Abstract
Purpose
Evaluate changes in rates of codeine and alternate opioid prescription filling after pediatric dental procedures, associated with the FDA codeine black box warning for tonsillectomy/adenoidectomy.
Methods
Patients ≤17 years who underwent any tooth extraction or restorative procedure in 8/2011–8/2016 were identified in Ohio Medicaid claims. Interrupted time series analyses were used to evaluate the impact of the FDA warning on the rates of codeine or alternative opioid prescription fills within 14 days post-procedure.
Results
In August 2011, the rates of codeine and alternative opioids prescription fills after a dental procedure were 2.3% and 3.3%, respectively. These rates decreased monthly by 0.02% and 0.01%, respectively, during the study period (p<0.0001). The FDA warning was associated with an immediate drop in codeine prescription filling of 0.4% (p<0.001), but there was no such association with alternative opioid prescription filling. The codeine decrease was only significant among extractions with/without concurrent restorative procedures, males, 12–17 year-olds, whites, and small metropolitan area residents.
Conclusions
Among children enrolled in Ohio Medicaid, opioid prescription fill rates after dental procedures decreased between 2011 and 2016. In addition, there was a significant immediate reduction in codeine prescription filling after the FDA codeine black box warning for tonsillectomy/adenoidectomy.
Keywords: codeine, prescriptions, drug, pediatric dentistry
Introduction
Children and adolescents continue to receive opioid prescriptions despite increased awareness of potential adverse effects and evidence to suggest that non-opioid strategies are effective for pain control.1 Opioid prescriptions for this population in the US increased slightly from 2.2 million in 1996 to 2.5 million in 2012 with 2.9% of individuals receiving an opioid prescription.2 Dental visits account for a significant proportion of opioid prescriptions in US children and adolescents (16%) and are second only to medical provider visits for traumatic injury.2 Given the large number of opioid prescriptions by dentists, there is concern that dentists inadvertently contribute to risk of opioid abuse in both children and adults.3
Historically, codeine has been prescribed in pediatric care because it was thought to be a safe alternative to morphine.4 However it is now known to have variability in therapeutic doses and the toxic dose is dependent on patient specific genetic polymorphisms.4 Specifically, patients with duplication of the CYP2D6 allele are more likely to have an ultra-rapid metabolizer phenotype,5 and as a consequence, may experience increased risk of death from codeine consumption due to toxic accumulation of morphine.6 Following the report of deaths associated with codeine among children who were shown to be ultra-rapid metabolizers, the US Food and Drug Administration (FDA) issued a safety announcement regarding the use of codeine after tonsil and adenoid surgery in August 2012.7, 8 In February 2013 the FDA added a “black box warning” to the drug label of codeine-containing preparations advising healthcare providers to use alternative analgesics following tonsillectomy and/or adenoidectomy and avoid codeine use in these patients.7, 9 Nationwide prescriptions of codeine after tonsillectomy and adenoidectomy decreased substantially, but not entirely, after the 2013 black box warning.20 In April 2017, the FDA issued a contraindication to the use of codeine to treat pain or cough in children younger than 12 years and a warning to recommend against its use in adolescents between 12 and 18 who are obese, have severe lung disease, or obstructive sleep apnea.10
Codeine prescribing by dentists for children and adolescents is an area of concern because dentists account for 14% of codeine prescriptions among US children.11 Although the initial FDA warnings regarding codeine primarily targeted tonsillectomy and adenoidectomy, it is unknown if they impacted the codeine prescribing patterns of dentists. The objective of this study was to evaluate the change in rates of codeine and alternate opioid prescription filling after restorative procedures and tooth extractions among children and adolescents in Ohio from before to after the 2013 FDA black box warning against codeine use for pediatric tonsillectomy and adenoidectomy, and prior to the 2017 contraindication warning.
METHODS
Data Source and Study Population
This study was approved by the Nationwide Children’s Hospital Institutional Review Board Institutional Review Board (IRB18–00409). We conducted a retrospective cohort study using Ohio Medicaid claims data from the period of August 2011 to August 2016. The database contained enrollment, health care, and pharmacy claims data on children and adolescents aged 0–19, from families with incomes up to 206% of the federal poverty level, who were enrolled in Ohio Medicaid/Children’s Health Insurance Program (CHIP) at any time during this period.12 We identified children and adolescents aged 0–17 years who underwent any tooth extraction or restorative procedure by searching for professional claims containing Code on Dental Procedures and Nomenclature (CDT) codes for restoration, pulp therapy, other restorative procedures (D2000-D3999), and extractions ( D7111, D7140, or D7210-D7251). In order to capture all opioid prescription fills within 14 days of each procedure, we excluded procedures from children and adolescents who were not enrolled in Ohio Medicaid during the month after their procedure and children and adolescents whose surgery occurred after August 17th, 2016. In addition, we excluded procedures by providers outside of Ohio in order to include a consistent set of providers and opioid prescribers.
