Variations in Current Practice and Protocols of Intraoperative Multimodal Analgesia: A Cross-Sectional Study Within a Six-Hospital US Health Care System

Abstract

BACKGROUND:

Multimodal analgesia (MMA) aims to reduce surgery-related opioid needs by adding nonopioid pain medications in postoperative pain management. In light of the opioid epidemic, MMA use has increased rapidly over the past decade. We hypothesize that the rapid adoption of MMA has resulted in variation in practice. This cross-sectional study aimed to determine how MMA practices have changed over the past 6 years and whether there is variation in use by patient, provider, and facility characteristics.

METHODS:

Our study population includes all patients undergoing surgery with general anesthesia at 1 of 6 geographically similar hospitals in the United States between January 1, 2017 and December 31, 2022. Intraoperative pain medications were obtained from the hospital’s perioperative information management system. MMA was defined as an opioid plus at least 2 other nonopioid analgesics. Frequencies, χ² tests (χ²), range, and interquartile range (IQR) were used to describe variation in MMA practice over time, by patient and procedure characteristics, across hospitals, and across anesthesiologists. Multivariable logistic regression was conducted to understand the independent contributions of patient and procedural factors to MMA use.

RESULTS:

We identified 25,386 procedures among 21,227 patients. Overall, 46.9% of cases met our definition of MMA. Patients who received MMA were more likely to be younger females with a lower comorbidity burden undergoing longer and more complex procedures that included an inpatient admission. MMA use has increased steadily by an average of 3.0% each year since 2017 (95% confidence interval =2.6%–3.3%). There was significant variation in use across hospitals (n = 6, range =25.9%–68.6%, χ² = 3774.9, P < .001) and anesthesiologists (n = 190, IQR =29.8%–65.8%, χ² = 1938.5, P < .001), as well as by procedure characteristics. The most common MMA protocols contained acetaminophen plus regional anesthesia (13.0% of protocols) or acetaminophen plus dexamethasone (12.2% of protocols). During the study period, the use of opioids during the preoperative or intraoperative period decreased from 91.4% to 86.0% of cases; acetaminophen use increased (41.9%–70.5%, P < .001); dexamethasone use increased (24.0%–36.1%, P < .001) and nonsteroidal anti-inflammatory drugs (NSAIDs) increased (6.9%–17.3%, P < .001). Gabapentinoids and IV lidocaine were less frequently used but also increased (0.8%–1.6% and 3.4%–5.3%, respectively, P < .001).

CONCLUSIONS:

In a large integrated US health care system, approximately 50% of noncardiac surgery patients received MMA. Still, there was wide variation in MMA use by patient and procedure characteristics and across hospitals and anesthesiologists. Our findings highlight a need for further research to understand the reasons for these variations and guide the safe and effective adoption of MMA into routine practice.

KEY POINTS.

• Question: How has multimodal analgesia use changed over the past 6 years, and is there variation in use by patient, provider, and facility characteristics?

• Findings: The use of multimodal analgesia for postoperative pain management increased significantly over the past 6 years and varied notably, not just by patient characteristics and medication combinations used but also by the anesthesia provider and facility.

• Meaning: These findings highlight a need for further high-quality research to understand the reasons for these variations.

Multimodal analgesia (MMA) is a postoperative pain management technique that involves using 2 or more nonopioid pain medications alongside opioids to reduce opioid-related adverse side effects. Not only do fewer opioid-related side effects reduce hospital length of stays and the risk of readmission, but an MMA approach may also help decrease unnecessary and excessive postoperative opioid prescriptions.^1–4 Although this approach is not new, it has gained popularity over the last decade due to its potential to reduce postoperative opioid use and address the opioid epidemic.^5,6 Efforts to increase the adoption of MMA protocols have included electronic health record system order set interventions and implementation as part of Enhanced Recovery After Surgery (ERAS) programs.^7,8 Despite its benefits, the adoption of MMA into practice is still in its early stages, and there is very little guidance for which combinations of medications are most effective and in which populations.⁹

