Abstract
Background
Despite targeted national quality improvement initiatives, opioid prescription and use patterns after lower limb amputation are not well studied.
Objective
To determine risk factors associated with extended opioid use and rates of high‐risk opioid prescribing after lower limb amputation.
Design and setting
A retrospective population‐based study was performed by querying a 20% national sample of Medicare claims of patients undergoing lower limb amputation from 2009 to 2018. Patients were excluded if they did not discharge home or had subsequent surgery 180 days after discharge. Extended opioid use was defined as one opioid prescription filled within 90 days after discharge and another within 90–180 days. Chronic opioid users were defined as having a 180‐day supply over the preoperative surgical period and at least one opioid prescription within 60 days of surgery. High‐risk opioid prescribing included overlapping opioid prescriptions, overlapping benzodiazepine‐opioid prescriptions, high daily doses >100 oral morphine milligram equivalents, multiple prescribers, and new long‐acting opioid use within 90 days postoperatively. Multilevel mixed‐effects logistic regression and the Cochran–Armitage Trend Test was used.
Main outcome
New persistent and prolonged opioid use following lower extremity amputation.
Results
The study included 5191 patients (nonchronic opioid users = 3759, chronic opioid users = 1432). Among the nonchronic opioid users, 873 (23.2%) had new persistent opioid use after surgery. In these patients, age < 55 years (odds ratio [OR], 1.41, p = .04), concurrent gabapentinoid fills (OR, 1.92, p < .001), and sedative/anxiolytic fills (OR, 1.75, p < .001) were associated with new persistent use. Of the chronic opioid users, 1306 (91.2%) had prolonged opioid use after surgery. Rates of high‐risk prescribing did not change significantly during the study period, and 82.3% of chronic opioid users were exposed to high‐risk prescribing.
Conclusions
Extended opioid use is common following lower extremity amputation. Moreover, high‐risk prescribing practices remain common and have not changed over time.
INTRODUCTION
Limb loss is expected to affect over 3.5 million Americans within the next 3 decades. 1 , 2 Unfortunately, for patients who undergo amputations, chronic postamputation pain (PAP) is a burdensome condition that limits daily activities and quality of life. 3 , 4 , 5 Chronic PAP is common following amputation, affecting 30% to 80% of patients living with amputations, and remains a significant clinical challenge after limb loss. 6 , 7 , 8 , 9 , 10 , 11 The current national health care costs for managing chronic PAP, estimated at $10 billion annually, are likely to increase. 1 , 6
Patients experiencing chronic PAP are treated with a variety of therapeutic interventions including rehabilitation (such as mirror visual feedback and cognitive behavioral therapy), medical treatment (such as opioids and gabapentinoids), and revision surgery. 11 , 12 , 13 While undergoing these treatments, many patients use opioid medications. 11 Several studies have demonstrated the unsatisfactory analgesic efficacy of long‐term opioids for chronic pain. 14 , 15 Despite this, reported rates of long‐term opioid prescriptions are high. 14 Furthermore, in one systematic review of 38 studies, individuals with chronic pain often demonstrated high‐risk opioid use, such a nonmedical use or use for unintended reasons, ranging from 21% to 29%. 16 Such patterns of high‐risk opioid use can lead to opioid overdose, dependence, and death. Recognizing the need for improved national guidance on opioid prescribing practices, the Centers for Disease Control and Prevention (CDC) published the CDC Guideline for Prescribing Opioids for Chronic Pain in 2016. 17 This guideline comprised 12 recommendations to provide safer and more effective care for patients with chronic pain to reduce opioid use disorder and overdose. 18 However, the impact of this policy on opioid use and misuse in patients with chronic pain remains unclear. 19
Although risk factors for extended postoperative opioid use and high‐risk opioid prescribing have been characterized for some surgical procedures, 20 , 21 , 22 , 23 , 24 , 25 opioid use and prescribing patterns for patients with chronic PAP following lower limb amputations remain poorly understood. Few studies have characterized opioid use in people with lower extremity amputations, who are often older, have multiple comorbid conditions, and are at high risk of polypharmacy. 26 The purpose of this study is to identify predictors of new persistent and prolonged opioid use and to characterize rates of high‐risk opioid prescribing among adult patients undergoing lower limb amputations.
METHODS
Data source and patient population
We included adult patients 65 years and older from a 20% national sample of Medicare fee‐for‐service claims data who had undergone a lower limb amputation between January 1, 2009 and June 30, 2018. Patients were included only if they had continuous Medicare Parts A, B, and D coverage during the year before and after surgery. Patients were identified using Current Procedural Terminology (CPT)‐4 codes for unilateral or bilateral lower limb amputations including hip and below hip, knee and below knee, and ankle and below ankle amputations (Supporting Information I‐Table I). Patients were excluded if (1) they did not discharge home, (2) had a length of stay longer than 30 days, or (3) had a subsequent surgery, based on anesthesia CPT‐4 codes for surgery, within 180 days after discharge. Patients were excluded if they did not discharge home because their postoperative opioid use was not available in the database. This study was deemed exempt by the University of Michigan Institutional Review Board.
Defining opioid use
Opioid prescriptions were identified from Part D claims files (Supporting Information II‐Table II). To stratify patients preoperatively according to their opioid use, we defined the preoperative opioid exposure period as starting 365 days before surgery and concluding 31 days before surgery. We then categorized the perioperative surgical period as 30 days before surgery and including their hospital stay. Nonchronic users included both opioid naïve and intermittent users. Opioid naïve users were defined as patients who did not fill an opioid prescription in the preoperative opioid exposure period. Intermittent opioid users were defined as patients who filled at least one prescription in the preoperative opioid exposure period but did not meet the criteria of chronic opioid use. Chronic opioid users were defined as having a 180‐day supply over the preoperative surgical period and at least one opioid prescription within 60 days of surgery. Opioid prescription size was measured as oral morphine milligram equivalent (OME) and calculated by multiplying strength per unit, quantity, and OME conversion factors.
Extended opioid use was defined as obtaining one opioid prescription filled in the 90 days after discharge and another fill in the 91–180 days after discharge. The term new persistent use was used to refer to extended opioid use in individuals who were classified as nonchronic users (both naïve and intermittent users) before surgery. In contrast, the term prolonged use was defined as extended opioid use in individuals who were chronic opioid users preoperatively. High‐risk opioid prescribing practice included documentation of overlapping opioid prescriptions, overlapping benzodiazepine‐opioid prescriptions, high daily doses >100 OMEs, multiple prescribers, or new long‐acting opioid use within 90 days postoperatively. 27 , 28 Prescription fill data were obtained from the 20% national sample of Medicare fee‐for‐service claims database. Opioid use terminology is clarified in Figure S1.
