Opioid Prescribing for Surgical Dental Procedures in Dental Clinics of Military Treatment Facilities

INTRODUCTION

Drug overdose deaths are increasing in number: there were 70,237 U.S. drug overdose deaths in 2017 compared to 63,632 in 2016; the latter a 21.4% increase from 2015 and two-thirds involved an opioid.¹ Non-fatal overdoses far exceed the number of overdose deaths.² To develop a public health response to opioid-related morbidity, researchers and policymakers are looking upstream to prevent first opioid exposure, long-term opioid use, and diversion. The first exposure to opioids is usually from a prescription or prescription diversion and larger quantities, higher dosages and longer duration play a significant role in later opioid use.^3–8

Dental clinicians’ prescribing of opioids appears to account for between 5% and 10% of all opioid prescriptions.^9–13 Many pills from dental opioid prescriptions go unused ¹⁴ and the necessity for opioids rather than other analgesics for dental and other surgeries has been questioned in recent studies.^15,16 Supporting this contention, prescriptions written by U.S. dentists for opioids are 37 times greater than the proportion written by English dentists.¹⁷ Studies have found substantial variation in opioid prescribing between different geographic regions or states and this variation has raised concerns about overprescribing; however, there is no literature on variation among dental providers.¹⁸

This study is the first to focus on dental clinicians’ prescribing of opioids for service members in the US Military Health System (MHS). The MHS possesses unique features for a dental study: there is a centralized computerized pharmacy record; soldiers are universally covered for medical and dental care; there are regular dental exams to prevent dental emergencies during deployment; a dental encounter record is part of the medical record system, and centralized policies are promulgated, including following CDC guidelines on opioid prescribing. These features permit this study to compare military dental practices with published findings on civilian practices and to address limitations we identified in prior dental literature related to matching opioid prescriptions to dental encounters, assigning dental opioids to surgical versus non-surgical encounters, and selected samples that may not be generalizable to the studied population.^10,19,20

The current study provides new information on this urgent and important public health issue. Using recent data, it updates estimates of prior dental studies conducted in civilian settings, separately examines opioid prescribing for dental extractions versus other surgical or invasive procedures, examines differences between younger and older age-group soldiers, and differences between early and later time periods. We advance knowledge on variation in dental prescribing by also comparing opioid prescribing rates among parent facilities of dental clinics.

METHODS

Design and Setting:

This retrospective secondary data analysis examines dental encounter records and filled opioid prescriptions from October 2008 to Sept 2017 for a sample of US Army soldiers who comprise the SUPIC study cohort. The Dental Program of the MHS operates 270 dental clinics in the US, US territories, and military bases in Europe, the Pacific, and Asia for active duty service members only.

Participants:

The SUPIC study is a longitudinal study of all active duty and activated reserve component Army soldiers (n=865,460) who returned from a deployment as part of Operation Enduring Freedom/Operation Iraqi Freedom/Operation New Dawn (OEF/OIF/OND) between fiscal years (October to September) 2008 and 2014.²¹ No data on military family members were obtained. Any sample member who had a Dental Program encounter during the study window, whether before or after their deployment, was eligible for inclusion. We defined the study’s participants as sample soldiers with one or more procedures performed by dentists that were surgical or invasive (hereafter, surgical) and therefore might be associated with an opioid prescription. Our methods to select procedure codes are described in the Appendix Methods. All procedures performed by non-dentists, providers who do not prescribe opioids, were excluded. We rolled up all included procedure records to the encounter level for matching to prescription claims and for analysis. We examined all the prescription claims of the sample and selected for inclusion opioid prescriptions using the Centers for Disease Control algorithm modified to remove opioids that are used to treat addiction disorders.^22,23 (See Appendix Methods)

The study protocol was reviewed and approved by the Institutional Review Boards (IRB) of Brandeis University and the Uniformed Services University/Department of Defense.

Outcomes:

Our primary outcome was an encounter with opioid exposure. We specified a 60-day window around each surgical encounter and matched all opioid prescriptions written by a dentist on the date of a surgical encounter, or pre-surgery (within 30 days prior), or post-surgery (within 30 days after surgery). A dental encounter with one or more pre-or post-surgery opioid prescriptions was classified as an opioid-exposed encounter. A second measure of opioid exposure was based on total morphine milligram equivalents (MME) dispensed for the dental encounter during the 60-day window (see Appendix Methods)²³ with dental encounters classified as more than 100 MME (yes/no), an equivalent to the typical 5-day supply found in the literature. To further characterize opioid exposure, we examined mean MME, mean days-supply, and frequency by opioid subclass for young and older patient groups.

