Abstract
Purpose:
To determine the rate and risk factors for new persistent opioid use after ophthalmic surgery in the United States (US).
Design:
Retrospective claims-based cohort analysis.
Participants:
Opioid-naive patients aged 13 years and older who underwent incisional ophthalmic surgery between January 1, 2012 to June 30, 2017 in Optum’s de-identified Clinformatics® DataMart Database.
Methods:
New persistent opioid use was defined as fulfillment of an opioid prescription both in the 90 day and the 91–180-day periods following the surgical procedure. The primary explanatory variable was an initial perioperative opioid prescription fill. The rate of new persistent opioid use was calculated, and multivariable logistic regression models were used to identify variables that increased the risk of new persistent use and refill of an opioid prescription after the initial perioperative opioid prescription in the 30 days after surgery.
Main Outcome Measures:
New persistent opioid use and refill.
Results:
A total of 327,379 opioid-naive patients (mean [SD] age, 67[16] years; 178,067 [54.4%] female) who underwent ophthalmic surgery were examined. Among these patients, 14,841 (4.5%) had an initial perioperative opioid fill. The rate of new persistent opioid use was 3.4% (498 out of 14,841 patients) in patients having an initial perioperative opioid fill compared to 0.6% (1833 out of 312,538 patients) in patients without an initial perioperative fill. After adjusting for patient characteristics, initial perioperative opioid fill was independently associated with an increased odd of new persistent use (adjusted OR 6.21; 95% CI, 5.57–6.91; p<0.001).
Among patients who had filled an initial perioperative prescription, a prescription size of >=150 morphine milligram equivalent (MME) was associated with an increased odd of refill (adjusted OR 1.87; 95% CI, 1.58–2.22; p<0.001).
Conclusions:
Exposure to opioids in the perioperative period is associated with new persistent use in patients who were previously opioid-naïve. This suggests that this exposure to opioids is an independent risk factor for persistent use in patients undergoing incisional ophthalmic surgery. Surgeons should be aware of those risks to identify at-risk patients in the current national opioid crisis and minimize prescribing opioids when possible.
PRECÍS
A national claims-based cohort analysis indicates that new, persistent use of opioids medications occurs postoperatively, even after low dose prescribing of opioids at the time of ophthalmic surgery.
Since the 1990s, the increasing incidence of opioid use has resulted in a national public health crisis.1 In recent years, many studies have also been conducted to investigate the opioid prescription patterns within the ophthalmology field itself.2–4 A Medicare study examined prescription patterns among ophthalmologists in the United States and found that the median number of opioid prescriptions compared with total prescriptions was 4%.2 Recently, a health care claims database analysis and found that the rate of filled opioid prescriptions has been increasing among incisional ocular surgery over time.3 In oculoplastics, surgeons wrote an average of 45 opioid prescriptions per year4, and retina surgeons 11 per year with 6% of prescribers accounting for a third of opioid prescriptions.5
Surgeons play a unique role in the opioid crisis by often prescribing opioids to previously opioid-naive patients. Moreover, prescriptions provided for procedural care are often in excess of what patients use, and many patients are able to manage pain with opioid alternatives, such as acetaminophen or nonsteroidal anti-inflammatory (NSAIDs) medications. Prescription opioids have traditionally been an important analgesic for treating postoperative acute pain.6 Opioids prescribed after many common procedures have been associated with development of persistent opioid use defined as opioid use beyond the postoperative recovery period.6–9 Prior studies suggest that 6% of opioid-naive patients continue to fill opioid prescriptions more than 3 months after undergoing major and minor surgical procedures.8
The risk of new persistent opioid use among opioid-naive patients after a perioperative opioid prescription is well described in other procedural-based and surgical specialties,6–10 but this risk is unknown after ophthalmic procedures or surgery. Studies in other surgical disciplines identified several patient-level factors associated with persistent opioid use, including prior mental health conditions and substance use disorders.11,12 It is unclear if similar risk should be expected following ophthalmic procedures. Prescribing patterns help to understand the overall opioid-related landscape in ophthalmology. However, the occurrence of new persistent opioid use is a clinically meaningful endpoint with stronger potential implications for opioid dependence and addiction. Using a large health care claims database of opioid-naive patients undergoing ophthalmic surgery in the United States, the incidence rate and risk factors for new persistent opioid usage among previously opioid-naive patients were explored for patients receiving perioperative opioids following ophthalmic procedures.
METHODS
Dataset and study population:
In this retrospective cohort study, private insurance claims from Optum’s de-identified Clinformatics® Data Mart Database were analyzed. The University of Michigan Institutional Review Board deemed this study of deidentified data exempt from review and informed consent. Optum’s national database contains de-identified records for more than 140 million enrollees and includes data on patient demographics, insurance coverage, use of inpatient and outpatient services, and outpatient pharmacy claims. The study population consisted of all patients ages 13 years or older in this database from January 2010 to June 2017 who underwent incisional ophthalmic surgery as identified by Current Procedural Terminology (CPT) codes. The CPT codes were further subdivided into subspecialty surgery categories: anterior segment (i.e., those likely to be performed by a comprehensive ophthalmologists), cornea (i.e., those likely to be performed by a corneal surgeon), glaucoma, oculoplastics, pediatric ophthalmology (including adult strabismus), and retina categories. Non-specialty “anterior segment” surgery included cataract, trauma, and any surgery potentially performed by a comprehensive ophthalmologist (Online Supplement 1 available at http://www.aaojournal.org/). An opioid referred to all opioid analgesics or prescription opioids as defined by the Centers for Disease Control.13
Enrollees in the claims who filled an opioid prescription in the perioperative period are defined as 30 days before and up to 3 days after surgery. A refill of an opioid was defined as any fill after initial postoperative fill in the 30 days after discharge. Enrollees were excluded who had filled 1 or more opioid prescriptions within 12 months to 31 days before their surgical procedure, consistent with prior works in order to focus the analysis on opioid naive enrollees.8,14 The analysis also excluded enrollees who did not have continuous insurance enrollment and pharmacy benefit coverage in the 12 months before to 6 months after the procedure date and enrollees who underwent additional surgical procedures during the study period using subsequent procedural codes for anesthesia in the 6-month postoperative period. Enrollees who had a length of stay greater than 30 days or were not discharged home were also excluded.