Independent variables
Demographic, socioeconomic, and clinical variables included gender, age, race, ethnicity, reason for Medicare eligibility, level of amputation, and provider specialty—listed as the provider who performed the surgery. We also classified patients using the Charlson Comorbidity Index (none to mild = 0–2, moderate = 3–4, and severe ≥5) to describe the severity of patient comorbidities. Additionally, based on the Healthcare Cost and Utilization Project Clinical Classification System, we identified preexisting mental and social health disorders (adjustment disorder, anxiety, disruptive, mood, personality, schizophrenia, suicidality, other mental health disorders, and drug and substance abuse) as well as preexisting pain diagnoses (arthritis, back pain, neck pain, and other pain) within the year before the surgery using International Classification of Diseases, Ninth Revision (ICD‐9) and Tenth Revision (ICD‐100 diagnosis codes (Supporting Informations III and IV‐Tables III and IV). To better characterize the simultaneous use of opioids and nonopioid pain medications, we documented concurrent filled prescriptions of gabapentinoids and separately all other controlled substance medications (Supporting Information II‐Table II), including prescriptions for benzodiazepines, sedatives, hypnotics, and anxiolytics within 30 days before admission.
Outcomes
The primary outcome of interest was rates of new persistent opioid use or prolonged opioid use following lower limb amputation. The secondary outcome of interest was rates of high‐risk opioid prescribing following lower limb amputation and postoperative opioid prescription size filled within 3 months after surgery. In addition, a direct comparison of the rates pre‐and postimplementation of the 2016 CDC guideline was performed to evaluate the impact on prescribing patterns following publication of these guidelines.
Statistical analyses
Demographic information, clinical characteristics, and trends in opioid use were analyzed using descriptive statistics. We used the Cochran–Armitage trend test to determine trends in high‐risk prescribing over time and chi‐square test to compare rates before and after the implementation of the CDC guideline in 2016. We used two‐level mixed‐effects logistic and linear regression models to account for the clustering effect of surgeons who operated on one or more patients in the study population. Two‐level mixed‐effects logistic regression models were used to determine the relationship between new persistent or prolonged opioid use and patient characteristics. We used two‐level mixed‐effects linear regression models to estimate the effect of patient characteristics on prescription size. For the linear regression model to be valid, one of the assumptions is residuals estimated based on the regression model to be normally distributed. This assumption was violated, so we modeled log‐scale prescription size and reported the effect of patient characteristics as multiplicative effects on prescription size after the beta coefficients were exponentiated. For this analysis 95% confidence intervals (CIs) are presented. Analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC) and statistical significance was defined as p < .05.
RESULTS
Baseline demographic and clinical characteristics
In total, 5191 patients were included (opioid naïve = 1564; intermittent opioid users = 2195; chronic opioid users = 1432, Figure S2; Table 1). Most patients were male (64.3%) and identified as non‐Hispanic White (62.2%). Overall, 47.5% of patients filled opioid prescriptions in the perioperative surgical period (31.0% of nonchronic users compared to 90.9% of chronic opioid users). Nearly 15% of all patients (8.6% non‐chronic users compared to 29.7% chronic users) filled prescriptions for benzodiazepines, sedatives, hypnotics, and anxiolytics during the perioperative period and more than half (54.6%) of all patients filled gabapentinoid prescriptions during the same period (47.8% nonchronic compared to 72.5% chronic users). Chronic opioid users were more likely to identify as non‐Hispanic White (73.2%). A higher proportion of non‐Hispanic White patients were also chronic opioid users (32.5%) compared to Black (21.6%) and Hispanic (15.5%) patients.
TABLE 1.
Patient demographics and characteristics.
| Number of patients (% total) | |||
|---|---|---|---|
| Overall (n = 5191) | Chronic opioid users (n = 1432) | Nonchronic opioid users (n = 3759) | |
| Level of amputation | |||
| Ankle and below ankle | 2828 (54.5%) | 638 (44.6%) | 2190 (58.3%) |
| Knee and below knee | 1228 (23.7%) | 424 (29.6%) | 804 (21.4%) |
| Hip and below hip | 1135 (21.9%) | 370 (25.8%) | 765 (20.4%) |
| Provider specialty | |||
| Vascular surgery | 1290 (24.9%) | 332 (23.2%) | 958 (25.5%) |
| General surgery | 1274 (24.5%) | 357 (24.9%) | 917 (24.4%) |
| Podiatry | 1259 (24.3%) | 275 (19.2%) | 984 (26.2%) |
| Orthopedic surgery | 1140 (22.0%) | 405 (28.3%) | 735 (19.6%) |
| Other surgical specialties | 101 (2.0%) | 24 (1.7%) | 77 (2.1%) |
| Plastic and reconstructive surgery | 65 (1.3%) | 19 (1.3%) | 46 (1.2%) |
| Medical specialties | 62 (1.2%) | 20 (1.4%) | 42 (1.1%) |
| Gender | |||
| Male | 3336 (64.3%) | 897 (62.6%) | 2439 (64.9%) |
| Female | 1855 (35.7%) | 535 (37.4%) | 1320 (35.1%) |
| Age group (years) | |||
| <55 | 1309 (25.2%) | 547 (38.2%) | 762 (20.3%) |
| 55–64 | 1147 (22.1%) | 375 (26.2%) | 772 (20.5%) |
| 65–74 | 1591 (30.7%) | 344 (24.0%) | 1247 (33.2%) |
| 75–84 | 772 (14.9%) | 121 (8.5%) | 651 (17.3%) |
| ≥85 | 372 (7.2%) | 45 (3.1%) | 327 (8.7%) |
| Race or ethnicity | |||
