Sage Journals: Discover world-class research

Abstract

Introduction

Limited evidence exists on health system characteristics associated with initial and long-term prescribing of opioids and nonsteroidal anti-inflammatory drugs (NSAIDs) following total hip and knee arthroplasty (THA/TKA), and if these characteristics differ among individuals based on preoperative NSAID exposure. We identified orthopedic surgeon opioid prescribing practices, hospital characteristics, and regional factors associated with initial and long-term prescribing of opioids and NSAIDs among older adults receiving THA/TKA.

Materials and Methods

This observational study included opioid-naïve Medicare beneficiaries aged ≥65 years receiving elective THA/TKA between January 1, 2014 and July 4, 2017. We examined initial (days 1-30 following THA/TKA) and long-term (days 90-180) opioid or NSAID prescribing, stratified by preoperative NSAID exposure. Risk ratios (RRs) for the associations between 10 health system characteristics and case-mix adjusted outcomes were estimated using multivariable Poisson regression models.

Results

The study population included 23,351 NSAID-naïve and 10,127 NSAID-prevalent individuals. Increases in standardized measures of orthopedic surgeon opioid prescribing generally decreased the risk of initial NSAID prescribing but increased the risk of long-term opioid prescribing. For example, among NSAID-naïve individuals, the RRs (95% confidence intervals [CIs]) for initial NSAID prescribing were 0.95 (0.93-0.97) for 1-2 orthopedic surgeon opioid prescriptions per THA/TKA procedure, 0.94 (0.92-0.97) for 3-4 prescriptions per procedure, and 0.91 (0.89-0.93) for 5+ opioid prescriptions per procedure (reference: <1 opioid prescription per procedure), while the RRs (95% CIs) for long-term opioid prescribing were 1.06 (1.04-1.08), 1.08 (1.06-1.11), and 1.13 (1.11-1.16), respectively. Variation in postoperative analgesic prescribing was observed across U.S. regions. For example, among NSAID-naïve individuals, the RR (95% CIs) for initial opioid prescribing were 0.98 (0.96-1.00) for Region 2 (New York), 1.09 (1.07-1.11) for Region 3 (Philadelphia), 1.07 (1.05-1.10) for Region 4 (Atlanta), 1.03 (1.01-1.05) for Region 5 (Chicago), 1.16 (1.13-1.18) for Region 6 (Dallas), 1.10 (1.08-1.12) for Region 7 (Kansas City), 1.09 (1.06-1.12) for Region 8 (Denver), 1.09 (1.07-1.12) for Region 9 (San Francisco), and 1.11 (1.08-1.13) for Region 10 (Seattle) (reference: Region 1 [Boston]). Hospital characteristics were not meaningfully associated with postoperative analgesic prescribing. The relationships between health system characteristics and postoperative analgesic prescribing were similar for NSAID-naïve and NSAID-prevalent participants.

Discussion

Future efforts aiming to improve the use of multimodal analgesia through increased NSAID prescribing and reduced long-term opioid prescribing following THA/TKA could consider targeting orthopedic surgeons with higher standardized opioid prescribing measures.

Conclusions

Orthopedic surgeon opioid prescribing measures and U.S. region were the greatest health system level predictors of initial, and long-term, prescribing of opioids and prescription NSAIDs among older Medicare beneficiaries following THA/TKA. These results can inform future studies that examine why variation in analgesic prescribing exists across geographic regions and levels of orthopedic surgeon opioid prescribing.

Keywords

geriatric medicine pharmacology systems of care total hip arthroplasty total knee arthroplasty

Introduction

Analgesic prescribing has changed dramatically in the last decade, particularly for postoperative pain management following high-volume procedures such as total hip and knee arthroplasty (THA/TKA). Prescribing patterns have shifted to reduce opioid exposure and increase the use of multimodal analgesia, where a variety of analgesics are incorporated into treatment regimens. Older adults have an increased risk of serious adverse events related to opioids (e.g., hospitalizations, falls and fracture), but they are often susceptible to medication-related adverse effects or have contraindications to therapy with nonopioid analgesics, particularly nonsteroidal anti-inflammatory drugs (NSAIDs).^1,2 Evidence-based guidelines do not make specific recommendations on which opioid and nonopioid combinations are preferred for older adults following THA/TKA.^3-8 This is likely due to a dearth of information on postoperative analgesic prescribing and use post-THA/TKA as well as the factors that influence analgesic prescribing practices. This foundational knowledge is essential to inform future work examining which combinations of multimodal analgesia are safest and most effective among older adults receiving THA/TKA. Optimizing postoperative pain management is especially important for older adults following THA/TKA because these procedures often result in severe pain, which may interfere with functional recovery if treated inadequately.

Further investigation is needed to understand how health system characteristics, such as orthopedic surgeon prescribing patterns, hospital characteristics, and regional factors impact postoperative analgesic prescribing following THA/TKA. Prior observational studies have examined surgeries that are heterogeneous in their pain etiology and course of recovery,^9-14 have considered younger, commercially-insured populations,^10-13,15-17 or have focused on patient-level characteristics (e.g., use of benzodiazepines, back pain) associated with postoperative opioid use after THA/TKA.^14,15,17-20 Understanding the relationships between health system level factors and postoperative analgesic prescribing following THA/TKA can guide future interventions aiming to improve postoperative pain management. For example, if patients treated by orthopedic surgeons who prescribe opioids to a greater extent have an increased risk of long-term opioid use, then interventions aiming to reduce long-term opioid use and dependence could potentially target orthopedic surgeons with such opioid prescribing patterns. Additionally, prior literature only considered use of opioids before THA/TKA and has not examined how preoperative analgesia with nonopioids impacts postoperative pain management.^15,17,21,22 Examining individuals separately based on exposure to preoperative nonopioid analgesics could reveal distinct patterns of postoperative analgesic prescribing that could inform future research efforts.

The objective of this study was to identify health system characteristics, including orthopedic surgeon, hospital, and regional characteristics, associated with initial and long-term prescribing of opioids and prescription NSAIDs among Medicare beneficiaries ≥65 years old receiving elective THA/TKA. We chose to examine NSAIDs specifically as they are the most commonly used prescription nonopioid analgesic following THA/TKA.^23-26 Additionally, we examined how the use of prescription NSAIDs prior to THA/TKA impacted the relationships between health system characteristics and postoperative analgesic prescribing. We hypothesized that the risk of initial and long-term opioid prescribing would be higher among patients with orthopedic surgeons who prescribe opioids to a greater extent, while the risk of initial and long-term prescription NSAID prescribing would be lower.