Covariates:

We defined covariates at two levels, the encounter level and the facility level. The encounter level variables were: types of surgical procedure (endodontic, periodontic, implant, oral & maxillofacial excluding extraction, dental extraction, emergency palliative care), type of dentist (general dentist vs specialist), patient age group at time of encounter (under 26, 26 and older), military rank (junior enlisted, senior enlisted, officer, warrant officer), gender, fiscal year of encounter, and patient’s opioid status at time of encounter (naïve yes/no defined as no filled opioid prescription in the past 90 days). We identified the ‘parent’ military treatment facility for each dental clinic and aggregated encounters to the facility level because of insufficient sample size of most dental clinics. We defined facility-level variables in two groups. First were facility characteristics: region or outside the continental US, facility type (medical center, hospital, clinic or other), and percentage of dental surgeries for young patients (under age 26). Second, we computed dental practice variables at the facility-level: proportion of surgeries for extractions, proportion of surgeries for periodontic procedures, and the proportion of dentists that were specialists.

Statistical methods:

Descriptive analysis was conducted on dental surgical encounters for the full study window (i.e., Main Study). We present the percentage of surgical encounters that resulted in opioid exposure and greater than 100 MME for patient subgroups and types of encounters. We separately examined dental extraction encounters and all other surgical encounters as some prior literature has focused on extractions. To address policymaker interest, we compared the proportion of all opioids prescribed for dental surgical procedures for early (2008–2013) and recent (2014–2017) time periods. To address our core question about the degree of variation between facilities, we nested dental encounters within specific facilities to examine variation in prescribing rates between facilities. For these analyses, we restricted the data to October 2014 to September 2017 (Facility study). Dentists in the military’s Dental Program move among facilities every three years or so and by restricting study years we attempted to stabilize the pool of dentists at each facility.

We first calculated the observed opioid prescribing rates for dental extractions and other surgical encounters; we present these data for the 30 facilities with the largest volume of dental encounters. We then performed multivariate analysis to estimate the variation in opioid prescribing within and among facilities due to encounter-level variables (within facility) or context/facility-level variables (between facilities). We used a two-level multilevel model, using the STATA command melogit estat ICC. In the Appendix Methods we present supplementary analysis that compares the observed to the expected opioid prescribing rate for facilities with at least 25 encounters which takes into account the facility case mix.

Statistical significance was defined at the p<.01 level. Analyses were performed using SAS 9.4 and STATA 15 procedures.

RESULTS

Dental Opioids Overview

Among all soldiers eligible for the study (n=865,460), nearly all (n=819,453) had at least one military Dental Program encounter and 317,128 dental patients (38.7%) had at least one surgical encounter during the study window of October 2008 to September, 2017. There were 743,459 surgical encounters, thus each dental patient with a surgical encounter had on average 2.3 surgical encounters. There were 4,472 dentists who performed these surgical procedures (Figure 1, Flow Diagram).

Figure 1:

Flow Diagrams: Main study and facility substudy

There were 4,109,048 opioid prescriptions filled by the sample for any purpose during the study window, of which 342,595 prescriptions were linked to a Dental Program surgical or non-surgical encounter (8.3%). Among the 580,599 soldiers who received an opioid prescription, 195,035 (33.6%) received at least one dental opioid prescription and 88.7% of dental prescriptions were for surgical procedures (Figure 1).

To compare time trends we examined changes within two age groups (Table 1). We hypothesized apriori that younger patients were more likely to have opioid prescriptions from dental encounters than older patients but we note that, because of the longitudinal cohort design, the sample aged over the study. Of all opioid prescriptions filled by younger patients, 16.5% were written by dentists in the earlier period compared to 6.9% of older patients. The percentage of opioids written by the Dental program declined for both age groups in the later period (p <.001).

Table 1.

