Table 4.

Univariable (T-test for comparisons) and multivariable results by liposomal bupivacaine use for the sensitivity analysis in hospitals with >50% liposomal bupivacaine use.

	Univariable results			Multivariable results
	Use of liposomal bupivacaine			Use of liposomal bupivacaine
	Median (IQR)	Median (IQR)	P-value	Percent change (95% CI)	P-value
INPATIENT OPIOID PRESCRIPTION
Total Opioids (mg OME)	170 (110–260)	235 (140–357)	<0.0001	−12.6% (−16.0%; −9.0%)	<0.0001
Day 0 Opioids (mg OME)	95 (55–138)	113 (60–180)	<0.0001	−9.4% (−12.9%; −5.7%)	<0.0001
Day 1 Opioids (mg OME)	45 (20–75)	56 (30–90)	<0.0001	−7.7% (−12.5%; −2.6%)	<0.0001
Day 1+ Opioids (mg OME)	23 (0–60)	45 (15–90)	<0.0001	−9.5% (−15.7%; −2.9%)	<0.0001
RESOURCE UTILIZATION
Cost of Hospitalization (US dollars)	14,356 (12,818–16,572)	14,081 (12,781–16,271)	0.0002	−3.1% (−10.6%; 5.0%)	>0.999
Length of Stay (days)	2 (2–3)	2 (2–3)	<0.0001	−13.1% (−15.9%; −10.3%)	<0.0001

Multivariable models adjusted for: age, volume of annual total knee arthroplasties per hospital, hospital size, Charlson-Deyo Index, sex, race, year, insurance, teaching status, hospital area, general anesthesia use, neuraxial anesthesia use, liposomal bupivacaine, NSAIDs, Cox-2 inhibitors, ketamine, gabapentinoids, patient-controlled analgesia, IV acetaminophen, and the comorbidities for substance use/abuse, pain conditions, and psychiatric conditions.

OME: Oral morphine equivalents

IQR: Interquartile range

CI: Confidence interval

NSAIDs: Nonsteroidal antiinflammatory drugs

Cox-2: Cyclooxygenase-2