Cannabis Use Disorder Linked to Slightly Shorter Hospital Stays After Hip and Knee Replacement
A 2023 study found that patients with a documented cannabis use disorder (CUD) had slightly shorter hospital stays and were more likely to go directly home after primary total hip or knee replacement. However, the findings show an association—not that cannabis use improves recovery.
The study published in Arthroplasty analyzed 8,740,798 total joint arthroplasty admissions recorded in the U.S. National Inpatient Sample between 2006 and 2015. Researchers compared patients whose hospital records included CUD with those without a documented disorder and used propensity-weighting methods to adjust for differences between the groups.
Patients with CUD stayed in the hospital an average of 3.04 days, compared with 3.24 days among patients without CUD. They also had higher odds of discharge to home and lower odds of discharge to a rehabilitation facility. The study found no meaningful difference in overall inpatient complications or in most individual complications examined.
The CUD group did not, however, have lower hospital charges. Daily charges averaged $22,614 compared with $17,955 for patients without CUD, while total admission charges averaged $58,507 versus $50,924. Charges are not the same as actual costs or out-of-pocket expenses, but the difference suggests that a shorter stay did not translate into lower billed amounts in this analysis.
Importantly, the study examined a diagnosis of cannabis use disorder—not ordinary past or occasional cannabis consumption. CUD refers to a pattern of cannabis use that causes clinically significant problems, such as difficulty cutting back, cravings, or continued use despite harm. The National Institute on Drug Abuse’s cannabis resources provide additional information about cannabis use and substance use disorders.
Because the research was retrospective and based on hospital administrative data, it cannot establish why patients with CUD were discharged sooner or whether cannabis itself played any role. The database could not reliably capture factors such as the amount or timing of cannabis use, use of other substances, social support at home, functional status, or a patient’s readiness to leave. It is therefore not appropriate to interpret the results as evidence that cannabis is a safe or effective postoperative pain treatment.
The data source, the Agency for Healthcare Research and Quality’s National Inpatient Sample, is designed to estimate patterns in U.S. inpatient care but is derived largely from discharge and billing records. Clinicians should use the findings as one piece of information when planning care for patients with CUD, while still screening for substance use, reviewing possible interactions with anesthetic or pain medications, assessing discharge support, and coordinating follow-up when treatment for a substance use disorder is indicated.
