Opioid prescribing by oral and maxillofacial surgeons in the United States, 2016–2019

Abstract

Objective:

To describe opioid prescribing trends among oral and maxillofacial surgeons (OMFS).

Methods:

Prescriptions by OMFS were identified from IQVIA Longitudinal Prescription Dataset, 2016–2019. OMFS-based, patient-based and population-based prescribing rates and changes in high-risk opioid prescribing were calculated annually. We used linear regression to describe trends.

Results:

There were 13.9 million opioid prescriptions among 12.5 million patients (627 prescriptions/OMFS/year). Hydrocodone and oxycodone decreased by 20.9% and 39.2% (p < 0.05), while tramadol and codeine increased by 24.3% and 6.1% (p < 0.05), respectively. Opioid prescribing rates significantly decreased by 27 prescriptions/OMFS/year, 18.6 patients/OMFS/year and by 0.9 prescriptions/100,000 population/year (p < 0.05 for all). From 2016 to 2019, the proportion of opioids >3 days decreased by 54.2% (p < 0.05) and prescriptions ≥50 MME/day decreased by 66.3% (p < 0.05). Although the number of opioid prescriptions by OMFS decreased in most states, 12% of states experienced increases.

Conclusion:

Opioid prescribing, especially high-risk prescribing, by OMFS has decreased. However, targeted interventions are warranted in some areas.

INTRODUCTION

Dental surgical procedures are a common indication for opioids. However, strong evidence based on third-molar impaction models support the efficacy of NSAIDs for dental pain [1]. Data suggest that dental opioids exceed CDC recommendations for acute pain [2]. This evidence coupled with the risks associated with opioids argues against the routine use of opioids even for procedures associated with severe pain [3]. However, oral and maxillofacial surgeons (OMFS) in the United States commonly prescribe opioids. Survey data from 2013 revealed 99% of OMFS prescribed an opioid after third-molar extraction [4]. Almost 30% of adolescents receive their first opioid after an extraction, and 54% of opioids prescribed after third-molar extraction are not used [5, 6]. Although opioids are common after surgical dental procedures, there is limited data on OMFS prescribing trends nationally. A recent national study among dentists and dental specialists found a 55% decrease in opioid prescribing in 2019 compared to 2011 [7]. However, this analysis did not include OMFS. Understanding opioid prescribing among OMFS is important to evaluate their response to the opioid epidemic and in creating targeted interventions to improve prescribing for dental pain.

Therefore, the objective of this study was to describe changes in opioid prescribing by OMFS from 2016 through 2019.

METHODS

We conducted a retrospective analysis of IQVIA Longitudinal Prescription Dataset (LRx) from January 1, 2016 to December 31, 2019. This dataset includes 92% of all dispensed outpatient prescriptions in the United States reimbursed by private and public payers and those paid for with cash. The analytic dataset was restricted to actively prescribing OMFS, defined as prescribing ≥20 of any medication annually. Variables included prescription data (i.e., generic name, National Drug Code [NDC], strength, dosage form, quantity dispensed, days’ supply, and date dispensed), and prescriber data (i.e., identifier, state) This study was reviewed and determined to be exempt by the University of Pittsburgh IRB.

Prescriptions for an oral solid dosage form (e.g., capsules/tablets) of an opioid analgesic (identified using NDCs) and patients between 1 and 99 years on date of dispensing were included. Opioids not primarily indicated for pain such as cough suppressant-containing opioids (e.g., promethazine/codeine) or buprenorphine-containing products (e.g., buprenorphine/naloxone) were excluded. Liquid and topical opioid formulations, if any patient, provider or prescription variables were missing, and if the days’ supply and/or quantity dispensed were ≤0 or >99th percentile were excluded.

Outcome measures included annual population-based prescribing rates (Rx/100,000 individuals), OMFS-based prescribing rates (Rx/OMFS) and patient-based prescribing rates (patients with an opioid Rx/OMFS), and high risk opioid prescribing. Population-based prescribing rates were calculated by dividing the annual number of opioids by the annual population size, as reported by the Census Bureau. OMFS-based prescribing rates were calculated by dividing the annual number of opioids by the annual number of actively prescribing OMFS. Patient-based prescribing rates were calculated by dividing the number of unique patients prescribed an opioid by the annual number of actively prescribing OMFS. Opioid dosages were calculated by multiplying the total daily doses by the strength and MME conversion factor [2]. Changes in high-risk opioid prescribing (>3 days’ supply; >7 days’ supply; ≥50 MME/day; ≥90 MME/day per CDC guidelines) [8] were calculated annually overall, by agent, and by state. We used descriptive statistics and linear regression analysis to describe prescribing trends. Statistical significance was determined as p ≤ 0.05. Analyses were conducted using SAS, version 9.4 and Microsoft Excel—2016.

RESULTS

From 2016 to 2019, there were 13.9 million opioids prescribed by OMFS to 12.5 million unique patients. Annually, a median of 3.5 million opioids were prescribed by a median of 5562 OMFS for a rate of 627.2 opioids/OMFS/year. From 2016 to 2019, the total number of opioids decreased significantly by 20.4% from 3.9 to 3.1 million opioid prescriptions (Figure 1a). However, differences were noted based on agent. Although the total number of hydrocodone prescriptions decreased by 20.9%, it was the most common agent prescribed annually. Oxycodone decreased by 39.2%, while codeine and tramadol increased by 6.1% and 24.3%, respectively.

FIGURE 1.