The translation of evidence into medical practice can vary widely and is influenced by numerous factors, including clinical relevance, the strength of the evidence, endorsements by professional medical societies, effective educational strategies, and even the regulatory environment of the hospital.^10,11 Understanding current practice patterns is a crucial step in developing evidence-based MMA protocols. By analyzing practice patterns, guideline developers can tailor recommendations to align with real-world clinical scenarios, ensuring that guidelines are relevant, practical, and effective in improving patient outcomes.¹² Thus, our objectives were to describe the adoption of MMA over the past 6 years and detail current use patterns in a sample of 6 hospitals in a single United States health care system. We hypothesize that MMA use for postoperative pain management has increased over the past 6 years, and practice patterns vary significantly by patient, provider, and facility characteristics.

METHODS

This is a retrospective cross-sectional study of patients undergoing elective noncardiac surgery at one of 6 geographically similar hospitals in a single United States health care system, the Veterans Health Administration (VA). The Stanford Institutional Review Board approved the study with a waiver of informed consent. This manuscript adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. Study methods, including the primary exposure and analytic methods, were established during the study design.

Data Sources and Study Population

Data was obtained from the Picis Clinical Solutions Surgery Management System and supplemented with VA Corporate Data Warehouse data. Picis is a clinical management software some VA facilities use to collect, manage, and report perioperative data for Veterans visiting an operating room, postanesthesia care unit, or intensive care unit. The VA Corporate Data Warehouse is the data repository for all VA clinical and administrative data.

The study population was identified as all patients undergoing an elective noncardiac surgical procedure with general anesthesia between January 1, 2017, and December 31, 2022. This analysis was conducted at the procedure level, so patients could have contributed more than 1 observation to the dataset. Patients were excluded if they underwent an ophthalmology, podiatry, or endovascular procedure.

Variables

The primary exposure for this analysis is MMA. MMA is defined as 2 or more nonopioid medications in addition to an opioid administered no more than 6 hours before the operation and before the end of the operation, per the current CMS quality measure.¹³ Nonopioid pain medications were classified as shown in Table 1.¹⁴ Our inclusion criteria for medications were intended to be broad to capture all potential medications. However, unlike our other nonopioid medications, dexamethasone is more commonly used as an antiemetic. The typical dexamethasone dose administered for postoperative nausea and vomiting is a dose of 4 to 8 mg.¹⁵ To reduce the inclusion of dexamethasone administrations that were intended for postoperative nausea and vomiting, we only included dexamethasone as a nonopioid pain medication when 8mg or more were administered during the perioperative period (57.6% of all administrations).¹⁵ Medications were identified from the Picis Anesthesia Manager and the CDW BCMA, RxIV, and Surgery Medication tables to ensure complete capture of all intraoperative medications.

Table 1.

Classification of Non-opioid Medications and Techniques Included as MMA

Medication class	Medication brand name	Perioperative role(s)
Acetaminophen	Acetaminophen	Analgesia
NSAID	Celecoxib	Analgesia Antipyretic Anti-inflammatory
	Diclofenac
	Ketorolac
	Meloxicam
Gabapentinoids	Gabapentin	Neuropathic pain Antiseizure
	Pregabalin
Dexamethasone (8 mg or more)	Dexamethasone	Antiemetic (1st line) Analgesia
NMDA	Ketamine	Analgesia Increases BP, HR, and CO Bronchodilation
	Magnesium	Analgesia Muscle relaxation
Alpha2	Dexmedetomidine	Analgesia Anxiolysis Perioperative sympatholysis Cardiovascular stabilization
Regional anesthesia	Epidural anesthesia	Analgesia
	Spinal anesthesia
	Peripheral nerve blocks
	Intravenous lidocaine

Abbreviations: Alpha2, Alpha2-adrenergic receptor agonists; BP, blood pressure; CO, cardiac output; HR, heart rate; MMA, multimodal analgesia; NMDA, N-methyl-D-aspartate receptor antagonists; NSAID, nonsteroidal anti-inflammatory drug.