Outcomes:
The primary outcome measure was new persistent opioid use, defined as fulfillment of an opioid prescription both in the 90 day and in the 91 – 180 day periods after the surgical procedure (i.e., following surgery)15. If the patient did not fill an opioid prescription during the perioperative period, any fulfillment of an opioid prescription during 4 to 90 day after discharge, plus at least one more fill in the 91 to 180 day after discharge was also defined as persistent opioid use.
Patient Variables:
Clinical and sociodemographic covariates that have previously been associated with new persistent opioid use were collected.7,11,12 Demographic and clinical variables included age, sex, race, location, median household income, highest education, and history of tobacco use. Location was treated categorically as Northwest, Midwest, South or West as defined by the U.S. Census Bureau. We used the Charlson Comorbidity Index (CCI) to identify number of comorbid conditions16 (none [0], mild [1–2], moderate [3–4], severe [>5]).
Coexisting mental health diagnoses and pain disorders were also identified using diagnosis codes, as both have been associated with pain perception and long-term opioid use.17,18 We identified mental health diagnoses using the Clinical Classification System (Agency of Healthcare Research and Quality), including adjustment disorder, anxiety, mood disorder, disruptive behavior disorders, personality disorders, suicidality or self-harm, schizophrenia and other psychotic disorders, substance use disorders, and other mental health disorders (Online Supplement 1 available at http://www.aaojournal.org/).8 Pain disorders were subdivided as follows: arthritis, back, neck pain, and other pain conditions (Online Supplement 1 available at http://www.aaojournal.org/). Comorbidities, mental health diagnoses, and pain disorders were identified in the 12 months before the surgery date.
Opioid Prescription Fills:
We examined pharmaceutical claims data to determine the type of opioid prescribed, dose, duration of prescription in days, and number of refills. Prescription medications were identified using National Drug Codes. Morphine milligram equivalents (MME) for each opioid prescription were calculated using the conversion formulae recommended by the Centers for Medicare and Medicaid Services.19
Statistical Analysis:
Demographic and clinical characteristics by surgery type were analyzed using chi-square tests for categorical variables. Differences between those with persistent opioid use and those without persistent opioid use were assessed with chi-square tests. Adjusted odds ratios (ORs) were calculated from a multivariable logistic regression model to identify variables that increased the risk of refilling a perioperative opioid prescription and the development of persistent opioid use. The p values were two-tailed, and significance was set at p < 0.05. Statistical analyses were performed using SAS, version 9.4 (SAS Institute Inc., Cary, NC).
RESULTS
Among 327,379 opioid-naive enrollees who underwent an incisional ophthalmic surgery between 2012 and 2017, 14,841 enrollees (4.5%) filled an opioid prescription in the perioperative period. The mean age and standard deviation of the study cohort was 67±16 years and there were more enrollees who were female (54.4%) and Caucasian (61.4%). Fifty-two percent of enrollees had at least one documented Charlson comorbidity. Seventeen percent of enrollees were documented as having a mood or anxiety disorder. Anterior segment procedures were the most common procedures performed (55.4%), followed by oculoplastics (27.2%) and retinal (8.7%). Table 1 displays all descriptive characteristics of the study population with chi-squared tests to test for differences between those with and without persistent use. For opioid prescriptions, the initial prescription for the 14,841 cohort had a median morphine milligram equivalent (MME) strength of 100 (interquartile range of 82.5) which equates to a median of 13.3 5mg oxycodone pills (7.5 MME) (interquartile range of 11.0).19
Table 1.