Materials and Methods

Study Design and Data Sources

This observational study linked data from Medicare fee-for-service (i.e., Traditional Medicare) insurance claims to several publicly available data sources. Medicare claims data included the Master Beneficiary Summary File (MBSF), which included beneficiary plan enrollment information, demographics, and information on conditions through the Chronic Conditions Warehouse; Medicare Provider Analysis and Review (MedPAR) claims for inpatient (i.e., hospital) services; and a 20% random sample of Medicare Part B claims for outpatient services and Part D claims for prescription drug dispensings. The validated Residential History File, which leverages Medicare claims data, provided information on the timing and location of healthcare services.²⁷ Publicly available data included the Medicare Provider Payment and Utilization Data, which was used to construct standardized measures of orthopedic surgeon characteristics; the Dartmouth Health Atlas Hospital Tracking File for hospital characteristics; and the Prescription Drug Abuse Policy System (PDAPS), which contained data on state policies for prescription drug monitoring program (PDMP) utilization.^28,29 All data sources included data from January 1, 2013 to December 31, 2017 to allow for a 12-month covariate ascertainment period prior to THA/TKA procedures. This study was approved by the Brown University Institutional Review Board (Protocol Number: 2010002823). The need for informed consent was waived. The STROBE checklist was used for this cohort study.

Study Population

The study population included community-dwelling Medicare Fee-for-Service beneficiaries aged ≥65 years on the date of elective THA/TKA procedure occurring between January 1, 2014 to July 4, 2017. THA/TKA procedures were identified using Medicare Severity Diagnosis Related Group codes 469 and 470 on inpatient claims. Eligible participants had 12 months of continuous Part A and B coverage and 6 months of continuous Part D coverage prior to THA/TKA. We restricted the study population to individuals without a prescription for opioids in the 6 months prior to THA/TKA (i.e., opioid naïve) because we aimed to estimate the risk of first-time opioid exposure and subsequent risk of long-term opioid use or dependence. A 6-month time frame without opioid use is commonly used to identify opioid naïve individuals.^30-34 Individuals were excluded if they were enrolled in Medicare Advantage or used hospice in the 12 months prior to THA/TKA, were admitted for fracture or non-arthroplasty procedures, or were missing data on orthopedic surgeon characteristics or other covariates (eTable 1). For individuals with multiple procedures, we sampled the first THA/TKA during the study period.

Initial and Long-Term Analgesic Prescribing

The outcomes of interest were initial and long-term prescribing of opioids or prescription NSAIDs following THA/TKA. Initial prescribing was defined as a claim for an opioid or NSAID during days 1-30 following THA/TKA, while claims during days 90-180 after THA/TKA were considered long-term prescribing. Systemic opioids and NSAIDs of interest are listed in eTable 2. Claims for analgesics were captured regardless of prescriber type (e.g., orthopedic surgeon, advanced practice professional).

Follow-Up

Eligible individuals were followed from the date of the THA/TKA procedure until disenrollment from Medicare Parts A, B, or D; enrollment in Medicare Advantage; administrative end of follow-up (180 days post-THA/TKA); or death, whichever occurred first.

Health System Characteristics

We ascertained health system characteristics related to the orthopedic surgeon, hospital, and region. We linked Medicare Part B claims from the THA/TKA to public Part B and D Provider Payment and Utilization Data and created several measures related to orthopedic surgeon characteristics (eTable 3). These included annual THA/TKA volume, opioid prescription volume, proportion of prescriptions for opioids relative to all medications (i.e., “opioid share”), ratio of opioid prescriptions to beneficiaries prescribed opioids (i.e., “refill ratio”), and opioid prescriptions per THA/TKA. Hospital characteristics, including critical access designation, total beds, and academic affiliation were obtained from the Dartmouth Atlas Hospital Tracking File.²⁸ The PDAPS was used to classify the rigor of the PDMP program.²⁹ Categories included: (1) high rigor: prescribers were required to check the PDMP database each time prior to prescribing opioids to all patients or chronic pain patients; (2) low rigor: there were no requirements for checking the PDMP or checking the PDMP was only required prior to prescribing opioids for the first time; and (3) no PDMP was in place. Finally, region was categorized based on the Department of Health and Human Services regions.

Patient Characteristics

We ascertained 127 person-level covariates for case-mix adjustment (eTable 4). Medicare dual eligibility, demographics (age, sex, race/ethnicity³⁵), and comorbidities (e.g., heart failure, chronic kidney disease) were derived from the Medicare Beneficiary Summary File and Chronic Conditions Data Warehouse. Preoperative medication use in the 180 days prior to THA/TKA was ascertained from Part D claims. MedPAR and Medicare Part B claims were used to estimate healthcare utilization in the 12 months prior to THA/TKA (e.g., hospitalizations, annual wellness visit), comorbidity score,³⁶ and a claims-based frailty index.³⁷

Statistical Analysis

We used a two-step process to understand how health system characteristics were associated with the primary outcomes. First, we fit logistic regression models with 127 person-level covariates as independent variables to obtain case-mix adjusted primary outcomes. Second, we estimated risk ratios (RRs) for the potential relationships between health system characteristics and case-mix adjusted outcomes using modified Poisson regression models with robust standard errors.^38,39 Ten orthopedic surgeon, hospital, and regional characteristics were assessed while adjusting for the other health system characteristics. Since prior NSAID use before THA/TKA likely influences medications received in the postoperative period, all analyses were stratified based on the presence of a claim for NSAIDs in the 6 months prior to surgery. Individuals were categorized as NSAID-naïve or NSAID-prevalent.

Intermediate Analyses

Although not the primary focus of this analysis, we reported the results from our intermediate analyses for completeness. These included patterns of analgesic prescribing following THA/TKA stratified by preoperative prescription NSAID exposure, including opioid or NSAID prescribing during days 1-180 following THA/TKA; the top 10 combinations of treatments within 30 days following THA/TKA; prescriptions for short-acting or long-acting opioids; opioid dosage and days’ supply; and sustained opioid use based on proportion of days covered (PDC) ≥80% in the 90 and 180 days following THA/TKA.⁴⁰ Opioid dosages were standardized using the Centers for Disease Control and Prevention conversion factors for morphine milligram equivalents (MME) per day.⁴¹

We also fit modified Poisson regression models with robust standard errors to estimate risk ratios for the potential relationships between 127 person-level characteristics and the primary outcomes, without adjusting for health system characteristics.

Software

Data were analyzed using SAS, version 9.4 (SAS Institute, Inc., Cary, NC) and Stata, version 16 (StataCorp LLC, College Station, TX).