| Non‐Hispanic White | 3227 (62.2%) | 1048 (73.2%) | 2179 (58.0%) |
| Black | 1147 (22.1%) | 248 (17.3%) | 899 (23.9%) |
| Hispanic | 613 (11.8%) | 92 (6.4%) | 521 (13.9%) |
| Others & unknown | 204 (3.9%) | 44 (3.1%) | 160 (4.3%) |
| Current reason for Medicare entitlement | |||
| DIB | 2300 (44.3%) | 887 (61.9%) | 1413 (37.6%) |
| Advanced Age and Survivors Insurance | 2188 (42.2%) | 376 (26.3%) | 1812 (48.2%) |
| Both DIB and ESRD | 378 (7.3%) | 95 (6.6%) | 283 (7.5%) |
| ESRD | 170 (3.3%) | 37 (2.6%) | 133 (3.5%) |
| Unknown | 155 (3.0%) | 37 (2.6%) | 118 (3.1%) |
| CCI | |||
| None to mild (CCI 0, 1, 2) | 510 (9.8%) | 176 (12.3%) | 334 (8.9%) |
| Moderate (CCI 3, 4) | 875 (16.9%) | 242 (16.9%) | 633 (16.8%) |
| Severe (CCI ≥5) | 3806 (73.3%) | 1014 (70.8%) | 2792 (74.3%) |
| Filled perioperative opioid prescriptions | |||
| Did not fill | 1523 (29.3%) | 56 (3.9%) | 1467 (39.0%) |
| Filled within 30 d before admission | 2465 (47.5%) | 1301 (90.9%) | 1164 (31.0%) |
| Only filled within 3 d after discharge | 1203 (23.2%) | 75 (5.2%) | 1128 (30.0%) |
| Presence of mental disorders | |||
| Adjustment (yes) | 271 (5.2%) | 89 (6.2%) | 182 (4.8%) |
| Anxiety (yes) | 1185 (22.8%) | 532 (37.2%) | 653 (17.4%) |
| Disruptive (yes) | 85 (1.6%) | 33 (2.3%) | 52 (1.4%) |
| Mood (yes) | 1725 (33.2%) | 668 (46.7%) | 1057 (28.1%) |
| Personality (yes) | 52 (1.0%) | 20 (1.4%) | 32 (0.9%) |
| Schizophrenia (yes) | 371 (7.2%) | 96 (6.7%) | 275 (7.3%) |
| Suicide (yes) | 80 (1.5%) | 32 (2.2%) | 48 (1.3%) |
| Other mental (yes) | 247 (4.8%) | 89 (6.2%) | 158 (4.2%) |
| Drug and substance use (yes) | 959 (18.5%) | 411 (28.7%) | 548 (14.6%) |
| Presence of pain disorders | |||
| Arthritis (yes) | 5096 (98.2%) | 1425 (99.5%) | 3671 (97.7%) |
| Back (yes) | 1735 (33.4%) | 811 (56.6%) | 924 (24.6%) |
| Neck (yes) | 626 (12.1%) | 272 (19.0%) | 354 (9.4%) |
| Other pain disorder (yes) | 2355 (45.4%) | 759 (53.0%) | 1596 (42.5%) |
| Pre‐ and Post CDC guideline group | |||
| Pre‐CDC guideline (2009–2015) | 3607 (69.5%) | 1000 (69.8%) | 2607 (69.4%) |
| Post‐CDC guideline (2016–2018) | 1584 (30.5%) | 432 (30.2%) | 1152 (30.7%) |
| Filled prescriptions for benzodiazepines, sedatives, hypnotics, anxiolytics within 30 d before admission | 748 (14.4%) | 425 (29.7%) | 323 (8.6%) |
| Filled gabapentinoid (gabapentin and pregabalin) prescriptions within 30 d before admission | 2834 (54.6%) | 1038 (72.5%) | 1796 (47.8%) |
| Tobacco use | 2674 (51.5%) | 934 (65.2%) | 1740 (46.3%) |
| Persistent opioid users/prolonged opioid users | 2179 (42.0%) | 1306 (91.2%) | 873 (23.2%) |
Abbreviations: CCI, Charlson Comorbidity Index; CDC, Centers for Disease Control and Prevention; DIB, disability insurance benefits; ESRD, end‐stage renal disease.
Predictors of prolonged or new persistent opioid use
Among the 3759 nonchronic opioid users, 873 (23.2%) had new persistent opioid use after surgery (Table 1). In these nonchronic users, younger age < 55 years had greater odds of new persistent opioid use compared to those aged 65–74 years (odds ratio [OR], 1.41 [95% CI, 1.01–1.97]). Compared to those individuals who did not fill their opioid prescriptions, nonchronic users who filled their prescription during the perioperative surgical period (OR, 4.46 [95% CI, 3.57–5.57]) were more likely to have new persistent opioid use after surgery (Table 2). Additionally, those with back pain were more likely to have new persistent opioid use after surgery (OR, 1.26 [95% CI, 1.03–1.53]). Nonchronic users with concurrent filling of gabapentinoid prescriptions (OR, 1.92 [95% CI, 1.60–2.30]) and filling of prescriptions for benzodiazepines, sedatives, hypnotics, and anxiolytics (OR, 1.75 [95% CI, 1.33–2.29]) were associated with greater odds of new persistent opioid use.
TABLE 2.
Adjusted odds ratios of prolong opioid use (for chronic opioid users) and new persistent opioid use (for nonchronic opioid users) by patient characteristics.
| Chronic opioid users | Nonchronic opioid users | |||
|---|---|---|---|---|
| Odds ratio (95% CI) | p value | Odds ratio (95% CI) | p value | |
| Level of amputation (Ref: amputation at ankle and below ankle amputation) | ||||
| Hip and below hip | 1.15 (0.69–1.92) | .592 | 1.02 (0.79–1.30) | .895 |
| Knee and below knee | 1.00 (0.60–1.66) | .991 | 0.95 (0.75–1.21) | .698 |
| Provider's specialty (Ref: general surgery) | ||||
| Medical specialties | a | 1.01 (0.43–2.36) | .981 | |
| Orthopedic surgery | 1.55 (0.90–2.67) | .118 | 0.90 (0.69–1.17) | .426 |
| Other surgical specialties | 0.55 (0.18–1.65) | .286 | 0.75 (0.41–1.37) | .343 |
| Plastic and reconstructive surgery | 2.01 (0.25–16.15) | .510 | 1.34 (0.63–2.83) | .448 |
| Podiatry | 2.63 (1.27–5.44) | .009 | 0.88 (0.67–1.16) | .364 |
| Vascular surgery | 0.87 (0.53–1.42) | .570 | 0.81 (0.64–1.03) | .081 |
| Female (Ref: male) | 0.98 (0.65–1.48) | .924 | 0.92 (0.76–1.11) | .361 |
| Age group (years; Ref: 65–74) | ||||
| <55 | 1.09 (0.50–2.37) | .833 | 1.41 (1.01–1.97) | .042 |
| 55–64 | 0.95 (0.44–2.09) | .904 | 1.38 (0.99–1.91) | .055 |
| 75–84 | 0.89 (0.45–1.75) | .728 | 0.77 (0.58–1.03) | .081 |
| >=85 | 0.82 (0.30–2.20) | .689 | 0.95 (0.66–1.38) | .801 |
| Race or ethnicity (Ref: Non‐Hispanic White) | ||||
| Black | 0.90 (0.54–1.51) | .701 | 1.18 (0.96–1.46) | .119 |
| Hispanic | 1.12 (0.50–2.51) | .790 | 0.68 (0.52–0.90) | .008 |
| Other and unknown | 0.91 (0.30–2.77) | .874 | 0.65 (0.41–1.03) | .067 |
| Current reason for Medicare entitlement (Ref: Advanced Age and Survivors Insurance) | ||||
| DIB | 1.81 (0.81–4.03) | .147 | 0.79 (0.56–1.11) | .170 |
| Both DIB and ESRD | 1.06 (0.36–3.14) | .911 | 0.81 (0.52–1.27) | .356 |