Demographic and procedure data were obtained solely from the CDW. We collected patient age, race, and gender from the CDW Patient domain. We measured comorbidity burden using the patient’s American Society of Anesthesiologists (ASA) classification at the time of surgery and the Care Assessment Needs (CAN) score calculated for the patient the week before the operation. The CAN score is a statistical model used to predict high-risk patients and is automatically generated from electronic health record data. It is determined using demographic, clinical, health care utilization, and social data and is often used in decision-making in the clinical setting. It was developed by VA to help primary care providers identify high-risk patients who could benefit from health care coordination.¹⁶ Procedure characteristics included operative time, work relative value unit (RVU), and surgical specialty.

We obtained the name of the anesthesiologist for each procedure from the Picis Anesthesia Manager. Picis includes several data fields to describe the anesthesia team for each case. These include Anesthesiologist, Certified Registered Nurse Anesthetist, Fellow, Resident, and Student. For our analyses, we used the name of the supervising anesthesiologist as the name of the anesthesia provider for each procedure. When the name of the supervising anesthesiologist was missing (52%), we attributed the name of the principal anesthesiologist to the case. After this algorithm, we identified 190 unique anesthesiologists. Sixty-six percent (66%) of our final list were identified as anesthesiologists in the CDW Staff tables (17% CRNA, 15% Fellow/Resident, and 2% Missing).

Statistical Analysis

Our analysis was primarily descriptive. Distributional characteristics and missingness were checked before beginning analyses. After visualizing the distribution of all continuous variables, the probability of mortality within 90 days and operative time were found to be nonnormally distributed. Outside of the high proportion of missing data points for the supervising anesthesiologist mentioned above, missing data among all other variables was minimal (<5%, Table 2). Individuals missing data points were not excluded from the analysis. χ² tests and t-tests were used to examine differences in MMA use.

Table 2.

Patient and Procedure Characteristics by Receipt of MMA

Characteristic	Multimodal analgesia (MMA) Used			Difference (95% CI)a
	Overall	No	Yes
	N = 25,386	n = 13,468 (53.1%)	n = 11,918 (46.9%)
Sex, n (%)				0.14 (0.11–0.16)
Female	2554	1091 (42.7)	1463 (57.3)
Male	22,814	12,368 (54.2)	10,446 (45.8)
Missing	18	9 (50.0)	9 (50.0)
Patient age, y				−4.1 (−3.7 to −4.4)
Mean (SD)	63.2 (13.8)	65.1 (13.2)	61.1 (14.1)
ASA class, n (%)				0.19 (0.17–0.22)
Normal health patient	631	260 (41.2)	371 (58.8)
Mild systemic disease	6445	3067 (47.6)	3378 (52.4)
Severe systemic disease	16,795	9138 (54.4)	7657 (45.6)
Severe systemic disease that is a constant threat to life	1515	1003 (66.2)	512 (33.8)
Probability of mortality within 90 d (CANS)
Median (IQR)	0.3 (0.1, 0.9)	0.4 (0.1, 1.1)	0.3 (0.1, 0.7)	0.45 (0.39–0.50)
Missing	807	448	359
Year, n (%)				0.22 (0.19–0.24)
2017	4571	2801 (61.3)	1770 (38.7)
2018	4756	2710 (57.0)	2046 (43.0)
2019	4787	2473 (51.7)	2314 (48.3)
2020	3252	1740 (53.5)	1512 (46.5)
2021	4109	1933 (47.0)	2176 (53.0)
2022	3911	1811 (46.3)	2100 (53.7)
Surgical specialty, n (%)				0.55 (0.52–0.57)
Orthopedics	8011	3327 (41.5)	4684 (58.5)
General surgery	6887	3557 (51.6)	3330 (48.4)
Urology	4685	3531 (75.4)	1154 (24.6)
Vascular	1603	1142 (71.2)	461 (28.8)
Neurosurgery	1094	552 (50.5)	542 (49.5)
Ear, nose, throat	852	377 (44.2)	475 (55.8)
Plastic surgery	556	298 (53.6)	258 (46.4)
Obstetrics, gynecology	521	180 (34.5)	341 (65.5)
Thoracic surgery	504	229 (45.4)	275 (54.6)
Podiatry	473	208 (44.0)	265 (56.0)
Oral surgery	200	67 (33.5)	133 (66.5)
Operative time, h				0.81 (0.85–0.76)
Median (IQR)	1.8 (1.1, 2.9)	1.5 (0.9, 2.4)	2.2 (1.5, 3.4)
Work relative value unit				3.2 (3.4–3.0)
Mean (SD)	14.2 (8.2)	12.7 (7.6)	15.9 (8.5)
Admission status, n (%)				0.29 (0.26–0.31)
Inpatient	10,268	4561 (44.4)	5707 (55.6)
Outpatient	15,118	8907 (58.9)	6211 (41.1)