Results

Study Population

The study population included 33,478 eligible Medicare beneficiaries with THA/TKA (eFigure 1). Of these, 23,351 (69.8%) were NSAID-naïve and 10,127 (30.2%) were NSAID-prevalent prior to THA/TKA (Table 1). Compared to NSAID-prevalent individuals, NSAID-naïve individuals were older (age 65-74 years: 58.8% vs 64.9%) and a smaller proportion were female (59.9% vs 65.5%), Medicare-Medicaid dual eligible (4.8% vs 7.3%) and were discharged to home with home health following THA/TKA (44.5% vs 47.2%). A greater proportion of NSAID-naïve participants were Non-Hispanic White (91.4% vs 88.8%); had heart failure (15.2% vs 11.8%), ischemic heart disease (40.3% vs 35.7%), and chronic kidney disease (19.7% vs 14.6%); and were more multimorbid (combined comorbidity score ≥3: 21.5% vs 17.5%) and prefrail (32.0% vs 29.8%) compared to NSAID-prevalent participants (Table 1).

Table 1.

Demographic and Clinical Characteristics of Medicare Beneficiaries following total Knee and Hip Arthroplasty Stratified by Pre-surgery Prescription NSAID Use, 2014-2017.

Characteristics	NSAID-Naïve (n = 23,351)	NSAID-Prevalent (n = 10,127)
Age, years, n (%)
65-74	13,737 (58.8)	6577 (64.9)
75-84	8259 (35.4)	3161 (31.2)
85+	1355 (5.8)	389 (3.8)
Female, n (%)	13,988 (59.9)	6635 (65.5)
Race/ethnicity, n (%)
Non-hispanic white	21,332 (91.4)	8992 (88.8)
Non-hispanic black	719 (3.1)	401 (4.0)
Hispanic	567 (2.4)	304 (3.0)
Other	733 (3.1)	430 (4.2)
Dual-eligible, n (%)	1120 (4.8)	735 (7.3)
Procedure, n (%)
THA	6957 (29.8)	3068 (30.3)
TKA	16,394 (70.2)	7059 (69.7)
Discharge location, n (%)
Home without home health	6666 (28.5)	2778 (27.4)
Home with home health	10,400 (44.5)	4778 (47.2)
SNF	3244 (13.9)	1316 (13.0)
Other	3041 (13.0)	1255 (12.4)
Comorbidities, n (%)
Alcohol use disorder	350 (1.5)	136 (1.3)
Tobacco use	1281 (5.5)	484 (4.8)
Prior opioid-related adverse events	319 (1.4)	134 (1.3)
Cancer	3490 (14.9)	1417 (14.0)
Alzheimer’s disease	988 (4.2)	410 (4.0)
AMI	663 (2.8)	144 (1.4)
Atrial fibrillation	3172 (13.6)	835 (8.2)
Heart failure	3550 (15.2)	1197 (11.8)
Hypertension	18,680 (80.0)	8160 (80.6)
Ischemic heart disease	9407 (40.3)	3617 (35.7)
Stroke/TIA	1976 (8.5)	724 (7.1)
Liver disease	1431 (6.1)	616 (6.1)
CKD	4601 (19.7)	1476 (14.6)
Asthma	2831 (12.1)	1267 (12.5)
COPD	3451 (14.8)	1470 (14.5)
Combined comorbidity score, n (%)
<0^a	3954 (16.9)	1883 (18.6)
0	6126 (26.2)	2955 (29.2)
1	4956 (21.2)	2163 (21.4)
2	3304 (14.1)	1356 (13.4)
3+	5011 (21.5)	1770 (17.5)
Claims-based frailty index, n (%)^b
Robust	15,769 (67.5)	7084 (70.0)
Prefrail	7462 (32.0)	3013 (29.8)
Mild or moderately-to-severely frail	120 (0.5)	30 (0.3)

Abbreviations: THA, total hip arthroplasty; TKA, total knee arthroplasty; SNF, skilled nursing facility; AMI, acute myocardial infarction; TIA, transient ischemic attack; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease.

^aCombined Comorbidity Score values <0 possible due to component value −1 in algorithm for HIV/AIDS and Hypertension.³⁶

^bMeasured using the Claims-based Frailty Index and categorized as: <0.15 (robust), 0.15-0.24 (prefrail), and ≥0.25 (mild or moderately-to-severely frail).³⁷

Initial and Long-Term Analgesic Prescribing Following THA/TKA

A smaller proportion of NSAID-naïve participants had initial (78.3% vs 82.4%) and long-term (12.6% vs 15.7%) postoperative opioid prescribing compared to NSAID-prevalent participants (Table 2). Initial (13.5% vs 24.8%) and long-term (6.8% vs 37.1%) postoperative NSAID prescribing was also lower among NSAID-naïve participants.

Table 2.

Observed Postoperative Analgesic Prescribing Outcomes Among Medicare Beneficiaries With Total Hip and Knee Arthroplasty Stratified by Pre-surgery NSAID Use, 2014-2017.

	NSAID-Naïve (n = 23,351)	NSAID-Prevalent (n = 10,127)
Initial opioid prescribing, n (%)	18,293 (78.3)	8343 (82.4)
Long-term opioid prescribing, n (%)	2947 (12.6)	1585 (15.7)
Initial NSAID prescribing, n (%)	3161 (13.5)	2510 (24.8)
Long-term NSAID prescribing, n (%)	1582 (6.8)	3761 (37.1)

Abbreviations: NSAID, non-steroidal anti-inflammatory drug.

Note: Initial prescribing was defined as a dispensing for an opioid or NSAID during days 1-30 following total hip or knee arthroplasty. Dispensings during days 90-180 after total hip or knee arthroplasty were considered long-term prescribing.

Intermediate Analyses

The mean MME/day, opioid days’ supply, proportion of individuals with opioid PDC ≥80%, prescriptions for long-acting opioids, and the top 10 most common initial postoperative analgesic treatment patterns were similar for NSAID-naïve and NSAID-prevalent participants (eTable 5). Risk ratios for the associations between person-level characteristics and primary outcomes without adjustment for health system characteristics are presented in eTables 6-7.