| ESRD | 5.61 (0.66–47.58) | .114 | 1.24 (0.75–2.07) | .396 |
| Unknown | 0.67 (0.20–2.24) | .518 | 1.04 (0.59–1.86) | .881 |
| CCI (Ref: none to mild (CCI 0, 1, 2)) | ||||
| Moderate (CCI 3, 4) | 1.12 (0.53–2.38) | .763 | 0.98 (0.68–1.41) | .912 |
| Severe CCI >= 5 | 1.12 (0.59–2.14) | .725 | 1.09 (0.79–1.51) | .591 |
| Filled perioperative opioid prescriptions (Ref: didn't fill) | ||||
| Filled within 30 d before admission | 2.21 (1.06–4.60) | .033 | 4.46 (3.57–5.57) | <.001 |
| Only filled within 3 d after discharge | 1.16 (0.45–3.03) | .755 | 1.54 (1.21–1.96) | <.001 |
| Mental disorders | ||||
| Adjustment (Ref: no) | 0.91 (0.44–1.89) | .802 | 1.52 (1.04–2.20) | .029 |
| Anxiety (Ref: no) | 1.17 (0.74–1.85) | .512 | 1.07 (0.85–1.35) | .574 |
| Disruptive (Ref: no) | 0.66 (0.21–2.09) | .483 | 0.91 (0.44–1.88) | .805 |
| Mood (Ref: no) | 0.84 (0.54–1.29) | .422 | 1.11 (0.90–1.37) | .318 |
| Personality (Ref: no) | 2.77 (0.31–24.76) | .361 | 0.57 (0.23–1.42) | .225 |
| Schizophrenia (Ref: no) | 0.52 (0.27–0.98) | .042 | 1.07 (0.78–1.48) | .674 |
| Suicide (Ref: no) | 0.41 (0.13–1.26) | .120 | 1.53 (0.75–3.10) | .241 |
| Other mental (Ref: no) | 0.77 (0.35–1.73) | .531 | 1.02 (0.69–1.52) | .912 |
| Drug and substance use (Ref: no) | 0.97 (0.59–1.61) | .912 | 1.23 (0.94–1.61) | .128 |
| Pain disorders | ||||
| Arthritis (Ref: no) | 1.25 (0.14–11.45) | .845 | 1.26 (0.62–2.54) | .524 |
| Back (Ref: no) | 1.36 (0.90–2.04) | .140 | 1.26 (1.03–1.53) | .022 |
| Neck (Ref: no) | 1.09 (0.63–1.87) | .762 | 1.15 (0.87–1.52) | .318 |
| Other pain disorder (Ref: no) | 0.85 (0.56–1.27) | .417 | 1.15 (0.96–1.37) | .134 |
| Pre‐CDC guideline (Ref: post‐CDC guideline) | 1.09 (0.67–1.79) | .720 | 1.51 (1.21–1.88) | <.001 |
| Filled gabapentinoid (gabapentin and pregabalin) prescriptions within 30 d before admission (Ref: didn't fill) | 1.09 (0.72–1.67) | .679 | 1.92 (1.60–2.30) | <.001 |
| Filled prescriptions for benzodiazepines, sedatives, hypnotics, anxiolytics within 30 d before admission (Ref: didn't fill) | 1.21 (0.75–1.93) | .434 | 1.75 (1.33–2.29) | <.001 |
| Intermittent opioid users (Ref: opioid naïve) | 2.60 (2.10–3.22) | <.001 | ||
| Tobacco use (Ref: no) | 0.94 (0.60–1.46) | .771 | 1.10 (0.91–1.33) | .337 |
Due to zero case identified as prolong opioid use, the model was not able to estimate odds ratio.
Abbreviations: CCI, Charlson Comorbidity Index; CDC, Centers for Disease Control and Prevention; CI, confidence interval; DIB, disability insurance benefits; ESRD, end‐stage renal disease.
Of the 1432 chronic opioid users, 1306 (91.2%) demonstrated prolonged opioid use after surgery (Table 1). In these patients, individuals filling their opioid prescription within the perioperative surgical period had greater odds (OR, 2.21 [95% CI, 1.06–4.60]) of prolonged opioid use compared to those not filling during the perioperative surgical period (Table 2). Concurrent filling of gabapentinoid prescriptions (OR, 1.09 [95% CI, 0.72–1.67]) and filling of prescriptions for benzodiazepines, sedatives, hypnotics, and anxiolytics (OR, 1.21 [95% CI, 0.75–1.93]) was not associated an increased rate of prolonged opioid use.
Rates of high‐risk opioid prescribing
Rates of high‐risk prescribing in nonchronic and chronic users are shown in Figure 1. Yearly rates of high‐risk prescribing in nonchronic users were analyzed for both opioid‐naïve and intermittent users. In opioid‐naïve users, the rates ranged from 11.9% in 2009 to 14.0% in 2018 (p = .736). In intermittent users, the rates ranged from 41.6% in 2009 to 37.1% in 2018 (p = .105). The yearly rates of high‐risk prescribing did not change significantly during the study period. The rates in chronic users ranged from 88.6% in 2009 to 75.9% in 2018 (p = .650). Overall, 15.5% of opioid‐naïve, 37.2% of intermittent, and 82.3% of chronic users were exposed to high‐risk prescribing (p < .001) (Supporting Information V‐Table V).
FIGURE 1.
Rates of high‐risk prescribing in chronic and nonchronic opioid users.
There were statistical differences in the rates of high‐risk prescribing before and after the implementation of the CDC guideline (44.3% vs. 40.5%; p = .011). When compared to the rates of high‐risk prescribing for non‐chronic users before the implementation of the CDC guideline (29.5%), the rate was lower after the implementation (25.2%; p = .007) although there were no differences among opioid naïve patients (16.5% vs. 13.3%; p = .106) and intermittent users (38.5% vs. 34.2%; p = .052). The rates of high‐risk prescribing among chronic users were not significant between before and after the implementation (82.8% vs. 81.2%; p = .480).
Opioid prescription sizes
Among nonchronic users, mean opioid prescription size was 767 OMEs with median 435 OMEs and interquartile range (IQR) 225–875 OMEs. Concurrent filling of gabapentinoid prescriptions (multiplicative effect = 1.35, p < .001) and filling of prescriptions for benzodiazepines, sedatives, hypnotics, and anxiolytics (multiplicative effect = 1.16, p = .014) was associated with higher opioid prescriptions sizes.
In chronic opioid users, mean prescription size was 4353 OMEs with median 2205 OMEs and QR 1072.5–4788 OMEs. Amputations at the hip and below the hip were associated with higher opioid prescription sizes compared to ankle amputations (multiplicative effect = 1.26, p = .003) (Table 3). Furthermore, younger age < 55 years (multiplicative effect = 1.77, p < .001) and 55–64 years (multiplicative effect = 1.43, p < .001) was also associated with higher opioid prescription sizes compared to patients aged 65–74 years.