Study Design
To examine overall trends in prescription fills for codeine and alternative opioids following dental procedures, we used a time series design. Observations were aggregated at the month level based on the date of procedure. Time trends in prescription fill practices could then be estimated from the change in monthly rates of prescription fills. To examine the effect of the FDA codeine warning on opioid prescription filling after dental procedures, we used an interrupted time series design. This type of quasi-experimental design is used to evaluate the impact of policy changes or system-level interventions on the rate of an outcome using observational data.13 After aggregating observations at the month level, months were classified into 3 time periods: pre-FDA safety announcement, August 2011-July 2012 (12 months); FDA review period, August 2012 to February 2013 (7 months); and post-FDA warning for tonsillectomy and adenoidectomy, March 2013-August 2016 (41 months). Data from the seven-month review period were excluded from the interrupted time series analysis.
Measures
The outcomes of interest were 1) the percentage of patients with at least 1 prescription fill for codeine within 0–14 days after their dental procedure and 2) the percentage of patients with at least 1 prescription fill for an alternative opioid within 0–14 days after their procedure. To identify opioid prescription fills, we searched the outpatient pharmacy claims for opioid national drug codes (NDCs) as specified by the Centers for Disease Control and Prevention.14 Using a substring search of the drug name, we categorized opioids as codeine products or alternative opioid medications (e.g. oxycodone, hydrocodone, etc.).
The following procedure and patient characteristics were examined in comparative analyses of codeine and alternative opioid prescription filling rates: procedure group, patient age, sex, race and ethnicity, and rural/urban location of residence. Procedures were initially classified into 3 groups as follows: (1) restorative only (restorations, pulp therapy, and other restorative procedures in the CDT range D2000-D3999; without concurrent extraction procedures for an extraction at the same encounter), (2) extractions only (surgical and non-surgical procedures in the CDT ranges D7111, D7140, or D7210-D7251; without concurrent restorative procedures at the same encounter), (3) restorative with concurrent extraction (presence of a CDT code from group 1 and presence of a CDT from group 2 at the same encounter). We subsequently combined all encounters with an extraction into one group because there were no significant differences in the rates of codeine prescription fills between groups 2 and 3. This resulted in two procedure groups: restorative only and extractions with or without concurrent restorative procedures. Age was classified into two groups as 0–11 and 12–17 years. For race, we used the information available in the Medicaid member eligibility database, which contains up to 13 racial/ethnic groups. The comparative analyses by race were restricted to white and black patients (of either Hispanic and non-Hispanic ethnicity) due to a small number of patients in any other racial/ethnic groups. Urban/rural location was defined using the National Center for Health Statistics 2013 county-level urban/rural classification scheme.15
Statistical Analysis
Frequencies were used to describe the study population characteristics, overall and by time period. Chi-squared tests were performed to examine differences in the proportion of patients with an opioid prescription fill by time period (pre-FDA safety announcement vs. post-FDA warning) and patient characteristics. All analyses were conducted separately for the outcomes of codeine and alternate opioid prescription fills. First, linear regression with autoregressive errors (SAS AUTOREG procedure) was used to model each outcome as a function of month of surgery, for the entire 61-month study period (August 2011-August 2016). The slope for month of surgery estimated the overall monthly trend in the outcome. Second, linear segmented regression with autoregressive errors was used to model each of the two outcomes of interest as a function of (1) month of surgery, (2) an indicator variable for pre- safety announcement versus post-FDA black box warning period, and (3) the number of months after the warning. This model allowed us to estimate pre- and post-FDA black box warning trends as well as immediate decreases in the outcomes from the month before the FDA review began to the month after the FDA black box warning was released. To adjust for potential confounding by patient characteristics, we initially included in the segmented regression model covariates for the average patient age at procedure, and monthly proportion of female patients, white patients, and residents of large central metropolitan area. Because none of these covariates were significant at P<0.05, we removed them from all subsequent models.