Health System Characteristics Associated with Initial and Long-Term Opioid Prescribing

Orthopedic Surgeon Characteristics. In general, higher orthopedic surgeon opioid prescription volume was associated with a slightly greater risk of initial opioid prescribing and slightly lower risk of long-term opioid prescribing (Figure 1, eTable 8). For NSAID-naïve individuals, the RRs (95% confidence intervals [CIs]) for initial opioid prescribing were 1.01 (1.00-1.02) for 100-199 opioid prescriptions, 1.03 (1.02-1.04) for 200-299 opioid prescriptions, 1.03 (1.00-1.05) for 300-399 opioid prescriptions, and 1.02 (0.99-1.04) for 400+ opioid prescriptions (reference: 11-99 opioid prescriptions). The RRs (95% CIs) for long-term opioid prescribing were 0.97 (0.95-0.98), 0.96 (0.94-0.98), 0.98 (0.95-1.01), and 0.93 (0.90-0.98), respectively. For NSAID-prevalent individuals, the RRs (95% CIs) for initial opioid prescribing were 1.02 (1.01-1.03) for 100-199 opioid prescriptions, 1.03 (1.01-1.05) for 200-299 opioid prescriptions, 1.04 (1.01-1.07) for 300-399 opioid prescriptions, and 1.01 (0.98-1.05) for 400+ opioid prescriptions (reference: 11-99 opioid prescriptions). The RRs (95% CIs) for long-term opioid prescribing were 0.97 (0.95-0.99), 0.96 (0.93-0.99), 0.94 (0.90-0.98), 0.99 (0.93-1.05), respectively.

Figure 1.

Adjusted associations between health system level characteristics and opioid prescribing. Notes: Presents risk ratios for the potential relationships between health system characteristics and case-mix adjusted outcomes, which accounted for 127 person-level covariates. Reference categories are omitted and can be found in eTable 8. Abbreviations: NSAID, non-steroidal anti-inflammatory drug; THA/TKA, total hip and knee arthroplasty; Rx, prescription; PDMP, Prescription Drug Monitoring Program.

An increase in the opioid share was associated with an increased risk of long-term opioid prescribing among NSAID-prevalent participants only (RR [95% CIs] 1.05 [1.01-1.08] for opioid share 25%–49%, 1.07 [1.03-1.11] for opioid share 50%–74%, and 1.05 [1.01-1.10] for opioid share 75%–100% [reference: opioid share 0%–24%]) (eTable 8).

An increase in the number of orthopedic surgeon opioid prescriptions per THA/TKA procedure generally decreased the risk of initial opioid prescribing and increased the risk for long-term opioid prescribing among NSAID-naïve and NSAID-prevalent beneficiaries (Figure 1, eTable 8). For NSAID-naïve individuals, the RRs (95% CIs) for initial opioid prescribing were 0.97 (0.96-0.98) for 1-2 opioid prescriptions per procedure, 0.97 (0.96-0.98) for 3-4 prescriptions per procedure, and 0.95 (0.93-0.96) for 5+ opioid prescriptions per procedure (reference: <1 opioid prescription per procedure). The RRs (95% CIs) for long-term opioid prescribing were 1.06 (1.04-1.08), 1.08 (1.06-1.11), and 1.13 (1.11-1.16), respectively. For NSAID-prevalent individuals, the RRs (95% CIs) for initial opioid prescribing were 0.98 (0.96-0.99) for 1-2 opioid prescriptions per procedure, 0.98 (0.96-1.00) for 3-4 prescriptions per procedure, and 0.95 (0.93-0.97) for 5+ opioid prescriptions per procedure (reference: <1 opioid prescription per procedure). The RRs (95% CIs) for long-term opioid prescribing were 1.02 (0.99-1.05), 1.06 (1.02-1.09), and 1.08 (1.05-1.12) respectively.

Hospital Characteristics. There were no statistically significant associations between hospital characteristics and initial or long-term opioid prescribing (Figure 1, eTable 8).

Regional Characteristics. The risk of initial opioid prescribing varied by region for both NSAID-naïve and NSAID-prevalent beneficiaries (Figure 1, eTable 8). For NSAID-naïve individuals the RR (95% CIs) for initial opioid prescribing were 0.98 (0.96-1.00) for Region 2 (New York), 1.09 (1.07-1.11) for Region 3 (Philadelphia), 1.07 (1.05-1.10) for Region 4 (Atlanta), 1.03 (1.01-1.05) for Region 5 (Chicago), 1.16 (1.13-1.18) for Region 6 (Dallas), 1.10 (1.08-1.12) for Region 7 (Kansas City), 1.09 (1.06-1.12) for Region 8 (Denver), 1.09 (1.07-1.12) for Region 9 (San Francisco), and 1.11 (1.08-1.13) for Region 10 (Seattle) (reference: Region 1 [Boston]). The RRs (95% CIs) for NSAID-prevalent individuals were 0.94 (0.91-0.97), 1.05 (1.03-1.08), 1.04 (1.01-1.07), 1.02 (1.00-1.05), 1.12 (1.09-1.15), 1.06 (1.03-1.09), 1.05 (1.02-1.09), 1.03 (1.00-1.06), and 1.07 (1.03-1.11), respectively. Regional differences for long-term opioid prescribing were only apparent for NSAID-naïve participants. Statistically significant RRs (95% CIs) were 1.12 (1.08-1.15) for Region 4 (Atlanta), 1.10 (1.07-1.14) for Region 6 (Dallas), and 1.04 (1.01-1.07) for Region 7 (Kansas City) (reference: Region 1 [Boston]).

Health System Characteristics Associated with Initial and Long-Term NSAID Prescribing

Orthopedic Surgeon Characteristics. The relationships between orthopedic surgeon characteristics and initial and long-term NSAID prescribing were similar for NSAID-naïve and NSAID-prevalent individuals (Figure 2, eTable 9).

Figure 2.

Adjusted associations between health system level characteristics and NSAID prescribing. Notes: Presents risk ratios for the potential relationships between health system characteristics and case-mix adjusted outcomes, which accounted for 127 person-level covariates. Reference categories are omitted and can be found in eTable 9. Abbreviations: NSAID, non-steroidal anti-inflammatory drug; THA/TKA, total hip and knee arthroplasty; Rx, prescription; PDMP, Prescription Drug Monitoring Program.

Higher opioid prescription volume was generally associated with an increased risk of initial, but not long-term NSAID prescribing (Figure 2, eTable 9). The RRs (95% CIs) for initial NSAID prescribing among NSAID-naïve individuals were 1.02 (1.00-1.03) for 100-199 opioid prescriptions, 1.04 (1.02-1.06) for 200-299 opioid prescriptions, 1.05 (1.02-1.09) for 300-399 opioid prescriptions, and 1.05 (1.01-1.09) for 400+ opioid prescriptions (reference: 11-99 opioid prescriptions). The RRs (95% CIs) for initial NSAID prescribing were 1.03 (1.01-1.04), 1.04 (1.02-1.06), 1.05 (1.02-1.09), and 1.06 (1.01-1.11), respectively, among NSAID-prevalent individuals.