TABLE 3.
Multiplicative effect of patient characteristics on the amount of opioid prescriptions (in OMEs) filled during 3 months after surgery for chronic and nonchronic opioid users.
| Chronic opioid users | Nonchronic opioid users | |||
|---|---|---|---|---|
| Multiplicative effect (95% CI) | p value | Multiplicative effect (95% CI) | p value | |
| Level of amputation (Ref: amputation at ankle and below ankle amputation) | ||||
| Hip and below hip | 1.26 (1.08–1.47) | .003 | 1.22 (1.10–1.35) | <.001 |
| Knee and below knee | 1.05 (0.90–1.22) | .523 | 1.23 (1.11–1.35) | <.001 |
| Provider's specialty (Ref: general surgery) | ||||
| Medical specialties | 1.21 (0.76–1.92) | .428 | 1.01 (0.70–1.44) | .969 |
| Orthopedic surgery | 1.09 (0.93–1.27) | .283 | 1.17 (1.05–1.30) | .005 |
| Other surgical specialties | 0.89 (0.57–1.37) | .586 | 1.34 (1.04–1.73) | .022 |
| Plastic and reconstructive surgery | 1.03 (0.64–1.66) | .909 | 1.00 (0.74–1.35) | .993 |
| Podiatry | 1.09 (0.91–1.31) | .368 | 0.91 (0.81–1.02) | .099 |
| Vascular surgery | 1.02 (0.87–1.19) | .841 | 1.11 (1.01–1.23) | .029 |
| Female (Ref: male) | 0.90 (0.80–1.01) | .065 | 0.99 (0.92–1.07) | .839 |
| Age group (years; Ref: 65–74) | ||||
| <55 | 1.77 (1.45–2.16) | <.001 | 1.44 (1.26–1.64) | <.001 |
| 55–64 | 1.43 (1.16–1.75) | <.001 | 1.16 (1.02–1.32) | .027 |
| 75–84 | 0.75 (0.59–0.95) | .017 | 0.91 (0.81–1.02) | .116 |
| >=85 | 0.74 (0.53–1.04) | .080 | 0.93 (0.79–1.08) | .323 |
| Race or ethnicity (Ref: Non‐Hispanic White) | ||||
| Black | 0.87 (0.75–1.01) | .074 | 0.90 (0.82–0.98) | .014 |
| Hispanic | 0.76 (0.60–0.95) | .018 | 0.86 (0.77–0.95) | .005 |
| Other and unknown | 0.73 (0.54–1.01) | .055 | 0.81 (0.67–0.98) | .029 |
| Current reason for Medicare entitlement (ref: Advanced Age and Survivors Insurance) | ||||
| DIB | 0.94 (0.75–1.18) | .612 | 1.08 (0.95–1.24) | .235 |
| Both DIB and ESRD | 0.61 (0.45–0.84) | .002 | 0.83 (0.69–1.00) | .046 |
| ESRD | 0.65 (0.43–0.96) | .030 | 1.08 (0.87–1.34) | .512 |
| Unknown | 0.60 (0.40–0.89) | .012 | 1.10 (0.88–1.37) | .404 |
| CCI (Ref: none to mild (CCI 0, 1, 2)) | ||||
| Moderate (CCI 3, 4) | 0.94 (0.76–1.15) | .523 | 0.97 (0.84–1.11) | .625 |
| Severe CCI >= 5 | 0.92 (0.77–1.10) | .377 | 0.93 (0.83–1.06) | .276 |
| Filled perioperative opioid prescriptions (Ref: didn't fill) | ||||
| Filled within 30 d before admission | 1.32 (1.00–1.76) | .054 | 1.35 (1.21–1.51) | <.001 |
| Only filled within 3 d after discharge | 0.92 (0.64–1.32) | .634 | 0.94 (0.85–1.05) | .297 |
| Mental disorders | ||||
| Adjustment (Ref: no) | 1.12 (0.89–1.42) | .329 | 1.04 (0.88–1.22) | .659 |
| Anxiety (Ref: no) | 1.07 (0.94–1.22) | .304 | 1.07 (0.97–1.18) | .199 |
| Disruptive (Ref: no) | 1.49 (1.04–2.15) | .030 | 0.92 (0.68–1.24) | .581 |
| Mood (Ref: no) | 1.03 (0.91–1.17) | .605 | 0.96 (0.88–1.04) | .342 |
| Personality (Ref: no) | 1.24 (0.76–2.02) | .383 | 1.15 (0.77–1.72) | .502 |
| Schizophrenia (Ref: no) | 0.93 (0.75–1.17) | .552 | 1.00 (0.87–1.14) | .978 |
| Suicide (Ref: no) | 0.77 (0.52–1.14) | .198 | 1.59 (1.17–2.17) | .003 |
| Other mental (Ref: no) | 1.02 (0.81–1.28) | .888 | 1.10 (0.93–1.30) | .272 |
| Drug and substance use (Ref: no) | 1.17 (1.02–1.35) | .029 | 1.06 (0.95–1.18) | .297 |
| Pain disorders | ||||
| Arthritis (Ref: no) | 1.42 (0.66–3.09) | .372 | 1.09 (0.85–1.39) | .509 |
| Back (Ref: no) | 1.25 (1.11–1.40) | <.001 | 1.08 (0.99–1.17) | .074 |
| Neck (Ref: no) | 1.01 (0.88–1.17) | .852 | 0.95 (0.85–1.07) | .442 |
| Other pain disorder (Ref: no) | 1.05 (0.93–1.17) | .438 | 1.02 (0.94–1.10) | .651 |
| Pre‐CDC guideline (Ref: Post‐CDC guideline) | 1.27 (1.10–1.45) | .001 | 1.34 (1.23–1.46) | <.001 |
| Filled gabapentinoid (gabapentin and pregabalin) prescriptions within 30 d before admission (Ref: didn't fill) | 1.04 (0.92–1.17) | .565 | 1.35 (1.25–1.45) | <.001 |
| Filled prescriptions for benzodiazepines, sedatives, hypnotics, anxiolytic within 30 d before admission (Ref: didn't fill) | 1.06 (0.94–1.20) | .332 | 1.16 (1.03–1.30) | .014 |
| Intermittent opioid users (Ref: opioid naïve) | 1.23 (1.13–1.34) | <.001 | ||
| Tobacco use (Ref: no) | 0.99 (0.88, 1.12) | .883 | 1.03 (0.96, 1.12) | .385 |
Note: In this model, the outcome variable is the natural log transformed opioid prescription size (in OMEs), and multiplicative effect was the exponentiated regression coefficient.