Percent of total opioid prescriptions attributable to the Dental Program in study sample

Sample and time period a		Opioid prescriptions for study sample
		Total prescriptions (A)	Dental program prescriptionsc
		N	N	% of total (A)
Under age 26, all years		848,760	138,085	16.3%
2009–2014b		780,188	129,037	16.5%
2015–2017		59,524	9,048	15.2%
Age 26+, all years		3,260,288	204,510	6.3%
2009–2014b		2,208,011	153,032	6.9%
2015–2017		1,062,277	51,478	4.9%
All ages, all years		4,109,048	342,595	8.3%
2009–2014		2,988,199	282,069	9.4%
2015–2017		1,120,849	60,526	5.4%

Notes:

^aYears are fiscal years; i.e., 2009 starts October 1, 2008.

^bA multiple logistic regression on proportion of opioid prescriptions found that the indicator variables for age group and time period and the interaction term of agegroup * timeperiod, were each significant, p<.001.

^cIncludes surgical and non-surgical dental opioid prescriptions.

Table 2 presents a description of all surgical encounters for the dental sample and the rate of opioid prescribing. One-third of encounters were made by patients under the age of 26, the vast majority of encounters were among male and enlisted patients, nearly one-half were among racial/ethnic groups other than white, and over 80% were opioid naïve (no opioid prescription fill in the prior 90 days) at the time of the encounter. The most frequent types of procedures were endodontic and dental extraction. Eighty-four percent (84%) of the encounters were to general dentists rather than specialists, and nearly one-half of the encounters were at dental clinics associated with a medium-size military treatment facility.

Table 2.

Characteristics of dental encounters by opioid exposure, October 2008 – September 2017

Encounter Characteristic	# of surgical encounters	% of surgical encounters (SE)	With opioid prescription (OP)		With total opioid dose > 100 MME
			N	% of SE	N	% of OP
Total Surgical Encounters a	743459	100.0	272718	36.7	138346	50.7
Age group
26+	471739	63.5	161323	34.2	76633	47.5
<26	271720	36.6	111395	41.0	61713	55.4
Sex
Female	101651	13.7	35882	35.3	17445	48.6
Male	641808	86.3	236836	36.9	120901	51.0
Rank
Enlisted	653126	87.8	245096	37.5	125473	51.2
Officer	90333	12.2	27622	30.6	12873	46.6
Race Ethnicity
Non-Hispanic, Asian or Pacific Islander	79388	10.7	29915	37.7	15411	51.5
Non-Hispanic, Black	166843	22.4	62414	37.4	31642	50.7
Hispanic	80796	10.9	32176	39.8	16796	52.2
Non-Hispanic, White	401485	54.0	142986	35.6	71982	50.3
American Indian, Alaskan Native or other	14947	2.0	5227	35.0	2515	48.1
Componentb
Active duty	565551	90.5	216329	37.1	11,308	51.0
National Guard	38443	6.1	9749	25.3	4195	43.0
Reserve	20559	3.3	5686	27.6	2482	43.7
Patient opioid status time of encounterc
Naïve	502941	80.5	169157	33.6	86673	51.2
Not naïve	121735	19.5	62630	51.4	30323	48.4
Type of surgical procedure (not mutually exclusive)d
Endodontic	357189	48.0	38118	10.7	9812	25.7
Periodontic	56057	7.5	41485	74.0	19268	46.4
Implant	46550	6.3	18976	40.8	10533	55.5
Oral & Maxillofacial (excl. extraction)	46977	6.3	24860	52.9	13626	55.0
Dental Extraction	203559	27.4	165757	81.4	96192	58.0
Palliative Emergency	78019	10.5	14997	19.2	5148	34.3
By Provider Type
Oral Surgeon	50226	6.8	43270	86.2	30362	70.2
Periodontist	17839	2.4	13671	76.6	5362	39.2
Prosthodontist	11090	1.5	1033	9.3	393	38.0
Orthodontist	1066	0.1	143	13.4	65	45.5
Oral Pathologist	689	0.1	264	38.3	21	7.9
Endodontist	36738	4.9	7905	21.5	1310	16.6
Dental Officer General	624319	84.0	206034	33.0	100688	48.9
Pedodontist	1492	0.2	398	26.7	145	36.4
By Opioid Subclass
Codeine	17625	6.5	17625	100.0	73	0.04
Hydrocodone	68352	25.1	68352	100.0	12622	18.5
Oxycodone	183888	67.4	183888	100.0	123355	67.1
Tramadol	1912	1.0	1912	100.0	880	46.0
Other	941	< 1.0	941	100.0	173	18.4
Facility Service Branch
Army	633913	85.3	248415	39.2	127014	51.1
Other branch	20815	2.8	7403	35.6	3759	50.8
Facility Typeassociated with dental clinic
Medical Center (largest)	185549	25.0	80050	43.1	47453	59.3
Hospital (medium)	368098	49.5	140449	38.2	74473	53.0
Clinic and other (smallest)	189812	25.5	52219	27.5	16420	31.4
Facility Regione
TRICARE OCONUS	42544	5.7	17878	42.0	5923	33.1
TRICARE North	155063	20.9	53137	34.3	21547	40.5
TRICARE South	255779	34.4	94401	36.9	49903	52.9
TRICARE West	215343	29.0	95120	44.2	54951	57.8
Other / Unknown	74730	10.1	12182	16.3	6022	49.4