Patients Characteristics (n=327,379)
| Characteristics | Overall | No persistent opioid use | Persistent opioid use | P value | |||
|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | ||
| Age | <0.001 | ||||||
| 13–19 | 4830 | 1.48 | 4823 | 1.48 | 7 | 0.3 | |
| 20–29 | 7741 | 2.36 | 7710 | 2.37 | 31 | 1.33 | |
| 30–39 | 12714 | 3.88 | 12660 | 3.89 | 54 | 2.32 | |
| 40–49 | 20963 | 6.4 | 20843 | 6.41 | 120 | 5.15 | |
| 50–59 | 37061 | 11.32 | 36793 | 11.32 | 268 | 11.5 | |
| 60–69 | 65880 | 20.12 | 65408 | 20.12 | 472 | 20.25 | |
| 70–79 | 103662 | 31.66 | 102906 | 31.66 | 756 | 32.43 | |
| 80 and over | 74528 | 22.77 | 73905 | 22.74 | 623 | 26.73 | |
| Female | 178067 | 54.39 | 176679 | 54.35 | 1388 | 59.55 | <0.001 |
| Race | <0.001 | ||||||
| White | 200987 | 61.39 | 199578 | 61.40 | 1409 | 60.45 | |
| Asian | 12612 | 3.85 | 12563 | 3.86 | 49 | 2.1 | |
| Black | 21384 | 6.53 | 21156 | 6.51 | 228 | 9.78 | |
| Hispanic | 28571 | 8.73 | 28351 | 8.72 | 220 | 9.44 | |
| Unknown | 63825 | 19.5 | 63400 | 19.50 | 425 | 18.23 | |
| Geographic region | <0.001 | ||||||
| Northeast | 42307 | 12.92 | 42101 | 12.95 | 206 | 8.84 | |
| Midwest | 80871 | 24.7 | 80438 | 24.75 | 433 | 18.58 | |
| South | 113165 | 34.57 | 112178 | 34.51 | 987 | 42.34 | |
| West | 89074 | 27.21 | 88388 | 27.19 | 686 | 29.43 | |
| Unknown | 1962 | 0.6 | 1943 | 0.6 | 19 | 0.82 | |
| Household income | <0.001 | ||||||
| <$40K | 62655 | 19.14 | 62062 | 19.09 | 593 | 25.44 | |
| $40K–$49K | 20239 | 6.18 | 20053 | 6.17 | 186 | 7.98 | |
| $50K–$59K | 23472 | 7.17 | 23292 | 7.17 | 180 | 7.72 | |
| $60K–$74K | 32038 | 9.79 | 31818 | 9.79 | 220 | 9.44 | |
| $75K–$99K | 45931 | 14.03 | 45631 | 14.04 | 300 | 12.87 | |
| $100K+ | 84770 | 25.89 | 84345 | 25.95 | 425 | 18.23 | |
| Unknown | 58274 | 17.8 | 57847 | 17.8 | 427 | 18.32 | |
| Education | <0.001 | ||||||
| Less than 12th Grade | 1907 | 0.58 | 1886 | 0.58 | 21 | 0.9 | |
| High School Diploma | 75074 | 22.93 | 74401 | 22.89 | 673 | 28.87 | |
| Less than Bachelor Degree | 173010 | 52.85 | 171731 | 52.83 | 1279 | 54.87 | |
| Bachelor Degree Plus | 59232 | 18.09 | 58960 | 18.14 | 272 | 11.67 | |
| Unknown | 18156 | 5.55 | 18070 | 5.56 | 86 | 3.69 | |
| Ophthalmic surgery procedures | 0.339 | ||||||
| Anterior | 181492 | 55.44 | 180166 | 55.43 | 1326 | 56.89 | |
| Cornea | 9767 | 2.98 | 9700 | 2.98 | 67 | 2.87 | |
| Glaucoma | 15845 | 4.84 | 15727 | 4.84 | 118 | 5.06 | |
| Oculoplastics | 89185 | 27.24 | 88564 | 27.25 | 621 | 26.64 | |
| Retina | 28496 | 8.7 | 28308 | 8.71 | 188 | 8.07 | |
| Strabismus | 2594 | 0.79 | 2583 | 0.79 | 11 | 0.47 | |
| History of tobacco use | 23093 | 7.05 | 22814 | 7.02 | 279 | 11.97 | <0.001 |
| Charlson Comorbidity Index | <0.001 | ||||||
| None (CCI 0) | 157326 | 48.06 | 156580 | 48.17 | 746 | 32 | |
| Mild (CCI 1, 2) | 99851 | 30.5 | 99101 | 30.49 | 750 | 32.18 | |
| Moderate (CCI 3, 4) | 42682 | 13.04 | 42224 | 12.99 | 458 | 19.65 | |
| Severe (CCI ≥5) | 27520 | 8.41 | 27143 | 8.35 | 377 | 16.17 | |
| Mental health disorders | |||||||
| Adjustment | 5024 | 1.53 | 4982 | 1.53 | 42 | 1.8 | 0.292 |
| Anxiety | 25932 | 7.92 | 25619 | 7.88 | 313 | 13.43 | <0.001 |
| Mood | 30523 | 9.32 | 30117 | 9.27 | 406 | 17.42 | <0.001 |
| Disruptive | 2300 | 0.7 | 2277 | 0.7 | 23 | 0.99 | 0.099 |
| Personality | 321 | 0.1 | 316 | 0.1 | 5 | 0.21 | 0.071 |
| Psychosis | 2597 | 0.79 | 2564 | 0.79 | 33 | 1.42 | <0.001 |
| Alcohol or substance abuse disorders | 6959 | 2.13 | 6865 | 2.11 | 94 | 4.03 | <0.001 |
| Suicide or self-harm | 321 | 0.1 | 311 | 0.1 | 10 | 0.43 | <0.001 |
| Other | 5867 | 1.79 | 5806 | 1.79 | 61 | 2.62 | 0.003 |
| Pain disorders | |||||||
| Back | 59195 | 18.08 | 58439 | 17.98 | 756 | 32.43 | <0.001 |
| Neck | 25628 | 7.83 | 25350 | 7.8 | 278 | 11.93 | <0.001 |
| Arthritis | 161608 | 49.36 | 160067 | 49.24 | 1541 | 66.11 | <0.001 |
| Other pain conditions | 82423 | 25.18 | 81616 | 25.11 | 807 | 34.62 | <0.001 |
| Initial postop fill | <0.001 | ||||||
| No | 312538 | 95.47 | 310705 | 95.59 | 1833 | 78.64 | |
| Yes | 14841 | 4.53 | 14343 | 4.41 | 498 | 21.36 | |
Of the 14,841 patients who filled an opioid prescription during the perioperative period, 498 (3.3%) continued to fill opioid prescriptions more than 90 days after surgery and constituted new persistent opioid use. Among the 312,538 patients who did not fill an opioid prescription during the perioperative period, 1,833 (0.6%) had an opioid prescription fill during 4 to 90 days after surgery date AND at least one more fill in the 91 to 180 days after surgery date and thus also accounted for new persistent use.