Increases in the opioid share and refill ratio were associated with a slightly decreased risk of initial NSAID prescribing, but no statistically significant associations were observed for long-term NSAID prescribing (Figure 2, eTable 9). For NSAID-naïve individuals, the RRs (95% CIs) for initial NSAID prescribing were 0.96 (0.93-0.98) for 25%–49% opioid share, 0.96 (0.94-0.99) for 50%–74%, and 0.97 (0.94-1.00) for 75%–100% (reference: 0%–24%). The RRs (95% CIs) were 0.96 (0.95-0.98) for an opioid refill ratio of 2-2.9 and 0.93 (0.90-0.97) for 3+ (reference: 1-1.9). Among NSAID-prevalent individuals, the RRs (95% CIs) for initial NSAID prescribing were 0.95 (0.93-0.98) for 25%–49% opioid share, 0.96 (0.93-0.98) for 50%–74%, and 0.96 (0.93-0.99) for 75%–100% (reference: 0%–24%). The RRs (95% CIs) were 0.97 (0.95-0.98) for an opioid refill ratio of 2-2.9 and 0.97 (0.93-1.02) for 3+ (reference: 1-1.9).

In general, increases in the number of orthopedic surgeon opioid prescriptions per THA/TKA procedure were associated with a decreased risk of initial NSAID prescribing and increased risk of long-term NSAID prescribing (Figure 2, eTable 9). For NSAID-naïve individuals, the RRs (95% CIs) for initial NSAID prescribing were 0.95 (0.93-0.97) for 1-2 opioid prescriptions per procedure, 0.94 (0.92-0.97) for 3-4 prescriptions per procedure, and 0.91 (0.89-0.93) for 5+ opioid prescriptions per procedure (reference: <1 opioid prescription per procedure). The RRs (95% CIs) for long-term NSAID prescribing were 1.01 (1.00-1.03), 1.04 (1.02-1.07), and 1.06 (1.03-1.08), respectively. For NSAID-prevalent individuals, the RRs (95% CIs) for initial NSAID prescribing were 0.99 (0.97-1.01) for 1-2 opioid prescriptions per procedure, 0.97 (0.95-1.00) for 3-4 prescriptions per procedure, and 0.96 (0.93-0.98) for 5+ opioid prescriptions per procedure (reference: <1 opioid prescription per procedure). The RRs (95% CIs) for long-term NSAID prescribing were 1.04 (1.02-1.06), 1.04 (1.02-1.07), and 1.06 (1.04-1.09), respectively.

Hospital Characteristics. Nearly all associations between hospital characteristics and initial or long-term NSAID prescribing did not reach statistical significance (Figure 2, eTable 9).

Regional Characteristics. Regional variation in the risk of initial and long-term NSAID prescribing was observed for NSAID-naïve and NSAID-prevalent participants (Figure 2, eTable 9). For NSAID-naïve individuals the RR (95% CIs) for initial NSAID prescribing were 0.97 (0.94-1.01) for Region 2 (New York), 1.07 (1.05-1.10) for Region 3 (Philadelphia), 1.05 (1.02-1.08) for Region 4 (Atlanta), 0.99 (0.96-1.01) for Region 5 (Chicago), 1.15 (1.11-1.18) for Region 6 (Dallas), 1.06 (1.03-1.09) for Region 7 (Kansas City), 1.07 (1.03-1.10) for Region 8 (Denver), 1.11 (1.08-1.14) for Region 9 (San Francisco), and 1.10 (1.06-1.14) for Region 10 (Seattle) (reference: Region 1 [Boston]). The RRs (95% CIs) for long-term NSAID prescribing were 0.99 (0.96-1.03), 1.00 (0.97-1.03), 1.09 (1.05-1.13), 0.94 (0.92-0.97), 1.08 (1.04-1.11), 0.96 (0.93-1.00), 0.95 (0.91-0.98), 1.02 (0.98-1.05), and 0.92 (0.88-0.96), respectively. Fewer comparisons reach statistical significance for NSAID-prevalent individuals. The statistically significant RRs for initial NSAID prescribing were 1.06 (1.03-1.10) for Region 6 (Dallas) and 1.06 (1.02-1.10) for Region 9 (San Francisco), while the significant RRs for long-term NSAID prescribing were 1.05 (1.01-1.08) for Region 6 (Dallas) and 1.05 (1.01-1.09) for Region 7 (Kansas City) (reference: Region 1 [Boston]).

Discussion

In this large observational study of opioid-naïve Medicare beneficiaries receiving THA/TKA, we found that approximately 30% of individuals used prescription NSAIDs prior to THA/TKA and NSAID-prevalent individuals were more likely to have prescriptions for postoperative opioids and NSAIDs. Orthopedic surgeon opioid prescribing measures and U.S. region were the greatest predictors of initial, and long-term, opioid and NSAID prescribing following THA/TKA. We also found that the relationships between most health system characteristics and postoperative analgesic prescribing were similar among NSAID-naïve and NSAID-prevalent participants.

Clinical guidance recommends prescribing the lowest dose of opioids and the fewest number of opioid pills that is clinically effective following THA/TKA to help prevent opioid-related adverse events and diversion of unused pills.^42,43 Multimodal analgesia with different oral analgesic classes is also recommended to provide pain relief while limiting the use of opioids. Routine use of NSAIDs is recommended for patients without contraindications following THA/TKA, although risk factors for adverse events, such as history of bleeding, gastroduodenal ulcer, cardiovascular morbidity, and kidney and liver disease should be taken into consideration prior to prescribing postoperative NSAIDs.⁵ Multiple studies have focused on person level characteristics associated with postoperative analgesic use,^{14,15,17-20,44-48} reporting that factors like benzodiazepine use, TKA (versus THA), and tobacco use were predictors of long-term opioid use. Very few studies have considered the relationships between health system level characteristics and postoperative analgesic prescribing following THA/TKA. We demonstrated that several measures of orthopedic surgeon opioid prescribing had consistent associations with initial and long-term opioid and NSAID prescribing following THA/TKA. Variation in postoperative opioid and NSAID prescribing was also observed across Department of Health and Human Services regions. However, there were no meaningful relationships between hospital characteristics and postoperative analgesic prescribing. These results can inform future studies that examine why variation in analgesic prescribing exists across geographic regions and levels of orthopedic surgeon opioid prescribing.

