, tender to palpation, radiolucency, draining sinus tract) were assessed jointly, all except presence of a draining sinus tract retained a p<.1.

Multivariate associations (Table 5).

Table 5.

Adjusted associations with dentist prescribing opioids after completing a non-surgical root canal treatment (RCT)

	Bivariate1			Full2 (final) model
Characteristics	Odds Ratio	95% Confidence Interval	p3	Odds Ratio	95% Confidence Interval	p4
Practitioner
Member group or Academic practice	0.22	0.07, 0.69	.003	0.72	0.17, 3.02	.7
Network geographic region	categorical		.01	Categorical		.9
Patient data
Patient race-ethnic group	categorical		.08	Categorical		.3
Severe pain at RCT	1.29	1.01, 1.65	.02	1.23	0.97, 1.57	.08
Somewhat to extremely afraid of RCT	1.44	1.17, 1.76	<.001	1.41	1.15, 1.74	.001
Clinical
RCT completed in one visit	1.56	1.21, 2.01	.002	1.51	1.13, 2.02	.008
Tender to palpation	1.28	1.01, 1.62	.04	1.36	1.05, 1.77	.03
Exhibits radiolucency	0.84	071, 1.00	.06	0.77	0.62, 0.96	.04

¹All characteristics that had a p<.1 when adjusted only for clustering of patients within dentist, and retained a p<.1 when adjusted for other characteristics in its domain

²Includes all characteristics listed under bivariate analysis; analysis based on 1,625 patients (94% of 1,723) with no missing data forvariables included. 98 of 1,723

³p: p-value from generalized estimating equations (GEE) adjusted only for clustering of patients within dentist

⁴p: p-value from generalized estimating equations (GEE) adjusted for clustering of patients within dentist and all characteristics listed

Dentist, patient, and clinical characteristics that were associated with prescribing opioids (at p<0.1) described above were entered into a full model comprising practitioner, patient, and clinical information. For patient demographics, when race-ethnicity (p=.08 bivariate) and patient age (p=.07 bivariate) were assessed jointly, patient race-ethnicity had a p=.1 (but not <.1) and patient age had a p=.2. For the full model we used patient race-ethnicity. Four characteristics were associated with opioid prescription at alpha of 5% (p<.05). Three characteristics were associated with increased odds of opioid prescription: dental fear (OR=1.41; p=.0001), RCT completed in one visit (OR=1.51; p=.008), and tooth being tender to palpation (OR=1.36; p=.03). The tooth exhibiting radiolucency was associated with reduced odds of opioid prescription (OR=0.77; p=.04).

We ran a second model including patient age and not race-ethnicity to evaluate the robustness of our findings (Supplemental Table 2). The same four characteristics were associated with opioid prescriptions at a 5% alpha (p<.05) level.

DISCUSSION

In this practice-based RCT study, opioids were prescribed in 15% of cases, a substantial minority and lower than historical rates reported for more invasive dental procedures such as extractions. (10, 23), it still reflects ongoing use in endodontic settings. Although there is no established “optimal” prescribing proportion, current guidelines emphasize minimizing opioid use and prioritizing non-opioid analgesics whenever possible. Opioid prescribing was associated with dental fear, tenderness to palpation, and single-visit RCT, and was lower when a periapical radiolucency was present, consistent with differences between acute and chronic pain presentations (24-26).

Dental anxiety was associated with opioid prescribing, consistent with prior evidence that patient fear and pain catastrophizing influence expectations and prescribing behavior (27, 28). This suggests a psychosocial component to opioid prescribing, in which fear and anxiety, in addition to pain, influence both patient expectations and clinician decisions (29). It may be that improved communication and patient education about pain management could reduce the reflex to prescribe opioids in these situations (23). For example, discussing the efficacy of NSAIDs and setting expectations that some post-treatment soreness is normal might reassure anxious patients without resorting to opioids (23). Dentists should be encouraged to manage dental fear with empathy and guideline-driven pain control strategies (23).

Clinical presentation also influenced prescribing in ways consistent with pain physiology. Teeth that were tender to palpation and associated with severe pain were associated with higher odds of opioid prescribing, consistent with clinicians anticipating greater post-treatment pain (30, 31). In contrast, radiolucent lesions typically represent chronic infections with minimal symptoms, where dentists were appropriately lower odds to prescribe opioids (31). Although this suggests a measure of clinical prudence, it is important to note that even acute dental pain can often be effectively managed with NSAIDs and local measures (6). Evidence consistently shows that ibuprofen and acetaminophen combinations provide analgesia superior to opioids with fewer risks (6). This highlights opportunities to further reduce opioid use by emphasizing non-opioid first-line management, even for severe pain.

Although not retained in the final model, lower opioid prescribing in academic and large group settings suggests that institutional policies and standardized protocols may influence prescribing behavior, consistent with prior research (10). Cultural norms, patient expectations, and prescribing culture may explain regional differences. Such findings highlight the need for tailored interventions. Severe tooth pain at time of RCT was associated with opioid prescribing in bivariate analysis, but not in final model. Unexpectedly, TMD was not associated with opioid prescribing even in bivariate analysis.

This study has limitations. Data were collected in 2017, before the ADA’s 2018 opioid policy and increased public awareness, so prescribing patterns may have shifted; however, these data provide an important baseline for current opioid stewardship efforts. Because Network dentists volunteer for research, findings may not fully represent all U.S. dentists. Dentists knew the study evaluated RCT outcomes, so behavior may have been influenced by observation, even though opioid prescribing was not a stated focus. We recorded whether an opioid was prescribed but not dosage or duration, limiting assessment of prescribing intensity and potential misuse. Pain and fear were self-reported at baseline only, and postoperative experiences were not measured. Other unmeasured factors (e.g., provider beliefs, patient history, prior opioid use) may have influenced prescribing. Interim prescribing was not captured for multi-visit RCTs, which may have led to underestimation, particularly among general dentists. We also did not measure dentists’ baseline opioid-prescribing tendencies; however, prior National Dental PBRN surveys show wide variation in opioid use and risk-mitigation practices that may have influenced prescribing (32-34). Despite these limitations, the study’s large, diverse sample and real-world design provide valuable insights into opioid use in endodontic care.

CONCLUSION

Opioid prescribing after RCT remains a concern in the U.S. and is associated with dental fear, acute pain, and specific clinical findings. Prescribing showed higher odds in single-visit and high-fear cases and lower odds in chronic, low-symptom conditions. Emphasizing NSAIDs/acetaminophen, anxiety management, and standardized protocols may help reduce unnecessary opioid use and improve stewardship in dental care.