Table 2 displays the logistic regression examining the likelihood of persistent opioid use. After adjusting for significant covariates, patients with the highest ORs of persistent opioid use included those with an initial perioperative opioid fill (adjusted OR 6.21; 95% CI, 5.57–6.91; p<0.001), those in the age range 50–59 (adjusted OR 1.43; 95% confidence interval [CI], 1.06–1.92; p = 0.019) compared to the reference of patients age ranges 30–39, female sex (adjusted OR 1.10; 95% CI, 1.00–1.20; p = 0.042), Black race (adjusted OR 1.20; 95% CI, 1.03–1.39; p = 0.017) compared to White race, living in the South (adjusted OR 1.58; 95% CI, 1.36–1.85; p < 0.001) or West (adjusted OR 1.52; 95% CI, 1.30–1.78; p < 0.001) compared to living in the Northeast, lower household income $40K–$49K (adjusted OR 1.27, 95% CI, 1.04–1.54; p = 0.019) compared to those with an income range of $60K–$74K, highest education less than a bachelor’s degree (adjusted OR 1.36, 95% CI, 1.18–1.56; p < 0.001) compared to at least or more than a Bachelor’s Degree, history of tobacco use (adjusted OR 1.25; 95% CI, 1.09–1.43; p = 0.001), higher CCI (severe CCI; adjusted OR 2.20, 95% CI, 1.92–2.53; p < 0.001), presence of anxiety (adjusted OR 1.26, 95% CI, 1.10–1.44; p < 0.001), presence of a mood disorder (adjusted OR 1.42, 95% CI, 1.26–1.60; p < 0.001).
Table 2.
Logistic regression examining the likelihood of persistent opioid use (n=327,379)
| Adjusted OR | CI_low | CI_high | p- value | |
|---|---|---|---|---|
| Age | ||||
| 30–39 (reference) | 1 | NA | NA | NA |
| 13–19 | 0.39 | 0.18 | 0.87 | 0.021 |
| 20–29 | 0.97 | 0.62 | 1.51 | 0.886 |
| 40–49 | 1.27 | 0.92 | 1.75 | 0.150 |
| 50–59 | 1.43 | 1.06 | 1.92 | 0.019 |
| 60–69 | 1.20 | 0.90 | 1.61 | 0.209 |
| 70–79 | 1.12 | 0.84 | 1.49 | 0.450 |
| 80 and over | 1.18 | 0.88 | 1.58 | 0.273 |
| Female | 1.10 | 1.00 | 1.20 | 0.042 |
| Race | ||||
| White (reference) | 1 | NA | NA | NA |
| Asian | 0.62 | 0.47 | 0.83 | 0.001 |
| Black | 1.20 | 1.03 | 1.39 | 0.017 |
| Hispanic | 0.87 | 0.75 | 1.01 | 0.067 |
| Unknown | 0.96 | 0.85 | 1.09 | 0.551 |
| Geographic region | ||||
| Northeast (reference) | 1 | NA | NA | NA |
| Midwest | 1.00 | 0.84 | 1.18 | 0.977 |
| South | 1.58 | 1.36 | 1.85 | <0.001 |
| West | 1.52 | 1.30 | 1.78 | <0.001 |
| Unknown | 1.74 | 1.08 | 2.81 | 0.023 |
| Household income | ||||
| $60K–$74K (reference) | 1 | NA | NA | NA |
| <$40K | 1.18 | 1.01 | 1.39 | 0.040 |
| $40K–$49K | 1.27 | 1.04 | 1.54 | 0.019 |
| $50K–$59K | 1.06 | 0.87 | 1.30 | 0.553 |
| $75K–$99K | 1.01 | 0.85 | 1.21 | 0.874 |
| $100K+ | 0.90 | 0.76 | 1.06 | 0.213 |
| Unknown | 1.21 | 1.02 | 1.44 | 0.033 |
| Education | ||||
| Bachelor Degree Plus (reference) | 1 | NA | NA | NA |
| Less than 12th Grade | 1.58 | 1.00 | 2.52 | 0.052 |
| High School Diploma | 1.44 | 1.22 | 1.69 | <0.001 |
| Less than Bachelor Degree | 1.36 | 1.18 | 1.56 | <0.001 |
| Unknown | 0.72 | 0.55 | 0.95 | 0.019 |
| Ophthalmic surgery procedures | ||||
| Anterior (reference) | 1 | NA | NA | NA |
| Cornea | 0.67 | 0.52 | 0.86 | 0.002 |
| Glaucoma | 1.03 | 0.85 | 1.25 | 0.763 |
| Oculoplastics | 0.82 | 0.74 | 0.91 | <0.001 |
| Retina | 0.73 | 0.63 | 0.86 | <0.001 |
| Strabismus | 0.32 | 0.17 | 0.58 | <0.001 |
| History of tobacco use | 1.25 | 1.09 | 1.43 | 0.001 |
| Charlson Comorbidity Index | ||||
| None (CCI 0) (reference) | 1 | NA | NA | NA |
| Mild (CCI 1, 2) | 1.36 | 1.22 | 1.52 | <0.001 |
| Moderate (CCI 3, 4) | 1.85 | 1.63 | 2.10 | <0.001 |
| Severe (CCI ≥5) | 2.20 | 1.92 | 2.53 | <0.001 |
| Mental health disorders | ||||
| Adjustment | 0.93 | 0.68 | 1.27 | 0.626 |
| Anxiety | 1.26 | 1.10 | 1.44 | <0.001 |
| Mood | 1.42 | 1.26 | 1.60 | <0.001 |
| Disruptive | 1.39 | 0.91 | 2.13 | 0.130 |
| Personality | 1.07 | 0.43 | 2.69 | 0.883 |
| Psychosis | 0.88 | 0.61 | 1.26 | 0.473 |
| Alcohol or substance abuse disorders | 1.20 | 0.96 | 1.50 | 0.117 |