Prior literature suggests that high-intensity opioid prescribers provide higher MMEs, greater days’ supply, and are more likely to prescribe opioids long-term vs low-intensity prescribers.^49,50 We found that participants who had orthopedic surgeons with higher standardized opioid prescribing measures (opioid share, opioid refill ratio, opioid prescriptions per THA/TKA procedure) were generally less likely to initiate NSAIDs, but were at an increased risk of long-term opioid prescribing following THA/TKA. It may be beneficial for future efforts aiming to improve the use of multimodal analgesia through increased NSAID prescribing and reduce long-term opioid prescribing following THA/TKA to target orthopedic surgeons with higher standardized opioid prescribing measures. Creating standardized evidence-based guidelines and institutional protocols are potential strategies to combat variability in postoperative prescribing and minimize the risk of unintended long-term analgesic use.^44,51-53

Notably, we found that standardized measures of orthopedic surgeon opioid prescribing created using publicly available provider-level data had opposite associations with postoperative analgesic prescribing compared to orthopedic surgeon opioid prescription volume. For example, higher orthopedic surgeon opioid prescription volume was associated with a lower risk of long-term opioid prescribing, but a greater number of opioid prescriptions per THA/TKA procedure was associated with a higher risk of long-term opioid prescribing for both NSAID-naïve and NSAID-prevalent participants. These results suggest that it is important for future work to consider orthopedic surgeon opioid prescribing in the context of their case load and overall medication prescribing, rather than solely considering the number of opioid prescriptions.

Limitations

There are several potential limitations to this analysis. First, our results may not generalize well to older adults without Medicare Fee-for-Service (i.e., Medicare Advantage), those who were not opioid naïve at the time of THA/TKA, or to those residing outside of the United States.

Second, due to the nature of our prescription drug dispensing data, we could not capture medication administrations (i.e., patterns of medication use). Additionally, we could not identify use of non-prescription (over-the-counter) analgesics. Thus, we could not ascertain use of non-prescription NSAIDs (e.g., ibuprofen, naproxen) or other analgesics that are used as part of multimodal analgesia (e.g., acetaminophen). However, to our knowledge, there is currently no data source with information on the use of non-prescription medications that could have been applied to this national population of Medicare beneficiaries to fill this gap in knowledge.

Third, potentially influential predictors of opioid and NSAID prescribing following THA/TKA, such as beneficiary preoperative and postoperative pain scores, hospital prescribing protocols, and enhanced recovery pathways were not available in our data. Thus, we could not assess the relationships between these potentially important predictors and postoperative analgesic prescribing.

Finally, standardized opioid prescribing measures calculated for orthopedic surgeons should be interpreted in light of several considerations. Since Medicare Part B and D Provider Utilization and Payment data relate to the procedures performed and prescriptions written by each provider in aggregate, these measures do not consider providers’ case mix, including patient health status or prior medication use, which may impact postoperative prescribing practices. Additionally, although we attempted to characterize orthopedic surgeon prescribing, it is possible that another clinician may have been responsible for managing postoperative pain (e.g., advanced practitioner working with the surgeon). In this case, our inferences may be incorrect when attributing postoperative prescribing to the surgeon. However, if the Part D dispensings were captured under the surgeon’s National Provider Identifier through collaborative agreements, our approach would still be valid. Finally, we assumed that the orthopedic surgeon would be responsible for prescribing analgesics in most THA/TKAs performed each year. In our study, 58.6% of beneficiaries overall had a prescription written by an orthopedic surgeon during days 1-89 after THA/TKA, and 70.6% of beneficiaries had a corresponding prescription when restricting to those with a Part D claim. Our finding that greater opioid prescribing intensity was significantly associated with a higher risk of long-term opioid prescribing and lower risk of initial NSAID prescribing could suggest that the relationships were strong enough for people whose pain was managed by the orthopedic surgeon that the results were still significant even when applied to the whole cohort (which included participants who did not receive a prescription from the orthopedic surgeon in the initial postoperative period). Further research is needed to validate these standardized opioid prescribing measures, understand the impact of orthopedic surgeon opioid prescribing intensity on clinical outcomes, examine the clinical appropriateness of opioid and NSAID prescribing, and elucidate the pathways through which these relationships operate.

Conclusion

Overall, we found that orthopedic surgeon opioid prescribing measures and U.S. region were the greatest health system level predictors of initial, and long-term, prescribing of opioids and prescription NSAIDs among older Medicare beneficiaries following THA/TKA. The relationships between health system characteristics and postoperative analgesic prescribing were similar among NSAID-naïve and NSAID-prevalent individuals. Future work aiming to improve the use of multimodal analgesia through increased NSAID prescribing and reduce long-term opioid use following THA/TKA could consider targeting orthopedic surgeons with higher standardized opioid prescribing practices.

Supplemental Material

Supplemental Material - Initial and Long-Term Prescribing of Opioids and Non-steroidal Anti-inflammatory Drugs Following Total Hip and Knee Arthroplasty

Supplemental Material for Initial and Long-Term Prescribing of Opioids and Non-steroidal Anti-inflammatory Drugs Following Total Hip and Knee Arthroplasty by Melissa R. Riester, Elliott Bosco, Francesca L. Beaudoin, Stefan Gravenstein, Andrew J. Schoenfeld, Vincent Mor, and Andrew R. Zullo in Geriatric Orthopaedic Surgery & Rehabilitation.

Footnotes

Declaration of Conflicting Interests

F.L.B. works as a Senior Medical Advisor for the Institute for Clinical and Economic Review with funds paid directly to Brown University. S.G. reports grants from Seqirus, Sanofi; and consulting or speaker fees from Sanofi, Seqirus, Merck, and the Gerontological Society of America related to vaccines or nursing home care quality. A.J.S. reports royalties from Wolters-Kluwer and Springer-Nature. V.M. is a paid consultant and Scientific Advisory Board chair to NaviHealth, Inc. Which offers post-acute care management and services to health plans and health systems. A.R.Z. reports prior grants from Sanofi paid directly to Brown University for research on the epidemiology of infections and vaccinations in older nursing home residents as well as respiratory syncytial virus in infants.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported in part by the National Institute on Aging [R01AG065722 and R01AG077620 to A.R.Z]. Some authors are VA employees (S.G., V.M., A.R.Z.). The content and views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government.

Ethical Statement

ORCID iD

Melissa R. Riester

Data Availability Statement

The data that support the findings of this study are available from the Centers for Medicare and Medicaid Services (CMS) but restrictions apply to the availability of these data, which were used under data use agreements for the current study, and so are not publicly available. However, other researchers can establish their own data use agreement and obtain the datasets employed through the Research Data Assistance Center (ResDAC), a CMS contractor that provides free assistance to researchers interested in CMS data.*

Supplemental Material

Supplemental material for this article is available online.