Opioid prescriptions by oral and maxillofacial surgeons (a) by agent and (b) proportion of opioids that are high risk in the United States, 2016–2019 [Color figure can be viewed at wileyonlinelibrary.com]

OMFS-based rates decreased from 663.9 opioids/OMFS in 2016 to 584.9 in 2019, a decrease of 27 prescriptions/OMFS/year (p < 0.05). Similarly, the number of patients receiving an opioid from an OMFS decreased from 585.2 patients/OMFS in 2016, to 530.9 in 2019, a decrease of 18.6 patients/OMFS/year (p < 0.05). The population-based rate also decreased from 12 prescriptions/100,000 population in 2016 to 9 prescriptions/100,000 population in 2019 at a rate of 0.9 prescription/100,000 population/year (p < 0.05).

High risk prescribing also decreased over the study period (Figure 1b). The proportion of OMFS opioids that exceeded 3 days’ supply decreased from 44.3% in 2016 to 29.3% in 2019 (p < 0.05). Following a similar trend, the proportion of prescriptions ≥50 MME/day decreased by almost two-thirds from 26.1% to 8.8% (p < 0.05). A small proportion of opioids prescribed by OMFS in 2016 exceeded 7 days’ supply or ≥90 MME/day (2.2% and 2.4% of opioids, respectively) and decreased in 2019 (0.6% and 0.5%; p < 0.05).

At the state-level, there was wide variation (Figure 2). Comparing 2019 with 2016, most states experienced a decrease in OMFS opioids. However, six states experienced increases ranging from 0.4% to 10.5% (Arizona, Arkansas, Kansas, Missouri, Idaho, and Montana). Similar state-level variation was observed for high risk opioid prescribing (data not shown).

FIGURE 2.

Changes in the number of opioids prescribed by oral and maxillofacial surgeons by state, 2019 compared with 2016 [Color figure can be viewed at wileyonlinelibrary.com]

DISCUSSION

Over time, opioid rates and high risk prescribing by OMFS decreased. However, we observed variation between states.

Following national trends, OMFS opioids have decreased. Policies and regulations such as prescription drug monitoring programs, CDC guidelines and statements by the American Dental Association (ADA) and the American Association of Oral and Maxillofacial Surgeons on opioid prescribing may have influenced these trends [9]. In our analysis, we identified a 20% decrease in opioids prescribed over 4 years. Our findings are in contrast to a national study of OMFS prescribing to a Medicare population, where an increase was observed [10]. The differences can be attributed to the population included in our analysis (all age groups) and using a dataset with all payers (vs. restricting to Medicare). In our analysis, the amount of decrease was smaller than that observed among dentists and specialists, who decreased their opioid prescribing by 41% during the same period [7]. This could be due to the nature of procedures performed by OMFS, which may be associated with higher pain levels compared to those by general dentists. Despite the decrease in the total number, hydrocodone still composed almost 65% of OMFS opioids and oxycodone composed 15%. In comparison, Yan and colleagues reported dentist prescribing of hydrocodone decreased by almost two-thirds over the same time period (vs. 21% herein) and by 46.1% for oxycodone (vs. 39%) [7]. Consistent with the Yan findings, OMFS increased prescribing of codeine and tramadol. The increase in codeine and tramadol could be due to their lower potency and fewer restrictions. However, these trends should be monitored as codeine is not indicated for children and tramadol is not effective for acute dental pain. In the absence of contraindications, NSAIDs remain the most effective and safest evidence-based option for managing acute dental pain [1].

High risk prescribing by OMFS has decreased. Although the proportion of prescriptions >3 days’ supply has decreased significantly by almost a third during our study, 29% of prescriptions in 2019 exceeded 3 days’. We observed an encouraging trend where the proportion of opioids ≥50 MME/day significantly decreased by two-thirds. Importantly, we found a low percentage of very high risk prescribing (>7 days’ supply, ≥90 MME/day) among OMFS in 2016 (2.2% and 2.4%, respectively) which decreased by >80% in 2019 (0.6% and 0.5%, respectively). Similar trends were observed among dentists and among OMFS prescriptions to Medicare beneficiaries [7, 10]. Currently, the ADA policy recommends no more than 7 days’ of opioids when indicated. However, our findings suggest that the maximum limit of opioid days’ supply recommended by ADA maybe worth revisiting to be concordant with CDC recommendations of no more than 3 days’ supply for acute pain.

States varied in prescribing rates and in high risk prescribing. The majority of states experienced decreases in opioid prescribing, with the greatest decreases in Alaska, Rhode Island, and Wyoming (−30%, −31%, and −34%, respectively). However, six states experienced increases, with the highest observed in Idaho and Montana (increase of 8% and 11%, respectively). State variation was also observed in high risk prescribing, and there were discrepancies within state depending on the definition used (>3 days’ supply vs. ≥50 MME). The variation between states is also seen among dentists and non-dentists and may be a reflection of different policies, state-specific guidelines, and prescribing cultures [11]. The observed variation could also be due to lack of national guidelines for acute dental pain. Upcoming ADA guidelines for acute pain management may improve prescribing for dental pain among OMFS and other providers.

Our study has limitations. As the dataset does not contain dental visit information (such as procedures), we are not able to adjust for the procedure complexity. Because our objective was to describe OMFS prescription trends, we did not adjust for patient-level characteristics (age and sex). However, our analysis included national data inclusive of all payers and all age groups and provides timely information on OMFS opioids at national- and state-levels.

CONCLUSION

OMFS opioid prescribing, especially high risk, has decreased significantly. However, opportunities remain to improve high risk and high rates in some states. Therefore, interventions tailored to OMFS are warranted.