Pharmacie-rural. Validation Of A Medication Related Readmission Risk Tool

• By: KROGH, Linda (University of Sydney, Australia)
• Co-author(s): Linda Krogh (University of Sydney, Sydney, Australia)
  Dr Shania Liu (University of Sydney, Sydney, Australia)
  Dr Stephen Carter (University of Sydney, Sydney, Australia)
  Dr Rebekah Moles (University of Sydney, Sydney, Australia)
  Dr Jonathan Penm (University of Sydney, Sydney, Australia)
  
• Abstract:

  Title
  Pharmacie-Rural. Validation of a medication related readmission risk tool
  Authors: Linda Krogh, Shania Liu, Stephen Carter, Rebekah Moles, Jonathan Penm and the Tic Toc team (full team to be named on poster)
  Affiliation: Faculty of Medicine and Health
  Background information
  Clinical pharmacists have limited capacity to identify patients at high risk of medication related re-admissions and to target pharmacist interventions towards them. PHarmacie-R is a bedside risk prediction tool developed to identify risk of readmission in a metropolitan population. No tool currently exists to streamline this task for rural clinical pharmacists.
  
  Purpose
  This project aimed to validate a tool initially developed to identify patients at high risk of readmission to a metropolitan hospital to a rural setting.
  Method
  Retrospective data from patients across 19 rural hospitals between June 2022 and May 2023 were collected. This included age, gender, living arrangements, postcode, indigenous status or if an interpreter was required, chronic comorbidities, medications, prior admissions and length of stay. Separate logistic regression models were carried out using these variables and then modified using the Modified Monash Model to scale the rural/remote variable.
  Results
  Of the 1308 patients included in the analysis, 50.4% were male, 49.6% were female. The average age was 74.5 years. Results of the validation indicate that the tool performs poorly in rural areas in regards to predicting both 30 and 90 day readmission. For 30 day readmission C statistic = 0.5581 (95% CI 0.5139-0.6023) and for 90 day readmission C statistic = 0.5384 (95% CI 0.5007-0.5761).
  Conclusion
  Risk of readmission in a rural setting cannot be predicted using the current Pharmacie-R tool. If conducting further research requiring identification of patients at high risk of readmission in rural areas an alternative method should be used. Further study into variables that impact on readmission to rural hospitals is recommended.