Subgroup differences in the effects of the FDA warning
We ran additional segmented regression models to examine differences in the effects of the FDA warning by procedure group and patient sociodemographic characteristics (sex, age, race, and rural/urban location of residence). To examine differences by procedure group we modeled each outcome as a function of procedure group (restorative only vs. any extraction), month of surgery, an interaction term between procedure group and month of surgery, an indicator variable for pre- review versus post-FDA warning, an interaction term between procedure group and the post-warning indicator, the number of months after the warning, and an interaction term between procedure group and the number of months after the warning. These models enabled the estimation of differences the monthly trends in outcomes before and after the FDA warning as well as the immediate changes in outcomes associated with the FDA warning separately among patients with restorative procedures only versus any extractions with or without concurrent restorative procedures. Analyses evaluating differences by sex, age, race, and rural/urban location of residence were analogous to those describe above. All of the linear and segmented linear regression analyses were conducted using the SAS AUTOREG procedure to correct for autocorrelation in the outcomes. Statistical significance was set at a two-sided p<0.05 for all tests. SAS Enterprise Guide 7.15 (Cary, NC, USA) was used for all statistical analyses.
RESULTS
A total of 1,020,878 encounters among 512,278 patients were included in the study. Table 1 describes the patient and encounter-level characteristics of the study population. When considering each patient’s first dental procedure encounter during the study period, 28% of the study population underwent tooth extraction. Most patients were younger than 12 years at first procedure (65.2%). Female and white patients comprised 50.5% and 66.9% of the population, respectively (Table 1). The monthly average number of total encounters during the study period was 18,920 (4,688 for extractions).
Table 1.
Patient and encounter-level characteristics among Ohio Medicaid children who underwent a dental procedure in August 2011-August 2016.
| Patients, N (%)a | Encounters, n (%) | ||
|---|---|---|---|
| Overall | 512,278 (100.0) | 1,020,878 (100.0) | |
| Procedure group | Fillings/other restorative procedures | 369,777 (72.2) | 768,023 (75.2) |
| Tooth extraction | 90,206 (17.6) | 167,170 (16.4) | |
| Extraction and concurrent restorative procedure | 52,295 (10.2) | 85,685 (8.4) | |
| Female (%) | 258,495 (50.5) | 520,041 (50.9) | |
| Age in years (%) | 1 | 1,645 (0.3) | 1,748 (0.2) |
| 2 | 8,587 (1.7) | 9,648 (1.0) | |
| 3 | 23,804 (4.7) | 30,457 (3.0) | |
| 4 | 37,960 (7.4) | 57,078 (5.6) | |
| 5 | 41,566 (8.1) | 72,840 (7.1) | |
| 6 | 43,076 (8.4) | 84,641 (8.3) | |
| 7 | 42,626 (8.3) | 90,587 (8.9) | |
| 8 | 39,640 (7.7) | 87,392 (8.6) | |
| 9 | 35,055 (6.8) | 75,700 (7.4) | |
| 10 | 30,768 (6.0) | 64,662 (6.3) | |
| 11 | 29,142 (5.7) | 58,949 (5.8) | |
| 12 | 29,141 (5.7) | 58,731 (5.8) | |
| 13 | 29,658 (5.8) | 61,029 (6.0) | |
| 14 | 29,812 (5.8) | 63,135 (6.2) | |
| 15 | 30,039 (5.9) | 66,799 (6.5) | |
| 16 | 29,985 (5.9) | 67,985 (6.7) | |
| 17 | 29,774 (5.8) | 69,497 (6.8) | |
| Race (%) (N=486706) | White | 325,790 (66.9) | 669,843 (68.9) |
| Black | 141,037 (29.0) | 262,716 (27.0) | |
| Asian or Pacific Islander | 6,296 (1.3) | 13,035 (1.3) | |
| Other/More than one | 13,583 (2.8) | 26,101 (2.7) | |
| Rural/urban location of residence (%) (N=511567) | Large central metro | 164,930 (32.2) | 311,796 (30.6) |
| Large fringe metro | 81,549 (15.9) | 162,937 (16.0) | |
| Medium metro | 131,163 (25.6) | 266,474 (26.1) | |
| Small metro | 26,325 (5.2) | 55,389 (5.4) | |
| Micropolitan or Noncore | 107,600 (21.0) | 223,117 (21.9) | |
| Codeine prescription fill within 14 days | 9,619 (1.9) | 16,154 (1.6) | |
| Alternate opioid prescription fill within 14 days | 15,421 (3.0) | 29,581 (2.9) | |
Patient-level data at the first encounter during the study period
Codeine prescription fills