Abbreviations: ASA, American Society of Anesthesiologists; CANS, Care Assessment Needs Score; CI, confidence interval, IQR, interquartile range; MMA, multimodal analgesia; SD, standard deviation.

^aStandardized mean difference with 95% confidence interval.

Variability in MMA use across hospitals and anesthesiologists was described using the range and interquartile range, respectively, and the χ² test statistic (χ²). We used a mixed-effects multivariable logistic regression to understand the independent contributions of patient and procedural factors to MMA use. Anesthesiologists and hospitals were included in the model as random effects, and all other covariates were included in the model as fixed effects. The relative contribution of each covariate to the model’s fit was calculated using the proportion of variance in MMA explained.¹⁷

We also conducted a sensitivity analysis excluding dexamethasone from our definition of MMA. While all other nonopioid medications included in our study are most commonly used for analgesic purposes in the context of surgery, dexamethasone is most commonly used for postoperative nausea and vomiting. Since we were unable to determine the indication for giving dexamethasone, we repeated all analyses excluding dexamethasone for our definition of MMA.

RESULTS

Over our 6-year study period, 25,386 eligible procedures were performed. Forty-seven percent (46.9%) of patients received at least 2 nonopioid medications in addition to an opioid (MMA). Patients who received MMA were very different from patients who did not (Table 2). They were more often females (57.3% vs 45.8%, P < .001) of younger age (mean 61.1 vs 65.1 years, P < .001) with a lower comorbidity burden. The use of MMA also varied drastically by surgical specialty. Orthopedic procedures used MMA in 58.5% of cases, while only 24.6% of urology procedures used MMA (P < .001). Procedures involving MMA were also significantly longer (mean 2.7 vs 1.9 hours, P < .001), more complex (mean work RVU 15.9 vs 12.7, P < .001), and more often inpatient procedures as opposed to outpatient (55.6% vs 41.1%, P < .001, Table 2). In our final adjusted logistic model of MMA use, the anesthesiologist, facility, surgical specialty, and operative time explained the most variation in MMA use (Table 3).

Table 3.

