(AO) One-Year Opioid Use Following Discharge from Post-Acute Care After Hip Fracture

Saturday, August 27, 2022

8:00 AM – 6:00 PM CEST

Location: Hall E South

Publication Number: 657

Presenting Author(s)

Andrew R. Zullo, PharmD, PhD
– Associate Professor, Brown University School of Public Health, Brown University, Providence, Rhode Island, United States

Co-Author(s)

Meghan A. Cupp
– Brown University School of Public Health, Providence, Rhode Island, United States
Francesca Beaudoin, MD, PhD
– Associate Professor, Brown University, Providence, RI, United States
Richa Joshi
– Brown University, Providence, United States
Melissa Riester
– Brown University, United States
Kaley N. Hayes, PhD
– Assistant Professor, Brown University School of Public Health, Brown University, Providence, Rhode Island, United States

Background: Hip fracture is a major traumatic event that often requires pain management with opioid therapy. Following hip fracture hospitalization, older adults commonly receive post-acute care (PAC) in Skilled Nursing Facilities (SNFs), Inpatient Rehabilitation Facilities (IRFs) and Long Term Acute Care Hospitals (LTACHs) to rehabilitate and eventually discontinue opioid use before returning home. Little is known about continued opioid use after PAC and whether setting is related to whether patients become long-term opioid users.

Objectives: To assess whether any and long-term opioid use following discharge from PAC after hip fracture varies by PAC setting across SNFs, IRFs and LTACHs.

Methods: We assessed any opioid dispensing and long-term use of opioids among Medicare beneficiaries aged >66 years with a hip fracture hospitalization followed by PAC in a SNF, IRF or LTACH between 2012 and 2018. Patients were followed from PAC discharge for up to one year or until censoring due to re-fracture, re-entering PAC, Medicare disenrollment, or death. Long-term opioid use was defined as an episode of 90 or more days of continuous use with gaps of no more than 30 days. Fine and Gray proportional subdistribution hazards regression models were used to estimate the risk of any post-PAC opioid dispensing and becoming a long-term opioid user following discharge from each PAC setting. Analyses were adjusted for age, sex, race, year, Combined Comorbidity Index, days in PAC and the competing risk of death.

Results: We included 513,497 patients discharged from PAC (76% SNFs, 24% IRFs, 0.4% LTACHs). SNF patients were older on average, while LTACH patients had more comorbidities and longer PAC stays. Opioids were dispensed to 89,591 (17%) patients following PAC discharge and the unadjusted one-year cumulative incidence was notably higher in SNF patients (21.3%) than IRF (18.8%) or LTACH (18.4%). LTACH patients had a significantly reduced hazard of post-PAC opioid use (Hazard Ratio (HR): 0.794 [95%CL 0.705, 0.894]) compared to IRF. There was no significant difference in long-term opioid use between LTACH and IRF (HR: 1.038 [95%CL 0.828, 1.302]). SNF patients had a significantly increased hazard of post-PAC opioid use (HR: 1.208 [95%CL 1.188, 1.229]) and of long-term opioid use (HR: 1.473 [95%CL 1.414, 1.535]), compared to IRF.

Conclusions: Opioid use is common upon discharge from PAC after hip fracture. Further research is needed to assess the effects of post-PAC opioid use and how management during PAC can be modified to reduce the need for continued opioid use after discharge home.