Notes:

^a.Note: number of encounters with opioids is less than total number of dental opioids (Table 1) because each encounter may be associated with more than one opioid prescription.

^b.Total includes 123 encounters where component at time of the encounter was not recorded.

^c.Opioid naïve defined as no opioid prescription fills for any purpose in the 90 days prior to the dental encounter.

^d.Includes palliative emergency and invasive procedures that are not surgical but we hypothesized would account for some opioid prescriptions.

^e.States included in TRICARE regions: North= Connecticut, Delaware, the District of Columbia, Illinois, Indiana, Kentucky, Maine, Maryland, Massachusetts, Michigan, New Hampshire, New Jersey, New York, North Carolina, Ohio, Pennsylvania, Rhode Island, Vermont, Virginia, West Virginia, Wisconsin and portions of Iowa (Rock Island Arsenal area), Missouri (St. Louis area) and Tennessee (Ft. Campbell area); South = Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, Oklahoma, South Carolina, Tennessee (excluding the Ft. Campbell area) and Texas (excluding the El Paso area); West = Alaska, Arizona, California, Colorado, Hawaii, Idaho, Iowa (excluding Rock Island Arsenal area), Kansas, Minnesota, Missouri (except the St. Louis area), Montana, Nebraska, Nevada, New Mexico, North Dakota, Oregon, South Dakota, Texas (the southwestern corner, including El Paso), Utah, Washington and Wyoming; OCONUS (outside continental US) = Hawaii, Guam, Puerto Rico, Europe, Asia, Pacific

All chi-square values for crosstabulations of each characteristic by opioid prescription and each characteristic by dose > 100 MME given an opioid prescription, are significant at p < .001

More than one-third of surgical encounters (36.7%) resulted in an opioid fill and for one-sixth of surgical encounters (18.6%) the total opioid dose was over 100 MME (Table 2). The percentage of encounters with an opioid and the percentage with higher total dose was larger among young, enlisted, Hispanic soldiers, and for periodontic and dental extraction procedures (all comparisons p < .01). Among these dental prescriptions, 83.1% were written on the same day as a surgical procedure, 13.8% written within 30 days before the surgery, and 3.1% written within 30 days after surgery date (data not on table).

Hydrocodone (25.1%) and oxycodone (67.4%) were the most frequent opioid prescriptions. The median opioid dose was 120 MME for dental extraction encounters and 90 MME for non-extraction surgical encounters, and the median days-supply was 3 (Table 3). Only 4.7% of prescriptions was for 1 day-supply; 24.0% was for 2 days, and 6.3% was for 21 or more days-supply of opioids. Younger soldiers received a higher mean MME than older soldiers for dental extractions, and mean MME for all dental extraction encounters were higher than other dental surgery procedures (Table 3).

Table 3.

Percent distribution on opioid days-supply and average opioid dose per surgical encounter with opioid^a, October 2008-September 2017

	Non-extract encounters			Encounters with extractions			All surgical encounters
	26 & older	Under 26	All	26 & older	Under 26	All	All ages
Sample n	81,224	25,737	106,961	80,099	85,658	165,757	272,718
Total days-supply	Percent distribution
1 day	5.6	7.2	6.0	3.8	3.2	3.7	4.7
2	26.5	28.2	26.9	22.6	21.7	22.1	24.0
3	29.3	26.3	28.6	32.1	30.5	31.3	30.2
4	11.1	10.4	10.7	11.9	13.5	12.7	12.0
5	10.0	8.6	9.9	11.6	11.5	11.6	10.7
6–10 days	5.1	4.8	5.0	5.0	5.0	4.5	4.9
11–20	3.2	3.0	3.1	2.7	2.6	2.6	2.8
21 or more	8.3	9.6	8.6	7.5	8.2	7.9	8.2
Total MME	Average morphine milligram equivalents
Mean (SD)b	109 (76)	104 (74)	108 (76)	126 (69)	134 (59)	131 (64)	121 (70)
Median	90	90	90	113	150	120	105
IQR (low, high)	75, 150	60, 120	67, 150	80, 150	90, 150	90 150	75, 150
Total MME > 100, % yes	40.5	36.0	36.2	54.6	61.2	58.0	50.7