Abbreviations: CCI, Charlson Comorbidity Index; CDC, Centers for Disease Control and Prevention; CI, confidence interval; DIB, disability insurance benefits; ESRD, end‐stage renal disease; OME, oral morphine milligram equivalent.
DISCUSSION
In this study of Medicare beneficiaries undergoing lower limb amputation, nearly a quarter (23.2%) of nonchronic opioid users developed new persistent opioid use and over 90% of chronic users exhibited prolonged opioid use after surgery. Predictors of new persistent opioid use and prolonged opioid use included prior opioid fills, presence of pain disorders, younger age, concurrent prescription fills of gabapentinoids, and all other pain adjuncts. Rates of high‐risk prescribing did not significantly change during the study period for both nonchronic and opioid users.
New persistent opioid use and prolonged opioid use
Among nonchronic users, our findings for rates of new persistent use were lower than those described by Park et al. in their review of patients undergoing lower extremity amputation between 2010 and 2018 using national insurance claims data. 29 The database used in their study (PearlDiver Patients Records Database) covers all U.S. states with all payer types including commercial, Medicare, Medicaid, government claims, cash, and other sources such as the National Inpatient Sample. 30 The authors looked at 2247 opioid‐naïve patients undergoing lower extremity amputation and found that 44.6% of the patients had new persistent opioid use after surgery (defined in their study as a filled opioid prescription between 90 and 180 days following surgery). The higher rate of new persistent opioid use by Park et al. compared to our study could largely be due to variations in the populations sampled by the queried claims databases. With multiple payors captured in the PearlDrive database compared to the Medicare database, it is possible that the actual rate of new persistent use is higher than the rate captured by our study (23.2%). However, the study by Park et al. did not exclude patients if they did not discharge home and because the PearlDrive database was queried for multiple health care settings including inpatient and outpatient scenarios, it is possible that new persistent opioid use was overcaptured in that study compared to ours. Nevertheless, both studies found relatively high rates of opioid use following lower extremity amputation in previously nonchronic use patients.
The high rate of prolonged opioid use among chronic opioid users undergoing amputation (90%) highlights a need for better pain management strategies in these patients. Similar results to a lesser degree have been reported with bariatric surgery, where 77% of preoperative chronic opioid users continued chronic opioid use years after surgery. 31 Interestingly, in that study the amount of chronic opioid use among chronic opioid users undergoing surgery was greater postoperatively than preoperatively. Although we did not specifically study the quantity of opioid use before and after surgery, it is concerning that rates of high‐risk opioid prescribing remained persistently high among chronic users. Of the 665 chronic opioid users who remained exposed to high‐risk prescribing, 50% had high daily‐doses>100 OMEs and 71% had new long‐acting opioid use within 90 days postoperatively. Given the possible potentiating effects of postsurgical pain in worsening opioid use among chronic users, and the adverse outcomes associated with high‐risk use, it is important to implement opioid sparing perioperative interventions to mitigate opioid use after surgery and encourage multimodal analgesia to ensure adequate pain relief. 32
Opioid, gabapentinoid, and benzodiazepene prescriptions
Our study also found that opioid fills during the perioperative surgical period were associated with a significantly increased risk of new persistent opioid use (OR, 4.46) and prolonged opioid use (OR, 2.21). A prior study by Katzman et al., found an association between preoperative opioid exposure and prolonged postoperative use. 33 In that study involving over 190,000 patients, patients with recent exposure to opioids (compared to opioid‐naïve patients) had an increased risk of prolonged fills after surgery. 33 In addition to thorough preoperative screening of patients for opioid use, it could be prudent for providers to prescribe opioids after surgery as opposed to preemptively before surgery for patients undergoing amputation, especially given that in our study almost half of the patients received perioperative opioid prescriptions.
Opioids and gabapentinoids are frequently coprescribed to patients undergoing amputation to help manage postoperative pain. However, the concurrent use of gabapentinoids and opioids as well as the combination of gabapentinoids, opioids, and benzodiazepines are associated with adverse events including respiratory depression and substance‐related overdose. 34 Gabapentinoids are being increasingly prescribed to decrease the need for opioid use and provide an alternative analgesic. However, other studies have also found that preoperative gabapentinoid use is associated with postoperative new persistent use of gabapentinoids and opioids, and may be a marker of patients at high risk of persistent use. 35 Even the coprescription of opioids with muscle relaxants, a common combination after amputations, has been associated with increased risks for opioid refills after controlling for preoperative opioid exposure. 36 These associations make it important for surgical teams caring for patients undergoing amputation to be mindful of the risks associated with commonly used medication prescriptions after surgery. It is also important for health care teams to better identify patients who are at risk for new persistent opioid use or prolonged opioid use after amputation. This includes various interventions in the preoperative, perioperative, and postoperative periods. 37 , 38 , 39 , 40 , 41 , 42
CDC guidelines and high‐risk opioid prescribing
Compared to post‐CDC guidelines, pre‐CDC guidelines were associated with greater persistent opioid use after surgery in chronic (OR, 1.09 [95% CI, 0.67–1.79) and nonchronic (OR, 1.51 [95% CI, 1.21–1.88) users. Following CDC recommendations for opioid prescribing in 2016, chronic users had 27% smaller opioid prescription sizes and nonchronic users had 34% smaller prescriptions sizes (multiplicative effect = 1.27, p < .001 and multiplicative effect = 1.34, p < .001, respectively) (Table 3). Although the Cochran–Armitage trend test shows rates of high‐risk opioid prescribing did not change significantly over the study period (2009–2018) and rates of high‐risk opioid prescribing in chronic users remained high following guideline implementation (75.9%) (Supporting Information II‐Table II), the chi‐square analysis demonstrated statistically significant lower rates of high‐risk prescribing in nonchronic users (p = .007) but not chronic users (p = .480) before and after implementation. These findings are similar to what has been reported in the literature, which suggest that implemented opioid health care policies overall have improved prescription patterns but benefited high‐risk patients the least. 43 Future policy and pain management strategies for the lower extremity amputation population should be targeted toward patients with high‐risk prescription patterns.
LIMITATIONS
First, the data captured in the study reference opioid prescription fills and do not explicitly capture opioid use. Nevertheless, prescription fills provide a reasonable proxy for opioid use. 44 Second, this study was limited to Medicare beneficiaries, and most patients in this study were male and 65 years of age or older. As such, the findings from this cohort may not be generalizable to other groups. Third, due to the limitations in querying the database, the indication for opioid prescription cannot be determined. Therefore, it is difficult to know if opioid prescriptions were provided specifically for postoperative pain or for other pain conditions as the timing of the prescription was linked to the timing of the procedure. Although we attempted to control for this by excluding patients undergoing multiple surgeries and patients not discharged home, future studies such as a prospective clinical trial examining postoperative opioid use after major lower limb amputation could be designed to better control for such limitations. Lastly, as with all large database studies, there are limited clinically relevant variables such as surgical indication for amputation, which may limit generalizability.