Adjusted Model of MMA Use

Characteristic	OR for MMA use	P value	Relative importancea
	(95% CI)
Gender			0.17
Female	—
Male	0.86 (0.77–0.97)	.010
Unknown	1.50 (0.50–4.49)	.47
Patient age, y	0.99 (0.98–0.99)	<.001	1.10
ASA class			0.51
Normal health patient	—
Mild systemic disease	1.08 (0.88–1.33)	.47
Severe systemic disease	0.90 (0.73–1.11)	.32
Severe systemic disease that is a constant threat to life	0.67 (0.52–0.87)	.002
Probability of mortality	0.96 (0.94–0.98)	<.001	0.39
Year			0.84
2017	—
2018	1.17 (1.05–1.30)	.003
2019	1.39 (1.25–1.55)	<.001
2020	1.19 (1.06–1.34)	.003
2021	1.80 (1.61–2.02)	<.001
2022	2.03 (1.80–2.28)	<.001
Surgical specialty			5.46
Orthopedics	—
General surgery	0.54 (0.50–0.59)	<.001
Urology	0.20 (0.18–0.22)	<.001
Peripheral vascular	0.14 (0.12–0.16)	<.001
Neurosurgery	0.32 (0.27–0.38)	<.001
Ear, nose, throat	0.60 (0.50–0.71)	<.001
Plastic surgery	0.33 (0.27–0.40)	<.001
Obstetrics, gynecology	0.88 (0.70–1.10)	.26
Thoracic surgery	0.56 (0.44–0.71)	<.001
Podiatry	0.62 (0.50–0.77)	<.001
Oral surgery	0.70 (0.49–1.00)	.052
Operative time, h	1.29 (1.26–1.32)	<.001	2.56
Work relative value unit	1.03 (1.02–1.03)	<.001	1.69
Admission status			1.11
Inpatient	—
Outpatient	0.67 (0.62–0.73)	<.001
Random effects
Facility			3.65
Anesthesiologist			11.37
Log-likelihood	−13,191
No. obs.	24,579

Abbreviations: ASA, American Society of Anesthesiologists; CANS, Care Assessment Needs Score; CI, confidence interval; MMA, multimodal analgesia; OR, odds ratio.

^aR² partitioned by averaging over orders (Lindeman et. al. 1980).

Over time, MMA use has increased by an average of 3.0% each year since 2017 (95% confidence interval = 2.6%–3.3%), with significant variation in use across hospitals (Figure, n = 6, range = 25.9%–68.6%, χ² = 1938.5, P < .001) and anesthesiologists (n = 190, IQR = 29.8%–65.8%, χ² = 3774.9, P < .001). As shown in the Figure, some hospitals exhibited near-stable use of MMA during the study period, while others either rapidly adopted or even reduced their use of MMA.

Figure.

Change in MMA use over time. Top panel: Adoption of MMA over time. Bottom panel: Adoption of MMA over time by facility. *Ribbons represent the 95% confidence interval. MMA indicates multimodal analgesia.

During the study period, 158 unique combinations of medications were used in MMA protocols. The most common protocols included acetaminophen (55.3% of all MMA protocols), regional anesthesia (34.3% of all MMA protocols), and/or dexamethasone (32.2% of all MMA protocols, Table 4). The most common protocols contain acetaminophen plus regional anesthesia (13.0% of all MMA protocols) or acetaminophen plus dexamethasone (12.2% of all MMA protocols).

Table 4.

Top 20 Most Common MMA Protocols During Study Period

n	% of all MMA Protocols (n = 11,918)	Combination
1555	13.0%	Acetaminophen + regional
1451	12.2%	Acetaminophen + dexamethasone
735	6.2%	Acetaminophen + dexamethasone + regional
692	5.8%	Dexamethasone + regional
491	4.1%	Acetaminophen + NMDA + regional
458	3.8%	Acetaminophen + dexamethasone + NMDA
454	3.8%	Acetaminophen + NMDA
352	3.0%	Acetaminophen + NSAID + regional
350	2.9%	Acetaminophen + dexamethasone + NMDA + regional
342	2.9%	Acetaminophen + Alpha-2
333	2.8%	Acetaminophen + NSAID
225	1.9%	Acetaminophen + dexamethasone + NSAID
224	1.9%	NMDA + regional
218	1.8%	Dexamethasone + NSAID
215	1.8%	Acetaminophen + dexamethasone + Alpha-2
194	1.6%	Acetaminophen + Alpha-2 + regional
183	1.5%	Dexamethasone + NMDA
159	1.3%	Acetaminophen + dexamethasone + NSAID + regional
148	1.2%	Acetaminophen + IV + lidocaine
140	1.2%	Acetaminophen + NMDA + NSAID + regional

Abbreviations: Alpha-2 = Alpha2-adrenergic receptor agonist; MMA, multimodal analgesia; NMDA = N-methyl-D-aspartate receptor antagonist; NSAID = nonsteroidal anti-inflammatory drug.