^aAnalysis based on opioid prescriptions that were written on date of surgical encounter or either pre-surgery (within 30 days prior), or post-surgery (within 30 days after surgery). If multiple prescriptions were on the same date or within this 60 day window, days-supply and total MME were summed for all prescriptions.

^bT-test of difference in means 26 and older relative to under age 26 for non-extraction procedures, unequal variances = 9.3 p < 0.0001; T-test for extraction procedures, unequal variances = −26.3, p<0.001

IQR = interquartile range; the values reported are for the 25th percentile and 75^th percentile

MME = morphine milligram equivalents

Facility-specific analysis: October 2014-September 2017

Facility-specific observed prescribing percentages varied by a factor of 30.1 (87.5% vs 2.9%) for facilities with at least 25 observations in the study. Figure 2 shows the difference in prescribing rates among the 30 facilities with the largest volume of surgical procedures, stratified by encounters for extractions vs other surgical procedure. The range of prescribing rates for dental extraction was 60% to 94% and for other surgical encounters was 10% to 29%. In supplemental analysis described in the Appendix Methods, we describe the expected prescribing percentage for each facility with at least 25 observations. Among these facilities, the range remained large (64.7% to 19.0%) after we adjusted for case-mix.

Figure 2.

Percent of dental surgical encounters (non-extraction, top panel; extraction, bottom panel) with opioid prescriptions by facility: top 30 of 130 facilities by encounter volume, October 2014-September 2017

Note: Observed percentage of encounters associated with dental opioid prescription on date of encounter, within 30 days before the surgery, or within 30 days after the surgery. Facilities numbered by total number of surgical encounters (1 = highest). Restricted to top 30 facilities by encounter volume; minimum volume of encounters = 1,968.

In the multilevel multinomial models, the facility-level (Level 2) variables were significant predictors of opioid prescribing after controlling for encounter (Level 1) factors (see Appendix Methods). The multilevel results also showed that facility-level regional variables and facility-level variables that captured the clinical practice behavior explained 80% of the between-facility variation.

DISCUSSION

Clinicians’ decisions regarding the prescription of opioids require a balance between the pain-reducing benefits and the possible adverse short-term (e.g. constipation) and long-term (e.g. habituation) risks. Recommendations of respected organizations urge caution in prescribing opioids. In this context, our finding that some military dental clinics prescribe opioids at a significantly lower rate than other clinics suggests a path to follow. In this retrospective study, we showed facility-specific opioid prescribing percentages varied by a factor of 30, and in supplemental analysis, that dentists at 11 of the 30 largest facilities prescribed at a rate 4 percentage points higher than expected given their case mix, and dentists at nine of these facilities had a rate 4 percentage points lower than expected. Based on this variation among dental clinics, it would be reasonable for dental leadership to compare clinical practices among facilities and attempt to identify the basis for lower reliance on opioid analgesics.

Our study provides comparison with studies in civilian populations. Opioid prescriptions were filled for 36.7% of dental encounters; 50.1% of opioid encounters received more than 100 MME. Dental care accounted for a substantial proportion of all opioid prescriptions in the population, particularly among young patients where it was 16.5% of all opioids in the period October, 2008 to September 2014 and 13.2% in the period starting October, 2014 to September, 2017. A unique aspect of military dentistry is that service members receive a dental readiness classification based on annual exams, thus they had much higher access to dental care than the civilian population. Further, the military pays for all dental care for active duty and activity service members to avoid emergency care and morbidity. In our sample, 4.3% of records had oral conditions classified as expected to result in dental emergencies within 12 months if not treated, 56.8% had oral conditions that if not treated had the potential, but were not expected, to result in dental emergencies within 12 months. This percentage would be higher among service members just starting a military care, reflecting the restricted access to dental care in the US.