CONCLUSION
This study characterized opioid use in Medicare beneficiaries undergoing lower extremity amputations and found high rates of prolonged opioid use and new persistent opioid use after surgery in this population. Furthermore, it found that high‐risk prescribing practices remain common among nonchronic and chronic opioid users and that this has not significantly changed over time despite dissemination of CDC prescribing guidelines. Because the limb amputation population is at risk for chronic pain after surgery, interventions that focus on tailoring postoperative pain management based on preoperative risk factors will be critical to ensure optimal pain management and safe opioid stewardship.
DISCLOSURES
There are no disclaimers and no conflicts of interest to declare for the authors of this publication.
Supporting information
Figure S1. Terminology.
Figure S2. Flow Chart.
Data S1. Supporting Information.
ACKNOWLEDGMENTS
This research received no specific grant from any funding agency in the public, commercial or not‐for‐profit sectors. Jennifer F. Waljee receives funding from the Michigan Department of Health & Human Services; E20221872‐00/AQ.
Khouri AN, Kumar NG, Hu H‐M, Yazdanfar M, Waljee JF, Kung TA. Opioid consumption after lower limb amputation in Medicare beneficiaries. PM&R. 2025;17(12):1452‐1463. doi: 10.1002/pmrj.13408
The findings of this manuscript were selected for a Poster Presentation at the American Association of Plastic Surgery 2022 Meeting in San Diego, CA from April 9–12, 2022.
REFERENCES
- 1. Hsu E, Cohen SP. Postamputation pain: epidemiology, mechanisms, and treatment. J Pain Res. 2013;6:121‐136. doi: 10.2147/JPR.S32299 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Ziegler‐Graham K, MacKenzie EJ, Ephraim PL, Travison TG, Brookmeyer R. Estimating the prevalence of limb loss in the United States: 2005 to 2050. Arch Phys Med Rehabil. 2008;89(3):422‐429. doi: 10.1016/j.apmr.2007.11.005 [DOI] [PubMed] [Google Scholar]
- 3. Horgan O, MacLachlan M. Psychosocial adjustment to lower‐limb amputation: a review. Disabil Rehabil. 2004;26(14–15):837‐850. doi: 10.1080/09638280410001708869 [DOI] [PubMed] [Google Scholar]
- 4. Haythornthwaite JA, Benrud‐Larson LM. Psychological aspects of neuropathic pain. Clin J Pain. 2000;16(2 Suppl):S101‐S105. doi: 10.1097/00002508-200006001-00017 [DOI] [PubMed] [Google Scholar]
- 5. Suckow BD, Goodney PP, Nolan BW, et al. Domains that determine quality of life in vascular amputees. Ann Vasc Surg. 2015;29(4):722‐730. doi: 10.1016/j.avsg.2014.12.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Luo Y, Anderson TA. Phantom limb pain: a review. Int Anesthesiol Clin. 2016;54(2):121‐139. doi: 10.1097/AIA.0000000000000095 [DOI] [PubMed] [Google Scholar]
- 7. Srivastava D. Chronic post‐amputation pain: peri‐operative management ‐ review. Br J Pain. 2017;11(4):192‐202. doi: 10.1177/2049463717736492 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Kooijman CM, Dijkstra PU, Geertzen JH, Elzinga A, van der Schans CP. Phantom pain and phantom sensations in upper limb amputees: an epidemiological study. Pain. 2000;87(1):33‐41. doi: 10.1016/s0304-3959(00)00264-5 [DOI] [PubMed] [Google Scholar]
- 9. Ephraim PL, Wegener ST, MacKenzie EJ, Dillingham TR, Pezzin LE. Phantom pain, residual limb pain, and back pain in amputees: results of a national survey. Arch Phys Med Rehabil. 2005;86(10):1910‐1919. doi: 10.1016/j.apmr.2005.03.031 [DOI] [PubMed] [Google Scholar]
- 10. Sherman RA, Sherman CJ. Prevalence and characteristics of chronic phantom limb pain among American veterans. Results of a trial survey. Am J Phys Med. 1983;62(5):227‐238. https://www.ncbi.nlm.nih.gov/pubmed/6624883 [PubMed] [Google Scholar]
- 11. Modest JM, Raducha JE, Testa EJ, Eberson CP. Management of post‐amputation pain. RI Med J. 2020;103(4):19‐22. [PubMed] [Google Scholar]
- 12. Ahuja V, Thapa D, Ghai B. Strategies for prevention of lower limb post‐amputation pain: a clinical narrative review. J Anaesthesiol Clin Pharmacol. 2018;34(4):439‐449. doi: 10.4103/joacp.JOACP_126_17 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Kent ML, Hsia HJ, Van de Ven TJ, Buchheit TE. Perioperative pain management strategies for amputation: a topical review. Pain Med. 2017;18(3):504‐519. doi: 10.1093/pm/pnw110 [DOI] [PubMed] [Google Scholar]
- 14. Chou R, Turner JA, Devine EB, et al. The effectiveness and risks of long‐term opioid therapy for chronic pain: a systematic review for a National Institutes of Health Pathways to Prevention workshop. Ann Intern Med. 2015;162(4):276‐286. doi: 10.7326/M14-2559 [DOI] [PubMed] [Google Scholar]
- 15. Garland EL. Treating chronic pain: the need for non‐opioid options. Expert Rev Clin Pharmacol. 2014;7(5):545‐550. doi: 10.1586/17512433.2014.928587 [DOI] [PubMed] [Google Scholar]
- 16. Vowles KE, McEntee ML, Julnes PS, Frohe T, Ney JP, van der Goes DN. Rates of opioid misuse, abuse, and addiction in chronic pain: a systematic review and data synthesis. Pain. 2015;156(4):569‐576. doi: 10.1097/01.j.pain.0000460357.01998.f1 [DOI] [PubMed] [Google Scholar]
- 17. Prevention CfDCa . CDC Clinical Practice Guideline for Prescribing Opioids for Pain — United States, 2022. https://www.cdc.gov/mmwr/volumes/71/rr/rr7103a1.htm. [DOI] [PMC free article] [PubMed]
- 18. Prevention CfDCa . CDC Guideline For Prescribing Opioids For Chronic Pain. https://www.cdc.gov/drugoverdose/pdf/Guidelines_At-A-Glance-508.pdf.