References

Weiss

Heslin

Barrett

Izar

Bierman

. Opioid-Related Inpatient Stays and Emergency Department Visits Among Patients Aged 65 Years and Older, 2010 and 2015: Statistical Brief #244. Healthcare Cost and Utilization Project (HCUP) Statistical Briefs 2018.

O'Neil

Hanlon

Marcum

. Adverse effects of analgesics commonly used by older adults with osteoarthritis: focus on non-opioid and opioid analgesics. Am J Geriatr Pharmacother. 2012;10(6):331-342. doi:10.1016/j.amjopharm.2012.09.004.

American Academy of Orthopaedic Surgeons (AAOS) . Surgical Management of Osteoarthritis of the Knee: Evidence-Based Clinical Practice Guideline. Rosemont: American Academy of Orthopaedic Surgeons (AAOS); 2015.

Chou

Gordon

de Leon-Casasola

, 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 committee, and administrative council. J Pain : Official Journal of the American Pain Society. 2016;17(2):131-157. doi:10.1016/j.jpain.2015.12.008.

Wainwright

Gill

McDonald

, et al. Consensus statement for perioperative care in total hip replacement and total knee replacement surgery: enhanced Recovery after Surgery (ERAS(®)) Society recommendations. Acta Orthop. 2020;91(1):3-19. doi:10.1080/17453674.2019.1683790.

Hannon

Fillingham

Nam

, et al. Opioids in total joint arthroplasty: the clinical practice guidelines of the American association of hip and knee surgeons, American society of regional anesthesia and pain medicine, American academy of orthopaedic surgeons, hip society, and knee society. J Arthroplasty. 2020;35(10):2709-2714. doi:10.1016/j.arth.2020.05.034.

Anger

Valovska

Beloeil

, et al. PROSPECT guideline for total hip arthroplasty: a systematic review and procedure-specific postoperative pain management recommendations. Anaesthesia. 2021;76(8):1082-1097. doi:10.1111/anae.15498.

Lavand'homme

Kehlet

Rawal

Joshi

Anaesthesia

PWGESR

Pain

. Pain management after total knee arthroplasty: PROcedure SPEcific Postoperative Pain ManagemenT recommendations. Eur J Anaesthesiol. 2022;39(9):743-757. doi:10.1097/EJA.0000000000001691.

Santosa

Brummett

, et al. New persistent opioid use among older patients following surgery: a Medicare claims analysis. Surgery. 2020;167(4):732-742. doi:10.1016/j.surg.2019.04.016.

10.

Chen

Jeffery

Krebs

, et al. Long-term trends in postoperative opioid prescribing, 1994 to 2014. J Am Acad Orthop Surg Glob Res Rev. 2020;4(1):00171. doi:10.5435/JAAOSGlobal-D-19-00171.

11.

Klueh

Howard

, et al. Transitions of care for postoperative opioid prescribing in previously opioid-naïve patients in the USA: a retrospective review. J Gen Intern Med. 2018;33(10):1685-1691. doi:10.1007/s11606-018-4463-1.

12.

Brummett

Waljee

Goesling

, et al. New persistent opioid use after minor and major surgical procedures in US adults. JAMA Surgery. 2017;152(6):e170504. doi:10.1001/jamasurg.2017.0504.

13.

Sun

Darnall

Baker

Mackey

. Incidence of and risk factors for chronic opioid use among opioid-naive patients in the postoperative period. JAMA Intern Med. 2016;176(9):1286-1293. doi:10.1001/jamainternmed.2016.3298.

14.

Reid

DBC

Shah

Shapiro

Ruddell

Akelman

Daniels

. Mandatory prescription limits and opioid utilization following orthopaedic surgery. J Bone Joint Surg Am. 2019;101(10):e43. doi:10.2106/JBJS.18.00943.

15.

Cancienne

Patel

Browne

Werner

. Narcotic use and total knee arthroplasty. J Arthroplasty. 2018;33(1):113-118. doi:10.1016/j.arth.2017.08.006.

16.

Sutherland

Wunsch

Pinto

, et al. Association of the 2016 US Centers for disease Control and prevention opioid prescribing guideline with changes in opioid dispensing after surgery. JAMA Netw Open. 2021;4(6):e2111826. doi:10.1001/jamanetworkopen.2021.11826.

17.

Zeng

, et al. Prevalence and risk factors for prolonged opioid use after total joint arthroplasty: a systematic review, meta-analysis, and meta-regression. Arch Orthop Trauma Surg. 2021;141(6):907-915. doi:10.1007/s00402-020-03486-4.

18.

Kent

Hurley

Oderda

, 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.

19.

Reid

DBC

Shapiro

Shah

, et al.

Has a prescription-limiting law in Rhode Island helped to reduce opioid use after total joint arthroplasty?

Clin Orthop Relat Res. 2020;478(2):205-215. doi:10.1097/CORR.0000000000000885.

20.

Klueh

Howard

, et al. Transitions of care for postoperative opioid prescribing in previously opioid-naive patients in the USA: a retrospective review. J Gen Intern Med. 2018;33(10):1685-1691. doi:10.1007/s11606-018-4463-1.

21.

Hannon

Fillingham

Nam

, et al. The efficacy and safety of opioids in total joint arthroplasty: systematic review and direct meta-analysis. J Arthroplasty. 2020;35(10):2759-2771. doi:10.1016/j.arth.2020.05.032.

22.

Blevins

Weeks

Pisansky

AJB

Yong

Stringer

. Effect of preoperative opioid use on adverse outcomes, medical spending, and persistent opioid use following elective total joint arthroplasty in the United States: a large retrospective cohort study of administrative claims data. Pain Med. 2020;21(3):521-531. doi:10.1093/pm/pnz083.

23.

Turppo

Sund

Huopio

Kroger

Sirola

. Pain medication purchases before and after total hip and knee arthroplasty: a register study of 329,743 arthroplasties. Acta Orthop. 2022;93:534-541. doi:10.2340/17453674.2022.2929.

24.

Rajamaki

Jr. Puolakka

Hietaharju

Moilanen

Jamsen

. Use of prescription analgesic drugs before and after hip or knee replacement in patients with osteoarthritis. BMC Muscoskel Disord. 2019;20(1):427. doi:10.1186/s12891-019-2809-4.

25.

Ohnuma

Raghunathan

Ellis

, et al. Effects of acetaminophen, NSAIDs, gabapentinoids, and their combinations on postoperative pulmonary complications after total hip or knee arthroplasty. Pain Med. 2020;21(10):2385-2393. doi:10.1093/pm/pnaa017.

26.