The higher rate of opioid prescribing to the under-26 sample (42% versus 34%) is noteworthy, given dental care is often the source of first prescribing in this age group and has been found to increase risk of subsequent opioid use and misuse.¹⁹ Eighty percent (80%) of our sample had not filled opioids for any reason in the prior 90 days. The decrease in percent of dental-to-overall opioid prescribing within age group over the study timeframe parallels a decrease seen in the civilian world²⁴ as the overall dental community learned to manage dental pain through non-opioid medications. Of note, dental opioid prescriptions in one Canadian province during the 2014–2017 timeframe were only 3.8% of total opioid prescriptions,²⁵ lower than what we report.

Dental extractions, 27.4% of surgical encounters, accounted for a disproportionate share of opioid-exposure. The rate of opioid exposure for extraction procedures in military clinics (81%) matches the rate (80%) in a civilian study.²⁶ For patients age 26 and over and under 26, dental extractions accounted for 43.3% and 73.6%, respectively, of dental opioids. Of note, the American Association of Oral and Maxillofacial Surgeons recommends NSAID’s as the first-line analgesic for extractions.²⁷

The median days-supply in this study, estimated at 3, matched civilian studies.¹⁰ Military dentists were more likely than their civilian counterparts to prescribe oxycodone (67% military; 11% civilian), and less likely to prescribe hydrocodone (25% military; 65% of civilian dental opioid prescriptions), codeine (7% military; 14% civilian), and tramadol (1% military; 5% civilian).¹⁰ Expert guidelines do not recommend one short-acting opioid over another ²⁸ and a study found little difference in pain relief between oxycodone/acetaminophen and hydrocodone/acetaminophen.²⁹

We found that younger age, enlisted members, Hispanic patients, and males were more likely than other soldiers to receive an opioid prescription. Contrariwise, one study in the civilian literature found Hispanics less likely than Whites and Blacks more likely than Whites to receive opioid prescriptions;³⁰ while another separate study found no difference between race/ethnic groups.²

Geographic variation in overall prescribing of opioids is well known: the mean per capita MME in the top quartile of US counties was approximately six times the amount prescribed in the lowest quartile.¹⁸ To our knowledge there are no published studies of variation in opioid prescribing among dental group practices. One might expect local leadership, shared norms, and instruction of junior dentists by experienced dentists to affect the decisions of individual group practices.

As with all retrospective cohort studies, this study has several limitations. First, we did not account for variation among individual prescribers, which was beyond the scope of this paper. Second, there were unmeasured patient characteristics, for example opioid use disorder. Third, a small number of dental procedures for this sample may have occurred outside of the Dental Program in emergency departments that were not studied here.

The combination of NSAID’s and acetaminophen is superior to opioids for controlling post-dental procedure pain in some studies.^31,32 Based on literature, the American Dental Association guidelines recommend screening patients for opioid ‘red flags’ or risk signs and also recommend that dentists consult a state’s prescription drug monitoring program (PDMP) before prescribing.³¹ Civilian dental practices have had low rates of requesting medical records and rare use of PDMP’s.³³ The military health system has advantages here because it integrates dental care into the medical record and a unified prescription record system provides military dental clinicians with readily available information and other flags for possible opioid overuse.

It is difficult to change the behavior of clinicians whose training has been to provide opioids to ameliorate significant pain. The MHS has undertaken many initiatives to address possible over-prescribing, for example the creation of an Opioid Registry in 2017 intended to help clinicians and clinic directors monitor care delivery. There are some signs that their efforts have been successful -- the military’s rate of deadly overdoses is a quarter of the national average.³⁴ However, this study suggests additional strategies may be warranted. Prior research has demonstrated that ‘academic detailing’, or direct outreach education to clinicians, is an effective communication strategy that actually leads to change in practice,³⁵ and it has been successfully tailored to encourage safer opioid prescribing.^36,37 Further strategies may be needed to address patient expectations for narcotic pain medication, which can translate into pressure on clinicians.

In conclusion, these findings imply that more attention should be paid to strategies utilized at dental clinics of military facilities achieving lower prescribing rates given nine of the thirty largest facilities demonstrated rates significantly lower than the average. We hope that our evidence that some dental groups can achieve lower levels of opioid prescribing may increase confidence in dental leadership discussions leading to lower rates of opioid prescribing.