| Suicide or self-harm | 1.76 | 0.89 | 3.47 | 0.102 |
| Other | 1.06 | 0.82 | 1.37 | 0.671 |
| Pain disorders | ||||
| Back | 1.69 | 1.53 | 1.86 | <0.001 |
| Neck | 1.02 | 0.89 | 1.17 | 0.739 |
| Arthritis | 1.44 | 1.31 | 1.58 | <0.001 |
| Other pain conditions | 1.15 | 1.05 | 1.26 | 0.002 |
| Initial postop fill | 6.21 | 5.57 | 6.91 | <0.001 |
CI: Confidence Interval
C-statistics=0.714, Hosmer and Lemeshow Goodness-of-Fit: p value=0.095
Compared with patients undergoing anterior segment surgery; enrollees undergoing corneal, oculoplastics, retina, and strabismus surgery had a decreased likelihood of persistent opioid use. In addition, patients aged 13–19 (adjusted OR 0.39, 95% CI, 0.18–0.87, p = 0.021) and Asian race (adjusted OR 0.62, 95% CI, 0.47–0.83; p = 0.001) were less likely to develop persistent opioid use.
Continuing to fill opioid prescriptions in the 30 days after procedure or surgery date among opioid-naïve patients who filled an initial prescription during the perioperative period did exist at higher odds for some populations (Online Supplement 2 available at http://www.aaojournal.org/). Enrollees with lower income <$40K compared to those with $60K–74K, had higher odds of refilling an opioid prescription. Enrollees with a prescription size of >=150 MME (adjusted OR 1.87; 95% CI, 1.58–2.22; p<0.001), back disorders (adjusted OR 1.23; 95% CI, 1.06–1.43; p=0.006), or neck disorders (adjusted OR 1.31; 95% CI, 1.08–1.59; p=0.006) had higher likelihood to refill an opioid prescription. Compared to anterior segment surgery, those who had strabismus surgery were less likely to refill an opioid prescription (OR 0.7, 95% CI, 0.51–0.95, p = 0.024).
The type of opioid prescribed is provided in the most recent year (2017) to contextualize the type of opioid prescribed by ophthalmic surgeons (Table 3). Hydrocodone was the most commonly prescribed (45%) followed by oxycodone (18.4%) and tramadol (18.2%).
Table 3.
Type of opioid for initial postop prescriptions in 2017.
| Type of opioid | Morphine Milligram Equivalent (MME) Unit for the opioid type | Number of prescription (%) | |
|---|---|---|---|
| Hydrocodone (mg) | 1 | 607 | (45.0%) |
| Oxycodone (mg) | 1.5 | 249 | (18.4%) |
| Tramadol (mg) | 0.1 | 246 | (18.2%) |
| Codeine (mg) | 0.15 | 242 | (17.9%) |
| Hydromorphone (mg) | 4 | 2 | (0.2%) |
| Meperidine (mg) | 0.1 | 2 | (0.2%) |
| Morphine (mg) | 1 | 1 | (0.1%) |
| Fentanyl (meg) | 7.2 | 1 | (0.1%) |
| Total | 1350 | (100%) | |
DISCUSSION
In this study based on a large administrative claims data, 3.4% of opioid-naive enrollees had new, persistent opioid use following ophthalmic surgery when they had their initial opioid prescription during peri-operative period. This result is consistent with reported rates of new, persistent opioid use of 3–11% at least 90 days postoperatively in other surgical disciplines.6,9,12,20 However, the median MME of the initial opioid prescription in our cohort was 100 mg. So new persistent use existed even though ophthalmic surgeons were prescribing less than the amount expected to be necessary after minor surgery21 and less than the <200 mg MME recommended by state regulatory agencies.22 Thus it is surprising that rates of persistent use was still equivalent to other surgical disciplines which prescribe greater MME doses perioperatively. For surgical procedures such as arthroscopic rotator cuff repair in orthopedic surgery or hysterectomy in gynecologic surgery, it is recommended that no more than 150 MME should be prescribed to opioid naïve patients on discharge.23 In our study, initial perioperative opioid fill was the risk factor with the highest OR (adjusted OR 6.21; 95% CI, 5.57–6.91; p<0.001) suggesting that eliminating the initial opioid prescription may have the largest impact in reducing persistent opioid use following ophthalmic surgery.