During the first month of the study period (August 2011), and prior to the FDA review, the codeine prescription fill rate was 2.3%. Prescription fill rates by procedure group were 1.2% among encounters with restorative procedures only, and 6.1% among encounters with tooth extractions (with or without a concurrent restorative procedure) (p<0.0001 for the difference). Overall, the rate of codeine prescription fills after any procedure decreased each month throughout the entire 61-month study period by an average of 0.02% (p<0.0001). Among extractions encounters, codeine prescription fills decreased monthly by 0.1% throughout the study period (p<0.0001). In segmented regression, the FDA codeine warning was associated with an overall immediate reduction in the rate of codeine prescription fills after any procedure of 0.4% (95% CI: −0.6%, −0.2%; p<0.001). In the last month of the study period (August 2016), the overall rate of codeine prescription fills after any procedure was 1.0% (0.4% among encounters with restorative procedures only, 2.7% among encounters with tooth extractions; p<0.0001). The immediate decrease in codeine prescription fill rate associated with the FDA warning was only significant among encounters with an extraction procedure (−0.7%, p<0.001) (Figure 1a; Table 2). Additional subgroup comparisons by patient sociodemographic characteristics were conducted among the extraction encounters. At the beginning of the study period, there were significant differences in the percentage of extraction encounters with a codeine prescription fill by patient sex, age, and location of residence. The immediate decrease in codeine prescription fill rate associated with the FDA warning was significant only among male patients, 12–17 year olds, whites, and residents of small metropolitan areas (Table 2). In the last month of the study period, there were significant differences in codeine prescription fill rates by sex (2.6% in males vs. 3.0% in females; p=0.02), age (1.4% in 0–11 year olds vs. 6.7% in 12–17 year olds; p<0.0001), race (3.6% in blacks vs. 2.7% in whites; p<0.0001), and location of residence (2.3% in large central metropolitan vs. 4% in medium metropolitan, and 3.7% in small metropolitan areas; p<0.0001 for both).
Figure 1a.
Percentage of children filling a codeine prescription within 14 days after dental procedures in August 2011 to August 2016. The lines show the predicted probabilities from the segmented regression model. The period between the FDA review and warning (August 2012-February 2013), shown between the two vertical lines, was excluded from the analysis. Extraction group includes extractions with or without concurrent restorative procedures.
Table 2.
Effects of the 2013 FDA codeine warning on the percentage of dental procedures with a codeine prescription fill.
| Intercept: % Rx fill in Aug 2011 | Immediate change in % Rx fill between Jul 2012 and Mar 2013 | Slope of monthly change Aug 2011-Jul 2012 | Slope of monthly change Mar 2013-Aug 2016 | ||
|---|---|---|---|---|---|
| Procedure group | Restorative only | 1.25 | −0.30 | −0.02 | 0.01 |
| Extraction† | 6.13 | −0.71 *** | −0.04 | −0.01 | |
| Difference | 4.88 *** | −0.41 | −0.02 | −0.02 | |
| Extractions† by sex | Male | 5.40 | −1.01 ** | 0.03 | −0.08 |
| Female | 6.69 | −0.43 | −0.10 | 0.05 | |
| Difference | 1.29 * | 0.58 | −0.13 | 0.13 | |
| Extractions† by age in years | 0–11 | 4.53 | −0.55 | −0.02 | −0.03 |
| 12–17 | 9.71 | −1.06 * | −0.09 | 0.07 | |
| Difference | 5.17 *** | −0.51 | −0.07 | 0.11 | |
| Extractions† by race | Black | 6.00 | 0.07 | −0.07 | 0.03 |
| White | 6.06 | −0.92 * | −0.02 | −0.03 | |
| Difference | 0.06 | −1.00 | 0.05 | −0.06 | |
| Extractions† by rural/urban location of residence | Large central metro | 5.32 | −0.17 | −0.12 | 0.08 |
| Large fringe metro | 4.87 | −0.70 | −0.08 | 0.05 | |
| Difference | −0.45 | −0.53 | 0.04 | −0.03 | |
| Medium metro | 8.02 | −0.50 | −0.03 | −0.05 | |
| Difference | 2.70 ** | −0.33 | 0.09 | −0.13 | |
| Small metro | 5.69 | −2.13 *** | 0.09 | −0.11 | |
| Difference | 0.37 | −1.95 * | 0.21 | −0.19 | |
| Micropolitan or Noncore | 6.00 | −1.08 | 0.01 | −0.07 | |
| Difference | 0.68 | −0.91 | 0.13 | −0.15 | |
=p<.05
=p<.01
=p<.001
with or without concurrent restorations.
