LAUSR

Question In patients with an acute hospital admission and polypharmacy, does an in-hospital, multifaceted pharmacist intervention reduce readmissions compared with usual care? Methods Design Randomized controlled trial (Odense Pharmacist Trial Investigating Medication Interventions at Sector Transfer [OPTIMIST]). ClinicalTrials.gov NCT03079375. Allocation Concealed.* Blinding Blinded* (patients and pharmacists for medication review). Follow-up period 6 months. Setting 4 hospitals in Denmark. Patients 1499 patients 18 years of age (median age 72 y, 54% women) who had a new acute hospital admission and polypharmacy (5 prescription drugs used daily). Exclusion criteria included aphasia, severe dementia, terminal illness, suicidality, need for isolation precautions, or incarceration. Intervention Extended pharmacist intervention (EPI) including a structured, patient-centered medication review; medical reconciliation on discharge; motivational patient interviews (1 in hospital and 2 by telephone after discharge); and follow-up with primary care physicians, pharmacists, and nursing homes if relevant (n =497); basic pharmacist intervention (BPI) including a single medication review (n =498); or usual care (n =503). All interventions were done by trained clinical pharmacists. Outcomes Primary outcomes were readmission at 30 and 180 days, and a composite of readmission or emergency department visit at 180 days. Secondary outcomes included all-cause mortality. Patient follow-up 98% included in intention-to-treat analysis. Main results The main results are in the Table. Conclusion In hospitalized adults with polypharmacy, a multifaceted pharmacist intervention reduced readmissions at 30 and 180 days compared with usual care. Extended (EPI) or basic (BPI) pharmacist intervention vs usual care (UC) in patients with an acute hospital admission and polypharmacy Outcomes Event rates At 180 d EPI BPI UC RRR (95% CI) NNT (CI) Readmission or emergency department visit 41% 49% 17% (5 to 29) 12 (8 to 41) 47% 49% 4.3% (9 to 16) NS Readmission 40% 49% 19% (7 to 30) 11 (7 to 29) 47% 49% 3.6% (9 to 16) NS At 30 d Readmission 14% 22% 35% (14 to 51) 13 (9 to 32) 20% 22% 9.8% (15 to 29) NS NS = not significant; other abbreviations defined in Glossary. RRR, NNT, and CI calculated from UC event rates and hazard ratios in article. Groups did not differ for all-cause mortality (data not shown). Commentary Effective strategies to improve appropriate use of polypharmacy are lacking, and the OPTIMIST trial found that an EPI was effective. Both the EPI and BPI included medication review, which extends beyond medication reconciliation and includes assessment of patients' medications in the context of their clinical condition. The EPI also included motivational interviewing, which is not commonly done by pharmacists and requires training. The trial has limitations. First, it is not clear which EPI elements are effective. For example, pharmacists called primary care physicians for only 262 of 476 patients analyzed in the EPI group, with calls conducted when a medication change was made in hospital. Second, we don't know what types of suggestions or how many of each type were implemented in each group. Understanding more about which treatments were started, stopped, or changed would help us understand the EPI's mechanism of action. Third, no cost data were reported, yet the EPI is probably cost-effectiveit took a mean 2 hours of pharmacist time per patient and would be offset by the cost of 1 hospital readmission. The results of OPTIMIST are not aligned with a review of interventions to improve appropriate polypharmacy, which showed no difference for hospital admissions with an intense intervention (1). Consistent with the BPI effect in OPTIMIST, systematic reviews of medication review in hospital (2) and the community (3) showed no effect on hospital readmission. Basic medication review is unlikely to affect hospital readmission. More intense interventions, such as the EPI, may be effective, but we need information about which interventions work, in which settings, for which types of patients, and at what cost.