Thijssen

Timmerman

Koning

, et al. Multimodal analgesia practices for knee and hip arthroplasties in The Netherlands. A prospective observational study from the PAIN OUT registry. PLoS One. 2022;17(12):e0279606. doi:10.1371/journal.pone.0279606.

27.

Intrator

Hiris

Berg

Miller

Mor

. The residential history file: studying nursing home residents' long-term care histories(*). Health Serv Res. 2011;46(1 Pt 1):120-137. doi:10.1111/j.1475-6773.2010.01194.x.

28.

Dartmouth Atlas supplemental research data. 2020. https://atlasdata.dartmouth.edu/downloads/supplemental#hospital

29.

Prescription drug Abuse policy system - PDAPS. http://pdaps.org/

30.

Hayden

Ellis

Asbridge

, et al. Prolonged opioid use among opioid-naive individuals after prescription for nonspecific low back pain in the emergency department. Pain. 2021;162(3):740-748. doi:10.1097/j.pain.0000000000002075.

31.

Shah

Hayes

Martin

. Factors influencing long-term opioid use among opioid naive patients: an examination of initial prescription characteristics and pain etiologies. J Pain. 2017;18(11):1374-1383. doi:10.1016/j.jpain.2017.06.010.

32.

Jeffery

Hooten

Hess

, et al. Opioid prescribing for opioid-naive patients in emergency departments and other settings: characteristics of prescriptions and association with long-term use. Ann Emerg Med. 2018;71(3):326-336. doi:10.1016/j.annemergmed.2017.08.042.

33.

Ukert

Huang

Sennett

Delgado

. State-level variation in opioid prescribing after knee arthroscopy among the opioid-naive in the USA: 2015-2019. BMJ Open. 2020;10(8):e035126. doi:10.1136/bmjopen-2019-035126.

34.

Berger

Strother

Talwar

, et al. National variation in opioid prescription fills and long-term use in opioid naive patients after urological surgery. J Urol. 2019;202(5):1036-1043. doi:10.1097/JU.0000000000000343.

35.

Eicheldinger

Bonito

. More accurate racial and ethnic codes for Medicare administrative data. Health Care Financ Rev. Spring. 2008;29(3):27-42.

36.

Gagne

Glynn

Avorn

Levin

Schneeweiss

. A combined comorbidity score predicted mortality in elderly patients better than existing scores. J Clin Epidemiol. 2011;64(7):749-759. doi:10.1016/j.jclinepi.2010.10.004.

37.

Kim

Schneeweiss

Glynn

Lipsitz

Rockwood

Avorn

. Measuring frailty in Medicare data: development and validation of a claims-based frailty index. The journals of gerontology Series A, Biological sciences and medical sciences. 2018;73(7):980-987. doi:10.1093/gerona/glx229.

38.

Zou

. A modified Poisson regression approach to prospective studies with binary data. American journal of epidemiology. 2004;159(7):702-706.

39.

Cummings

. The relative merits of risk ratios and odds ratios. Arch Pediatr Adolesc Med. 2009;163(5):438-445. doi:10.1001/archpediatrics.2009.31.

40.

Bryson Chris

Rumsfeld

. Medication adherence: its importance in cardiovascular outcomes. Circulation. 2009;119(23):3028-3035. doi:10.1161/CIRCULATIONAHA.108.768986.

41.

Dowell

Haegerich

Chou

. CDC guideline for prescribing opioids for chronic pain - United States, 2016. MMWR Recommendations and reports : MMWR Recomm Rep (Morb Mortal Wkly Rep). 2016;65(1):1-49. doi:10.15585/mmwr.rr6501e1.

42.

Hannon

Fillingham

Nam

, et al. The efficacy and safety of opioids in total joint arthroplasty: systematic review and direct meta-analysis. J Arthroplasty. 2020;35(10):2759-2771. doi:10.1016/j.arth.2020.05.032.

43.

Hannon

Fillingham

Nam

44.

Sceats

Ayakta

Merrell

Kin

. Drivers, beliefs, and barriers surrounding surgical opioid prescribing: a qualitative study of surgeons’ opioid prescribing habits. J Surg Res. 2020;247:86-94. doi:10.1016/j.jss.2019.10.039.

45.

Hooten

Brummett

Sullivan

, et al. A conceptual framework for understanding unintended prolonged opioid use. Mayo Clinic proceedings. 2017/12/01/. 2017;92(12):1822-1830. doi: 10.1016/j.mayocp.2017.10.010.

46.

Bicket

Brat

Hutfless

Nesbit

Alexander

. Optimizing opioid prescribing and pain treatment for surgery: review and conceptual framework. Am J Health Syst Pharm. 2019;76(18):1403-1412. doi:10.1093/ajhp/zxz146.

47.

Lawal

Gold

Murthy

, et al. Rate and risk factors associated with prolonged opioid use after surgery: a systematic review and meta-analysis. JAMA Netw Open. 2020;3(6):e207367. doi:10.1001/jamanetworkopen.2020.7367.

48.

Rajamaki

Puolakka

Hietaharju

Moilanen

Jamsen

. Predictors of the use of analgesic drugs 1 year after joint replacement: a single-center analysis of 13,000 hip and knee replacements. Arthritis Res Ther. 2020;22(1):89. doi:10.1186/s13075-020-02184-1.

49.

Chang

H-Y

Murimi

Jones

Alexander

. Relationship between high-risk patients receiving prescription opioids and high-volume opioid prescribers. Addiction. 2018;113(4):677-686. doi:10.1111/add.14068.

50.

Barnett

Olenski

Jena

. Opioid-prescribing patterns of emergency physicians and risk of long-term use. N Engl J Med. 2017;376(7):663-673. doi:10.1056/NEJMsa1610524.

51.

Goel

Feinberg

McGuiness

, et al. Postoperative opioid-prescribing patterns among surgeons and residents at university-affiliated hospitals: a survey study. Canadian journal of surgery Journal canadien de chirurgie. 2020;63(1):E1-E8. doi:10.1503/cjs.016518.

52.

Lipof

Thirukumaran

Greenstein

Zmich

Lander

Ricciardi

. Postdischarge opiate-prescribing habits for primary THA and TKA: a survey of American association of hip and knee surgeons members. Orthopedics. 2019;42(6):361-367. doi:10.3928/01477447-20190723-06.

53.

Woelber

Wurster

Brandt

Mecum

Gundle

Anissian

. Minimizing opioids after joint operation: protocol to decrease postoperative opioid use after primary total knee arthroplasty. Fed Pract. 2021;38(2):e1-e7. doi:10.12788/fp.0092.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.69 MB