All estimates were derived from segmented linear regression with autoregressive errors. The period between the FDA review and warning (August 2012-February 2013) was excluded from the analysis. A separate model was run for each of the study factors shown (procedure group, sex, age, race, location of residence).
Alternate prescription fills
During the first month of the study period, and prior to the FDA review, the alternate opioid prescription fill rate was 3.3%. Prescription fill rates by procedure group were 0.4% among encounters with restorative procedures only, and 11.3% among encounters with tooth extractions (p<0.0001). Overall, the rate of alternate opioid prescription fills decreased each month throughout the entire study period by an average of 0.01% (p<0.0001). Among extraction encounters, alternate opioid prescription fills decreased monthly by 0.04% throughout the study period (p<0.0001). In segmented regression, the FDA codeine warning was not associated with significant changes in the rate of alternate opioid prescription fills following any of the dental procedure types, either overall or among encounters with an extraction procedure (Figure 1b; Table 3). Although during the first month of the study period there were significant differences in the percentage of alternate opioid prescription fills after extraction encounters by patient sex, age, race and location of residence, the FDA codeine warning was not associated with significant changes in alternate opioid prescription fills among any of the patient sociodemographic subgroups.
Figure 1b.
Percentage of children filling an alternate opioid prescription within 14 days after dental procedures in August 2011 to August 2016. The lines show the predicted probabilities from the segmented regression model. The period of the FDA review and warning (August 2012-February 2013), shown between the two vertical lines, was excluded from the analysis. Extraction group includes extractions with or without concurrent restorative procedures.
Table 3.
Effects of the 2013 FDA codeine warning on the percentage of dental procedures with an alternate opioid prescription fill.
| Intercept: % Rx fill in Aug 2011 | Immediate change in % Rx fill between Jul 2012 and Mar 2013 | Slope of monthly change Aug 2011-Jul 2012 | Slope of monthly change Mar 2013-Aug 2016 | ||
|---|---|---|---|---|---|
| Procedure group | Restorative only | 0.4 | −0.1 | 0.0 | 0.0 |
| Extraction† | 11.3 | −0.1 | −0.1 | 0.1 | |
| Difference | 10.3 *** | 0.0 | −0.1 | 0.1 | |
| Extractions† by sex | Male | 8.8 | −0.4 | 0.0 | 0.0 |
| Female | 13.9 | 0.3 | −0.2 * | 0.2 * | |
| Difference | 5.1 *** | 0.7 | −0.2 | 0.2 | |
| Extractions† by age in years | 0–11 | 0.2 | 0.0 | 0.0 | 0.0 |
| 12–17 | 36.2 | −0.8 | −0.3 | 0.2 | |
| Difference | 36.0 *** | −0.9 | −0.3 | 0.2 | |
| Extractions† by race | Black | 12.5 | 0.3 | −0.3 * | 0.2 * |
| White | 10.0 | −0.2 | 0.0 | 0.0 | |
| Difference | −2.6 * | −0.5 | 0.3 | −0.3 | |
| Extractions† by rural/urban location of residence | Large central metro | 11.4 | 0.4 | −0.3 * | 0.2 |
| Large fringe metro | 11.4 | −0.2 | −0.1 | 0.0 | |
| Difference | 0.0 | −0.6 | 0.2 | −0.2 | |
| Medium metro | 12.1 | −0.9 | −0.1 | 0.1 | |
| Difference | 0.7 | −1.3 | 0.2 | −0.2 | |
| Small metro | 8.8 | −0.6 | 0.0 | 0.0 | |
| Difference | −2.6 * | −1.0 | 0.3 | −0.2 | |
| Micropolitan or Noncore | 11.2 | 0.8 | −0.1 | 0.0 | |
| Difference | −0.2 | 0.4 | 0.2 | −0.2 | |
=p<.05
=p<.01
=p<.001
with or without concurrent restorations.