The use of opioids in the preoperative or intraoperative period has decreased only slightly from 91.4% in 2017 to 86.0% in 2022, while the use of all non-opioid medications in our study increased markedly. Acetaminophen use has risen from 41.9% in 2017 to 70.5% in 2022 (P < .001), and dexamethasone use increased from 24.0% in 2017 to 36.1% in 2022 (P < .001). Although not as frequently used, NSAIDs also more than doubled over the study period, from 6.9% in 2017 to 17.3% of cases in 2022 (P < .001). Gabapentinoids and intravenous lidocaine were less frequently used but still increased over our study period (0.8% to 1.6% and 3.4% to 5.3%, respectively, P < .001 for both, Supplemental Digital Content 1, Supplemental Table 1, http://links.lww.com/AA/F99).

Dexamethasone was used in 32.2% of MMA protocols and was more commonly used in female patients, even after limiting our definition of MMA dexamethasone to quantities of 8mg or more administered during the perioperative period (female 46.9% vs male 30.5%, P < .001, Supplemental Digital Content 1, Supplemental Table 2, http://links.lww.com/AA/F99). When dexamethasone was removed from our definition of MMA, we found that 36.3% of patients received MMA during the study period. The same trends of increasing MMA use over time remained (29.5% in 2017 to 43.2% in 2022, P < .001). MMA use also still varied significantly across hospitals (n = 6, range = 23.0%–60.9%, χ² = 2171.0, P < .001) and anesthesiologists (n = 190, IQR = 16.7%–54.6%, χ² = 3857.5, P < .001).

DISCUSSION

Over our study period, MMA use increased by an average of 3.0% per year. MMA practice patterns for postoperative pain management in this 6-hospital US sample varied notably over this time period, not just by the patient characteristics and medication combinations but also by the anesthesia provider and facility where the surgery was performed. We found that patients who received MMA were more likely to be younger and female, with a lower comorbidity burden. We also found significant variation in MMA use by procedure characteristics, including specialty and operative time. Further, not only did MMA use vary across anesthesia providers, but we found trends of both adoption and deimplementation in our sample of 6 US hospitals. Lastly, our exploration of common MMA protocols revealed substantial variation in the type of protocol used (n = 158 different combinations), further supporting wide variation in adoption.

Our study extends on prior research exploring variations of MMA practice patterns as it is being adopted across the United States. In 2016, Ladha and colleagues also observed significant variation in MMA use in surgical patients (2007–2014, United States) that was not explained by patient and procedure characteristics. In this population, 56% of patients received at least one nonopioid medication in addition to an opioid during the study period.¹⁸ Despite our more restrictive definition of 2 or more nonopioids, we still found similar rates of MMA use. Extending on this study, we were able to look at variation attributable to the facility and anesthesiologist, which we found to explain the largest proportion of variation in MMA use outside of procedure characteristics.

Prior research has shown that variations in practice patterns across health care providers may indicate a lack of awareness and education and, thus, a reluctance to change current practices.^19–21 Addressing these barriers through the promulgation of evidence-based guidelines and provider education may facilitate the effective adoption of MMA. One example of an evidence-based method is the Perioperative Surgical Home (PSH) model. This model emphasizes the importance of preemptive/preventive MMA techniques to optimize pain management outcomes. Implementing evidence-based models such as PSH or ERAS protocols can aid in optimizing MMA strategies and potentially reduce these variations.^22,23

The implementation of MMA also involves a collaborative effort between surgeons and anesthesiologists, and this interplay could potentially explain some of the variations we observed.²⁴ In MMA, anesthesiologists typically take the lead in intraoperative pain management, leveraging their specialized training in pharmacology and pain management. They are responsible for selecting and administering anesthetic agents and analgesics during surgery, determining appropriate dosages, monitoring patient vital signs, and adjusting medications as needed.^25,26 Surgeons, while often deferring to anesthesiologists on specific medication choices, still play an important role in MMA, especially in the postoperative period, and their influence on choices could also lead to more variation in the use of MMA.