In this ophthalmology cohort, patient risk factors for new persistent use were age 50–59, female sex, Black race, lower household income, education less than a bachelor’s degree, residence in the West and South, history of tobacco use, higher CCI scores and initial perioperative opioid fill. Patients with anxiety and mood disorders had higher odds of new persistent postoperative opioid use, consistent with the surgical literature; however, there was no correlation of new persistent use if the patient had with alcohol or other substance abuse disorders.9 Sex differences in prescription opioid use and patterns of use exist across surgical disciplines, with females being more likely than males to have prolonged opioid fills after surgery.24 It remains unclear what is the cause of sex differences in prescription opioid use. Physician prescribing difference between sociodemographic groups warrants more research.24
Compared with other surgical specialists, ophthalmologists prescribe overall low rates of opioids, likely due to the nature of less invasive surgery.2–4 Patel and Sternberg examined Medicare Part D Prescriber Data and found that the median number of opioid prescriptions compared with total prescriptions was 4% among ophthalmologists, which is lower than the national mean of 6.8%.2 That study also explored geographic trends with Southern states having tended to have an increased number of opioid prescriptions written per physician. In the current work, patients living in the South had 58% higher odds of persistent opioid use; however, claims data are not population-based data so implied differences should be interpreted cautiously. This result indicates that the harm of higher prescribing rates are potentially compounded by higher persistent use perhaps due to differences in state regulation of opioid prescriptions.25 Many states have enacted laws to establish prescription drug monitoring programs to address these issues at a policy level.26,27 The enactment of Michigan Opioid Laws in 2017 and 2018 led to a reduction in opioid prescriptions for oculoplastic and orbital procedures.28
Advances in surgical instrumentation and technique have made incisional ocular surgery less traumatic to the eye in the last decade. However, in 2020 Kolomeyer et al. found that the rate of filled opioid prescriptions was increasing for all types of incisional ocular surgery over time in Optum, the same health care claims database used for the current analysis.3 Particularly, trauma, strabismus, and retina surgeries had the highest rates of filled opioid prescriptions per surgery.3 While fill rates increased in those domains, our analysis showed patients in different domains who underwent cornea, oculoplastics, retina, and pediatic/strabismus surgeries had lower odds of developing persistent use in comparison with non-specialty anterior segment surgery. Based on our knowledge of the pain associated with non-specialty anterior segment surgery, when compared to the other surgeries we studied, there is no obvious anatomic or physiologic reason why this surgery should require more opioids to control pain nor lead to more persistent use. Additionally, our study found that those 13–19 years of age had lower odds of new persistent use compared to the reference group of 30–39 years. In younger age groups, refills may be likely dictated by guardians, so the lines between persistent use, who controls and fills prescriptions, and addiction are blurred. Discrepancies between our work and those of others may underscore the differences between prescription patterns and persistent use.
Current post-operative opioid prescribing practices are not always built around standardized guidelines and are typically at the discretion of the surgeon.21 There are differences among surgeons. For example, male surgeons are reported to write higher numbers of prescriptions than female surgeons.3,4 Interventions to limit patients’ exposure to unnecessary postoperative opioids can occur at the state legislative and the provider level. In a recent study, patients’ opioid use and fill rates after corneal surgery were actually lower than the volume prescribed by the provider.29,30 Once aware, providers lowered their volume of prescribed opioids. Then, after physicians lowered their opioid prescription volumes, patients concomitantly decreased their use of opioids further still after corneal surgery. In 2020, guidelines for opioid prescribing were tested in a large ophthalmic clinic system.31 Those guidelines became included in a continuing medical education activity through the American Board of Ophthalmology.32
While these trends are consistent with the published literature, our study has limitations. Health insurance claims data are limited to participants with private insurance. The results may not be generalizable to patients without insurance or those with exclusively state or federal health care coverage, such as Medicare or Medicaid programs. Furthermore, the differences in geography shown by these data may not be truly representative of the geographic differences because it is based on the population covered by Optum and does not include other private payers, Medicare, Medicaid and other federal sources. Future work with population-based data that are inclusive of private payers, Medicare, Medicaid, and other federal sources (e.g. VA) could provide information on geographical differences in prescribing. Second, opioid prescription fills were used as a proxy for opioid consumption and did not capture actual opioid consumption. As such, this study does not take into account the possibility of that prescriptions could be filled for other use, such as diversion for illicit sales to others.33 In addition, patients who procured opioid prescriptions outside the Optum database such as from a local emergency room were not accounted for. This may limit the accuracy of prescription opioid fills. Finally, analysis was restricted to opioid naive patients who did not fill an opioid prescription one year prior to their surgery date. It is unclear if these patients had used any opioids prior to one year. And it thus does not also include patients with recent prior use who may be at higher risk of persistent use.