All estimates were derived from segmented linear regression with autoregressive errors. The period between the FDA review and warning (August 2012-Februarv 2013) was excluded from the analysis. A separate model was run for each of the study factors shown (procedure group, sex, age, race, location of residence).
DISCUSSION
This study found that the 2013 FDA black box warning against codeine prescription after tonsillectomy and adenoidectomy was associated with an immediate 0.4% reduction in codeine prescription fill rates after pediatric dental procedures in Ohio. The FDA codeine warning was not associated with changes in alternate opioid prescription fill rates in the same population. In addition, both the rates of codeine and alternate opioid prescription fills decreased in this population throughout the period of August 2011 through August 2016 at monthly rates of 0.02% and 0.01%, respectively.
The data suggest that rates of codeine prescriptions by dentists in Ohio prior to the FDA codeine black box warning was higher than the overall national average. Livingston11 found overall codeine prescriptions for children and adolescents were 1.6% in 1996 compared to our study findings of 2.3% codeine prescription fills in the August 2011. However, the decreasing trend in the rate of codeine prescription filling that was observed throughout the study period resulted in an overall codeine prescription fill rate of 1% during the last month of the study period (August 2016), which in turn is lower than the reported national rate of 1.5% in 2013.11 Likewise, the initial rate of filling a prescription for non-codeine narcotics (3.3% in August 2011) was higher than national overall values of 2.9% in 2012, but decreased throughout the study period, although at a slower rate than codeine.2 Caution must be used in making these comparisons because data is from the nationally representative Medical Expenditure Panel Survey and not Medicaid enrollees only.2, 11 In contrast to other findings, non-codeine narcotics were prescribed more commonly to Ohio children and adolescents than codeine throughout the period.16, 17
The overall trend in the reduction of codeine and alternative narcotic prescriptions in Ohio may be related to other interventions in the state. From 2010 to 2014, there was a 39.7% decrease in opioid prescriptions in Ohio emergency departments. This reduction was in part a result of the April 2012 introduction of guidelines aimed at reducing inappropriate prescribing of opioids by emergency physicians in Ohio.18 In this study, we found that the FDA black box warning contributed to the reduction in codeine prescription fills even after adjusting for the overall decreasing trend that was observed throughout 2011– 2016. Despite the reduction in codeine prescriptions, there was not a reciprocal increase in non-codeine narcotic prescriptions. This may indicate dentists increased the use of nonnarcotic pain medications such as acetaminophen and nonsteroidal anti-inflammatories (NSAIDS) for pain control. Evidence suggests that NSAIDS may be a sufficient analgesic to treat most postoperative dental pain.1 There was no significant decrease in codeine prescription filling rates for encounters that involved restoration procedures only, but this is not surprising due to initial low rates that were unlikely to significantly decrease.
Changes in guidelines do not always have an impact, even in the groups they specifically target. In 1997 the American Academy of Pediatrics issued and reaffirmed in 2006 guidelines to warn about codeine’s potential dangers and lack of documented efficacy in children and adolescents who have cough or upper respiratory infection (URI).19 In 2006 the American College of Chest physicians’ guideline on treatment of pediatric cough recommended against codeine.19 Yet these guidelines were not associated with a decline in codeine prescriptions for cough or URI visits in the Emergency Department.19 Following the FDA warning of codeine use following tonsillectomy/adenoidectomy, there was a significant 13.3% level decrease in the proportion of children with at least one prescription for codeine.20 Despite the FDA warning specifically centering around tonsillectomy and/or adenoidectomy procedures, 5.1% of children had one or more prescription fills for codeine 2 years after the warning.20 Lack of awareness is one of the most common barriers to adherence to clinical guidelines followed by familiarity (experience with alternative options), agreement, self-efficacy, outcome expectancy, and inertia.21
The number of codeine prescriptions may have been impacted by the reclassification in hydrocodone drug schedule from III to II in October 2014 by the US Drug Enforcement Administration.22 Codeine remained a Schedule III drug. Schedule II drugs require a written prescription, whereas schedule III can be telephoned in to pharmacies.16, 22
Dentists have been identified as a significant source of opioid prescriptions.3, 17, 22 However, data from this study suggest that dentists prescribing patterns in Ohio have changed as evidenced by the reduction in codeine prescription fill rates. This reduction appears to be in part an effect of the FDA codeine black box warning even though the warning was primarily intended for surgeons following tonsillectomy and/or adenoidectomy. Therefore, this study is unique because it shows a significant change in a group of providers that were not targeted in the guideline change. Although the 0.4% reduction in codeine prescription fills associated with the FDA warning appears to be small, it impacts a meaningful number of patients. Based on an average total of 18,920 monthly encounters during the study period and a codeine prescription fill rate of 2.3% pre- review, this reduction translates into 76 fewer monthly prescriptions following dental procedures for children and adolescents in the state of Ohio alone. In children and adolescents, prescriptions of opioids have been associated with an adverse event in one of every 2611 prescriptions even when the majority (71.2%) did not deviate from the prescription regimen.17 This decrease in codeine prescription fills after dental procedures reduces the risk of adverse events for children and adolescents. Also, the overall downward trend in both codeine and alternate opioid prescription fills in Ohio occurred before the publication of articles implicating dentistry in the opioid epidemic and prior to mandated opioid training.
