Impact of a Geriatric Medication Safety Initiative (GEMSI) Pharmacist-Driven Procedure on Opioid Use in a Transitional Care Unit

<b>Background:</b> Millions of older people are hospitalized each year, and they often use pain medications during their stay. At the same time, about seven million people 65 years of age and older have filled at least one opioid prescription. Studies have found pharmacist-led interventions to minimize inappropriate prescribing and optimize medication regimens. These also may reduce adverse events and length of stay. <b>Objective</b> A pharmacist-driven procedure, the Geriatric Medication Safety Initiative (GEMSI), was created in May 2017. The primary outcome of this study was to examine the opioid morphine milligram equivalent per day (MMED) pre- and post-implementation of GEMSI. Secondary outcomes included length of stay (LOS) and acetaminophen use per day. <b>Design</b> This was a retrospective, single-center cohort study with pre- and post-implementation groups. Data were extracted via chart review to determine the reason for admission, length of stay, non-opioid pain medication usage, and MMED. Data are expressed as mean, standard deviation (SD), and n (%). Chi square, Mann Whitney U, and two sample t-tests were performed as appropriate. <b>Setting</b> A transitional care unit (TCU) in the state of New York that has achieved an Institute for Healthcare Improvement (IHI) for Age-Friendly health level 2 center designation. It is a short-term (usual maximum of 21 days) rehabilitation and acute care setting. Patients who are medically stable after their acute admission but need additional physical and/or occupational therapy are eligible for admission. <b>Patients, Participants</b> Patients greater than or equal to 65 years of age admitted to the TCU in 2016 or 2018 were included. Patients were excluded if they were acutely ill and/or transferred off the TCU. <b>Intervention</b> This was a single-center (ie, a short-term inpatient rehabilitation facility), retrospective, cohort chart review study that was exempted from Institutional Review Board (IRB) review. <b>Results</b> In total, 566 patients were included. The overall mean MMEDs pre- and post-GEMSI were 9.5 (+/- 18.0) compared to 8.5 (+/- 22.7) mg/day, respectively (<i>P</i> = 0.186). The TCU length of stay decreased from a mean of 12.4 to 11.1 days (<i>P</i> = 0.005) and the average acetaminophen use increased from 673 to 722 mg/day from pre- to post-GEMSI (<i>P</i> < 0.001). Lidocaine doses utilized per day were not statistically different between pre- to post-GEMSI (0.14 doses per day vs 0.20 doses per day; <i>P</i> = 0.798). In a subgroup analysis of patients who reported pain, the mean MMEDs were 15.3 (+/- 21.3) compared to 10.9 (+/- 18.7) mg/day, respectively (<i>P</i> = 0.038). <b>Discussion</b> This retrospective study demonstrates that the implementation of a pharmacist-driven procedure, such as GEMSI, is associated with decreased opioid exposure and increased use of non-opioid analgesics such as acetaminophen. By decreasing the exposure to opioid medications, we expect that patients are less likely to experience dose-dependent adverse effects of opioids. <b>Conclusion</b> A pharmacist-driven procedure for multimodal pain management was associated with a decreased MMED and reduced LOS. Future large-scale studies need to be conducted to replicate these results in different practice settings.