- 19. Nadeau SE, Wu JK, Lawhern RA. Opioids and chronic pain: an analytic review of the clinical evidence. Front Pain Res (Lausanne). 2021;2:721357. doi: 10.3389/fpain.2021.721357 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Gil JA, Gunaseelan V, DeFroda SF, Brummett CM, Bedi A, Waljee JF. Risk of prolonged opioid use among opioid‐naive patients after common shoulder arthroscopy procedures. Am J Sports Med. 2019;47(5):1043‐1050. doi: 10.1177/0363546518819780 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Marcusa DP, Mann RA, Cron DC, et al. Prescription opioid use among opioid‐naive women undergoing immediate breast reconstruction. Plast Reconstr Surg. 2017;140(6):1081‐1090. doi: 10.1097/PRS.0000000000003832 [DOI] [PubMed] [Google Scholar]
- 22. Lapidus JB, Santosa KB, Skolnick GB, et al. Opioid prescribing and use patterns in postsurgical facial trauma patients. Plast Reconstr Surg. 2020;145(3):780‐789. doi: 10.1097/PRS.0000000000006588 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Brescia AA, Harrington CA, Mazurek AA, et al. Factors associated with new persistent opioid usage after lung resection. Ann Thorac Surg. 2019;107(2):363‐368. doi: 10.1016/j.athoracsur.2018.08.057 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Harbaugh CM, Lee JS, Hu HM, et al. Persistent opioid use among pediatric patients after surgery. Pediatrics. 2018;141(1):e20172439. doi: 10.1542/peds.2017-2439 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Peahl AF, Morgan DM, Dalton VK, et al. New persistent opioid use after acute opioid prescribing in pregnancy: a nationwide analysis. Am J Obstet Gynecol. 2020;223:566.e1‐566.e13. doi: 10.1016/j.ajog.2020.03.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Molina CS, Faulk J. Lower Extremity Amputation. StatPearls; 2024. [PubMed] [Google Scholar]
- 27. Delaney LD, Gunaseelan V, Rieck H, Dupree JM, Hallstrom BR, Waljee JF. High‐risk prescribing increases rates of new persistent opioid use in total hip arthroplasty patients. J Arthroplasty. 2020;35(9):2472‐2479. doi: 10.1016/j.arth.2020.04.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Vargas GM, Gunaseelan V, Upp L, et al. High‐risk opioid prescribing associated with postoperative new persistent opioid use in adolescents and young adults. Ann Surg. 2023;277(5):761‐766. doi: 10.1097/SLA.0000000000005193 [DOI] [PubMed] [Google Scholar]
- 29. Park RH, Liston JM, Samuel AR, Forster GL, DeGeorge BR. Risk factors for prolonged opioid consumption in lower extremity amputees. Plast Reconstr Surg Glob Open. 2022;10(2):4026. doi: 10.1097/GOX.0000000000004026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Bolognesi MP, Habermann EB. Commercial claims data sources: PearlDiver and individual payer databases. J Bone Joint Surg Am. 2022;104(Suppl 3):15‐17. doi: 10.2106/JBJS.22.00607 [DOI] [PubMed] [Google Scholar]
- 31. Raebel MA, Newcomer SR, Reifler LM, et al. Chronic use of opioid medications before and after bariatric surgery. Jama. 2013;310(13):1369‐1376. doi: 10.1001/jama.2013.278344 [DOI] [PubMed] [Google Scholar]
- 32. Hah JM, Bateman BT, Ratliff J, Curtin C, Sun E. Chronic opioid use after surgery: implications for perioperative management in the face of the opioid epidemic. Anesth Analg. 2017;125(5):1733‐1740. doi: 10.1213/ANE.0000000000002458 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Katzman C, Harker EC, Ahmed R, et al. The association between preoperative opioid exposure and prolonged postoperative use. Ann Surg. 2021;274(5):e410‐e416. doi: 10.1097/SLA.0000000000003723 [DOI] [PubMed] [Google Scholar]
- 34. Olopoenia A, Camelo‐Castillo W, Qato DM, et al. Adverse outcomes associated with concurrent gabapentin, opioid, and benzodiazepine utilization: a nested case‐control study. Lancet reg Health Am. 2022;13:100302. doi: 10.1016/j.lana.2022.100302 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Billig JI, Sears ED, Gunaseelan V, et al. Inappropriate preoperative gabapentinoid use among patients with carpal tunnel syndrome. J Hand Surg Am. 2020;45(8):677‐689.e5. (In eng). doi: 10.1016/j.jhsa.2020.04.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Santosa KB, Wang CS, Hu HM, et al. Opioid coprescribing with sedatives after implant‐based breast reconstruction. Plast Reconstr Surg. 2022;150(6):1224e‐1235e. doi: 10.1097/PRS.0000000000009726 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Memtsoudis SG, Poeran J, Zubizarreta N, et al. Association of Multimodal Pain Management Strategies with perioperative outcomes and resource utilization: a population‐based study. Anesthesiology. 2018;128(5):891‐902. doi: 10.1097/ALN.0000000000002132 [DOI] [PubMed] [Google Scholar]
- 38. Burns S, Urman R, Pian R, Coppes OJM. Reducing new persistent opioid use after surgery: a review of interventions. Curr Pain Headache Rep. 2021;25(5):27. doi: 10.1007/s11916-021-00943-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Kent ML, Hurley RW, Oderda GM, et al. American Society for Enhanced Recovery and Perioperative Quality Initiative‐4 joint consensus Statement on persistent postoperative opioid use: definition, incidence, risk factors, and health care system initiatives. Anesth Analg. 2019;129(2):543‐552. doi: 10.1213/ANE.0000000000003941 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Howard R, Waljee J, Brummett C, Englesbe M, Lee J. Reduction in opioid prescribing through evidence‐based prescribing guidelines. JAMA Surg. 2018;153(3):285‐287. doi: 10.1001/jamasurg.2017.4436 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Zhang DDQ, Sussman J, Dossa F, et al. A systematic review of behavioral interventions to decrease opioid prescribing after surgery. Ann Surg. 2020;271(2):266‐278. doi: 10.1097/SLA.0000000000003483 [DOI] [PubMed] [Google Scholar]
- 42. Kumar K, Kirksey MA, Duong S, Wu CL. A review of opioid‐sparing modalities in perioperative pain management: methods to decrease opioid use postoperatively. Anesth Analg. 2017;125(5):1749‐1760. doi: 10.1213/ANE.0000000000002497 [DOI] [PubMed] [Google Scholar]
- 43. Scherrer JF, Tucker J, Salas J, Zhang Z, Grucza R. Comparison of opioids prescribed for patients at risk for opioid misuse before and after publication of the Centers for Disease Control and Prevention's opioid prescribing guidelines. JAMA Netw Open. 2020;3(12):e2027481. doi: 10.1001/jamanetworkopen.2020.27481 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Santosa KB, Lai YL, Oliver JD, et al. Preoperative opioid use and mortality after minor outpatient surgery. JAMA Surg. 2020;155(12):1169‐1171. doi: 10.1001/jamasurg.2020.3623 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Terminology.
Figure S2. Flow Chart.
Data S1. Supporting Information.