A strength of this study is that it examines the impact that a major nationwide review and warning on codeine safety had on prescription fill rates after pediatric dental procedures. Both the extent of opioid prescribing in the pediatric dental population and the potential effects of interventions on opioid prescribing rates to pediatric dental patients are underexplored.23 Another strength of this study is that the data comes from prescription fill data and does not rely on provider or patient family recall. One disadvantage of this study that is that it is state specific and therefore may have limited generalizability which is a gap in the research literature on opioid prescribing practices.23 Ohio is among the states that have been most affected by the opioid crisis and this may have driven provider awareness. This study only examined Medicaid claims and cannot be generalized to the private insurance or uninsured populations. Also, this study only examined prescriptions filled, not all prescriptions that were written, so the decrease detected could have reflected parental-level rather than provider-level changes. However, past studies in dentistry and medicine that examined trends in drug prescribing have tracked dispensed prescriptions.16–18, 20, 22, 24–26 A final limitation is that changes in the use of non-opioid analgesics such as acetaminophen or NSAIDS could not be examined because these are primarily purchased over the counter.
CONCLUSIONS
An analysis of Ohio Medicaid claims from 8/2011 to 8/2016 found:
The FDA warning regarding codeine use after tonsillectomy and/or adenoidectomy was associated with an immediate drop in codeine prescription filling of 0.4% (p<0.001) with no associated change in alternative opioid prescription fills.
The decrease in codeine prescription fills was significant among extractions with or without restorative concurrent procedures, males, 12–17-year olds, whites, and residents of small metropolitan areas.
The rates of codeine and alternative opioid prescriptions fills after a dental procedure decreased each month by 0.02% and 0.01% respectively during the study period.
Acknowledgements
This study was funded in part by Grant Number R21 MD011767 from the National Institute of Minority Health and Health Disparities of the National Institutes of Health.
REFERENCES
- 1.Wong YJ, Keenan J, Hudson K, et al. Opioid, NSAID, and OTC Analgesic Medications for Dental Procedures: PEARL Network Findings. Compend Contin Educ Dent 2016;37(10):710–18. [PubMed] [Google Scholar]
- 2.Groenewald CB, Rabbitts JA, Gebert JT, Palermo TM. Trends in opioid prescriptions among children and adolescents in the United States: a nationally representative study from 1996 to 2012. Pain 2016;157(5):1021–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Schroeder AR, Dehghan M, Newman TB, Bentley JP, Park KT. Association of Opioid Prescriptions From Dental Clinicians for US Adolescents and Young Adults With Subsequent Opioid Use and Abuse. JAMA Intern Med 2019;179(2):145–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kelly L, Madadi P. Is There a Role for Therapeutic Drug Monitoring With Codeine? Ther Durg Monit 2012;34(3):249–56. [DOI] [PubMed] [Google Scholar]
- 5.Johansson I LE, Bertilsson L, Dahl ML, Sjoqvist F, Ingelman-Sundberg M. Inherited amplification of an active gene in the cytochrome P450 CYP2D locus as a cause of ultrarapid metabolism of debrisoquine. Proc Natl Acad Sci U S A 1993;90(24):11925–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ciszkowski C, Madadi P, Phillips MS, Lauwers AE, Koren G. Codeine, ultrarapid-metabolism genotype, and postoperative death. N Engl J Med 2009;361(8):827–8. [DOI] [PubMed] [Google Scholar]
- 7.Tobias JD, Green TP, Cote CJ, et al. Codeine: Time to Say “No”. Pediatrics 2016;138(4). [DOI] [PubMed] [Google Scholar]
- 8.U.S. Food and Drug Administration. Drug safety communication. Codeine use in certain children after tonsillectomy and or adenoidectomy may lead to rare but life threatening adverse events or death. Rockville, MD: US. Food and Drug Administration; 2012. [Google Scholar]
- 9.US Food and Drug Administration. Drug safety communication. Safety review update of codeine use in children: a new boxed warning and contraindication on use after tonsillectomy and or adenoidectomy. Rockville, MD: US Food and Drug Administration; 2013. [Google Scholar]
- 10.US Food and Drug Administration FDA Drug Safety Communication: FDA restricts use of prescription codeine pain and cough medicines and tramadol pain medicines in children; recommends against use in breastfeeding women. Rockville, MD: US Food and Drug Administration; 2017. [Google Scholar]
- 11.Livingstone MJ, Groenewald CB, Rabbitts JA, Palermo TM. Codeine use among children in the United States: a nationally representative study from 1996 to 2013. Paediatr Anaesth 2017;27(1):19–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ohio Department of Medicaid https://www.medicaid.ohio.gov/FOR-OHIOANS/Programs/Children-Families-and-Women. Accessed Accessed 5/18/2019.