In this cohort of previously opioid-naive patients who were prescribed opioids after ophthalmic surgery, despite the fact that prescriptions were not at high morphine equivalent quantities, led to a 3.4% incidence of continued use opioids more than three months after surgery. Risk factors for new persistent opioid use included those age 50–59, female sex, Black race, lower household income, education less than a bachelor’s degree, residence in the West and South, history of tobacco use, higher CCI scores, anxiety and mood disorders, and initial perioperative opioid fill. New persistent opioid use is an important and underrecognized complication of perioperative care in ophthalmology and other surgical specialities.8 Given the incidence of new persistent use among postsurgical ophthalmology patients, surgeons need to play an active role in counseling patients on the risks of opioids and in minimizing the use of postoperative opioids when possible.33 Further study is needed to develop evidence-based guidelines in order to reduce unnecessary opioid prescribing and to counsel at-risk patients.
Supplementary Material
Online Supplement 1: Diagnosis codes and Classification systems used to generate data analysis. First list is the Current Procedural Terminology (CPT) codes for ophthalmic surgeries which are subdivided into subspecialty categories: anterior segment, cornea, glaucoma, oculoplastics, pediatric ophthalmology, retina categories. Non-specialty “anterior segment” surgery included cataract, trauma, and any surgery potentially performed by a comprehensive ophthalmologist. Other lists include the Clinical Classification System Codes for mental health conditions, in addition to ICD-9 and ICD-10 codes for pain disorders
Online Supplement 2: Logistic regression examining the likelihood of opioid refill (n=14841, eligible patients with an initial opioid fill). Observing the continuation of refilling opioid prescriptions 30 days after procedure or surgery date among opioid-naïve patients who filled an initial prescription during the perioperative period.
Financial Support:
Vitreo-Retinal Surgery Foundation Award (CU, YY), American College of Surgeons and the American Foundation for Surgery of the Hand (JFW), Centers for Disease Control and Prevention (E20182818-00 MA-2018 Master Agreement Program, [JFW]), Michigan Department of Health and Human Services (E20180672-00 Michigan DHHS - MA-2018 Master Agreement Program [JFW]), National Institute of Arthritis and Musculoskeletal and Skin Diseases (P50 AR070600 [JFW]), National Institute on Drug Abuse (RO1 DA042859 [JFW]), Substance Abuse and Mental Health Administration (SAMHSA: E20180568-00 MA-2018 Master Agreement Program [JFW]), University of Michigan School Dean’s Office—Michigan Genomics Initiative and Precision Health Initiative (JFW), National Eye Institute (R01EY031033 [MAW]), Research to Prevent Blindness, Career Advancement Award (MAW). The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures: Unrelated to submitted work. 3M Health Information Systems (JFW, unpaid consultant), Alcon (YY, consultant), Allergan (YY, consultant), Alimera (YY, consultant) Phoenix Clinical (YY, unpaid consultant), Regeneron (YY, consultant). Vortex Surgical (MAW, equity), Aviceda Ophthalmics (MAW, equity).
Conflict of interest: The authors have no proprietary or commercial interest in any of the materials discussed in this article.
REFERENCES
- 1.Rudd RA, Seth P, David F, Scholl L. Increases in Drug and Opioid-Involved Overdose Deaths — United States, 2010–2015. MMWR Morb Mortal Wkly Rep 2016;65:1445–1452. [DOI] [PubMed] [Google Scholar]
- 2.Patel S, Sternberg P. Association between opioid prescribing patterns and abuse in ophthalmology. JAMA Ophthalmol 2017;135:1216–1220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Vanderbeek BL, Kolomeyer AM, Yu Y. Association of Opioids with Incisional Ocular Surgery. JAMA Ophthalmol 2019;137:1283–1291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Charlson ES, Feng PW, Bui A, et al. Opioid Prescribing Patterns among American Society of Ophthalmic Plastic and Reconstructive Surgery Members in the Medicare Part D Database. In: Ophthalmic Plastic and Reconstructive Surgery.Vol 35. Lippincott Williams and Wilkins; 2019:360–364. [DOI] [PubMed] [Google Scholar]
- 5.Ung C, Ung R, Yonekawa Y. Opioid Prescribing Patterns Among Retina Specialists in the United States. Ophthalmol Retin 2020. [DOI] [PubMed] [Google Scholar]
- 6.Clarke H, Soneji N, Ko DT, et al. Rates and risk factors for prolonged opioid use after major surgery: Population based cohort study. BMJ 2014;348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Alam A, Gomes T, Zheng H, et al. Long-term analgesic use after low-risk surgery: A retrospective cohort study. Arch Intern Med 2012;172:425–430. [DOI] [PubMed] [Google Scholar]
- 8.Brummett CM, Waljee JF, Goesling J, et al. New persistent opioid use after minor and major surgical procedures in us adults. JAMA Surg 2017;152:170504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sun EC, Darnall BD, Baker LC, MacKey S. Incidence of and risk factors for chronic opioid use among opioid-naive patients in the postoperative period. JAMA Intern Med 2016;176:1286–1293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Soneji N, Clarke HA, Ko DT, Wijeysundera DN. Risks of developing persistent opioid use after major surgery. JAMA Surg 2016;151:1083–1084. [DOI] [PubMed] [Google Scholar]