We also found wide variation in the combinations of medications used for MMA. In our sample, the most commonly used nonopioid medications were acetaminophen, regional anesthesia, and dexamethasone. While acetaminophen is often a component of many MMA protocols, some studies have suggested that intravenous acetaminophen may not significantly reduce postoperative pain or opioid consumption in certain surgical procedures, such as abdominal surgery.^27,28 Similar to acetaminophen, the evidence for dexamethasone’s effectiveness in reducing postoperative opioid use is also mixed. Some studies have reported that dexamethasone, when administered perioperatively, can lead to reduced postoperative pain scores, decreased opioid consumption, and longer time to first analgesia in adults undergoing abdominal surgery.²⁹ However, others have reported no significant effect on opioid consumption for other types of procedures.^30,31 Dexamethasone may also be used for postoperative nausea and vomiting, so it is possible that some of the dexamethasone administrations were not intended for MMA protocols.³² Future studies should explore the interaction of acetaminophen and dexamethasone on opioid and pain outcomes to determine which patient populations may benefit most from these medications.

This was a retrospective observational study with a cross-sectional study design. While the cross-sectional study design may limit our ability to explore longitudinal outcomes, it is ideal for describing practices at a point in time, which was the goal of this study. Our most prominent limitation is the population studied, which consists mainly of older male Veterans with a greater comorbidity burden at the time of surgery. Further, we are limited to the practices at 6 geographically similar hospitals and only assessed variation in use by the anesthesia provider. Our cohort included 190 anesthesiologists and 728 surgeons, resulting in 4308 unique surgeon/anesthesiologist pairings. This made it impossible to include both anesthesiologist and surgeon as a repeated effect in our models. While these hospitals contain a mix of both academic and nonacademic hospitals, the practice patterns may still not be representative of those in other hospitals nationwide. It is also important to note that we opted not to include adjustments for multiple comparisons despite a large sample size. Thus, less emphasis should be placed on significant p-values and more emphasis on the magnitude of difference.

Lastly, while every effort was made to broadly define and accurately capture all potential nonopioid pain medications used for MMA during the perioperative period, it is still possible that we may have missed some medications or misclassified some medications as MMA when they were not intended to be. As an example, dexamethasone is also commonly used to prevent postoperative nausea and vomiting and may not have been intended as part of an MMA protocol. We are also unable to identify patients with contraindications for certain medications, given our data.

To conclude, despite the demonstrated benefits of MMA,³³ such as reduced opioid use and improved patient outcomes, MMA use varies not just in the combinations of medications used but also by patient, provider, and facility characteristics. This study represents the most recent description of the adoption of MMA into practice in the United States Studying the adoption of new health care practices such as MMA into practice is integral to achieving better health outcomes, optimizing health care delivery, fostering innovation, and addressing the evolving needs of patients and the health care system. Monitoring the adoption of new practices helps ensure that patients receive the most effective and up-to-date care, thereby enhancing the overall quality of health care delivery. Our findings highlight a need for further high-quality research to understand the reasons for these variations, develop evidence-based guidelines on optimal medication combinations, and guide the safe and effective adoption of MMA into routine practice.

DISCLOSURES

Conflicts of Interest: None. Funding: This material is based upon work supported by the Veterans Integrated Services Network 21 Early Career Award Program. This manuscript was handled by: Michael J. Barrington, MBBS, FANZCA, PhD.

Supplementary Material

ane-141-847-s001.docx ^{(20.9KB, docx)}