- 13.Penfold RB, Zhang F. Use of interrupted time series analysis in evaluating health care quality improvements. Academic pediatrics 2013;13(6):S38–S44. [DOI] [PubMed] [Google Scholar]
- 14.Centers for Disease Control and Prevention Analyzing prescription data and morphine milligram equivalents (MME). https://www.cdc.gov/drugoverdose/resources/data.html. Accessed Accessed 05/01/2018.
- 15.Centers for Disease Control and Prevention NCHS Urban-Rural Classification Scheme for Counties. https://www.cdc.gov/nchs/data_access/urban_rural.htm: 2017. Accessed 4-1-2019 2019.
- 16.Etminan M, Nouri MR, Sodhi M, Carleton BC. Dentists’ Prescribing of Analgesics for Children in British Columbia, Canada. J Can Dent Assoc 2017;83:h5. [PubMed] [Google Scholar]
- 17.Chung CP, Callahan ST, Cooper WO, et al. Outpatient Opioid Prescriptions for Children and Opioid-Related Adverse Events. Pediatrics 2018;142(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Weiner SG, Baker O, Poon SJ, et al. The Effect of Opioid Prescribing Guidelines on Prescriptions by Emergency Physicians in Ohio. Ann Emerg Med 2017;70(6):799–808 e1. [DOI] [PubMed] [Google Scholar]
- 19.Kaiser SV, Asteria-Penaloza R, Vittinghoff E, et al. National patterns of codeine prescriptions for children in the emergency department. Pediatrics 2014;133(5):e1139–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chua KP, Shrime MG, Conti RM. Effect of FDA investigation on Opioid Prescribing to Children After Tonsillectomy/Adenoidectomy. Pediatrics 2017;140(6). [DOI] [PubMed] [Google Scholar]
- 21.Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA 1999;282(15):1458–65. [DOI] [PubMed] [Google Scholar]
- 22.Gupta N, Vujicic M, Blatz A. Opioid prescribing practices from 2010 through 2015 among dentists in the United States: What do claims data tell us? J Am Dent Assoc 2018;149(4):237–45 e6. [DOI] [PubMed] [Google Scholar]
- 23.Lutfiyya MN, Gross AJ, Schvaneveldt N, Woo A, Lipsky MS. A scoping review exploring the opioid prescribing practices of US dental professionals. J Am Dent Assoc 2018;149(12):1011–23. [DOI] [PubMed] [Google Scholar]
- 24.McCauley JL, Hyer JM, Ramakrishnan VR, et al. Dental opioid prescribing and multiple opioid prescriptions among dental patients: Administrative data from the South Carolina prescription drug monitoring program. J Am Dent Assoc 2016;147(7):537–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Bushnell GA, Sturmer T, Swanson SA, et al. Dosing of Selective Serotonin Reuptake Inhibitors Among Children and Adults Before and After the FDA Black-Box Warning. Psychiatr Serv 2016;67(3):302–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Busch SH FR, Leslie DL, Martin A, Rosenheck RA, Martin EG, Barry CL. Antidepressants and suicide risk: how did specific information in FDA safety warnings affect treatment patterns? Psychiatr Serv 2010;61(1):11–16. [DOI] [PMC free article] [PubMed] [Google Scholar]