- 11.Harbaugh CM, Nalliah RP, Hu HM, et al. Persistent opioid use after wisdom tooth extraction. JAMA - J Am Med Assoc 2018;320:504–506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Harbaugh CM, Lee JS, Hu HM, et al. Persistent opioid use among pediatric patients after surgery. Pediatrics 2018;141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.CDC. Commonly Used Terms | Drug Overdose | CDC Injury Center. Centers Dis Control Prev Natl Cent Inj Prev Control; 2021. Available at: https://www.cdc.gov/drugoverdose/opioids/terms.html [AccessedMarch 10, 2021]. [Google Scholar]
- 14.Howard R, Fry B, Gunaseelan V, et al. Association of Opioid Prescribing with Opioid Consumption after Surgery in Michigan. JAMA Surg 2019;154:184234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Keller DS, Kenney BC, Harbaugh CM, et al. A national evaluation of opioid prescribing and persistent use after ambulatory anorectal surgery. In: Surgery (United States). Mosby Inc.; 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol 1994;47:1245–1251. [DOI] [PubMed] [Google Scholar]
- 17.Campbell G, Nielsen S, Bruno R, et al. The Pain and Opioids in Treatment study: Characteristics of a cohort using opioids to manage chronic non-cancer pain. Pain 2015;156:231–242. [DOI] [PubMed] [Google Scholar]
- 18.Goesling J, Henry MJ, Moser SE, et al. Symptoms of Depression Are Associated with Opioid Use Regardless of Pain Severity and Physical Functioning among Treatment-Seeking Patients with Chronic Pain. J Pain 2015;16:844–851. [DOI] [PubMed] [Google Scholar]
- 19.Centers for Medicare and Medicaid Services. Opioid Oral Morphine Milligram Equivalent ( MME ) Conversion Factors. Centers Medicaire Medicaid Serv 2018:2. [Google Scholar]
- 20.Lee JSJ, Hu HM, Edelman AL, et al. New persistent opioid use Among patients with cancer after curative-intent surgery. J Clin Oncol 2017;35:4042–4049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Thiels CA, Anderson SS, Ubl DS, et al. Wide Variation and Overprescription of Opioids after Elective Surgery. Ann Surg 2017;266:564–573. [DOI] [PubMed] [Google Scholar]
- 22.Anon. Opioid Prescribing Work Group: Acute Pain Prescribing Recommendations. 2020.
- 23.Overton HN, Hanna MN, Bruhn WE, et al. Opioid-Prescribing Guidelines for Common Surgical Procedures: An Expert Panel Consensus. J Am Coll Surg 2018;227:411–418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Serdarevic M, Striley CW, Cottler LB. Sex differences in prescription opioid use. Curr Opin Psychiatry 2017;30:238–246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Paulozzi LJ, Strickler GK, Kreiner PW, Koris CM. Morbidity and Mortality Weekly Report Centers for Disease Control and Prevention MMWR Editorial and Production Staff (Weekly) MMWR Editorial Board. Cent Surveillance, Epidemiol Lab Serv Centers Dis Control Prev 2015;64. [Google Scholar]
- 26.Davis CS, Pierce M, Dasgupta N. Evolution and convergence of state laws governing controlled substance prescription monitoring programs, 1998–2011. Am J Public Health 2014;104:1389–1395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Brady JE, Wunsch H, DiMaggio C, et al. Prescription drug monitoring and dispensing of prescription opioids. Public Health Rep 2014;129:139–147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Xie Y, Joseph AW, Rudy SF, et al. Change in Postoperative Opioid Prescribing Patterns for Oculoplastic and Orbital Procedures Associated with State Opioid Legislation. JAMA Ophthalmol 2020;139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Inciardi JA, Surratt HL, Lugo Y, Cicero TJ. The Diversion of Prescription Opioid Analgesics. Law Enforc Exec Forum 2007;7:127–141. [PMC free article] [PubMed] [Google Scholar]
- 30.Woodward MA, Zhang Y, Tannen B, et al. Association of Limiting Opioid Prescriptions with Use of Opioids after Corneal Surgery. JAMA Ophthalmol 2020;138:76–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Starr MR, Patel SV., Bartley GB, Bothun ED. Impact of Standardized Prescribing Guidelines on Postoperative Opioid Prescriptions after Ophthalmic Surgery. Ophthalmology 2020;127:1454–1459. [DOI] [PubMed] [Google Scholar]
- 32.Woodward MA, Waljee J. Decisions Ahead of Time: The Power of Guidelines to Change Opioid Prescribing in Today’s World. Ophthalmology 2020;127:1460–1461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Chou R, Gordon DB, De Leon-Casasola OA, et al. Management of postoperative pain: A clinical practice guideline from the American pain society, the American society of regional anesthesia and pain medicine, and the American society of anesthesiologists’ committee on regional anesthesia, executive commi. J Pain 2016;17:131–157. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Online Supplement 1: Diagnosis codes and Classification systems used to generate data analysis. First list is the Current Procedural Terminology (CPT) codes for ophthalmic surgeries which are subdivided into subspecialty categories: anterior segment, cornea, glaucoma, oculoplastics, pediatric ophthalmology, retina categories. Non-specialty “anterior segment” surgery included cataract, trauma, and any surgery potentially performed by a comprehensive ophthalmologist. Other lists include the Clinical Classification System Codes for mental health conditions, in addition to ICD-9 and ICD-10 codes for pain disorders
Online Supplement 2: Logistic regression examining the likelihood of opioid refill (n=14841, eligible patients with an initial opioid fill). Observing the continuation of refilling opioid prescriptions 30 days after procedure or surgery date among opioid-naïve patients who filled an initial prescription during the perioperative period.