Olson, Catherine2015-11-062015-11-062014-11https://hdl.handle.net/11299/175285University of Minnesota Ph.D. dissertation. November 2014. Major: Health Informatics. Advisor: Bonnie Westra. 1 computer file (PDF); vii, 124 pages.Title: Exploration of the Clinical Utility of High Risk Medication Regimens Background: Unnecessary hospital readmissions are a costly problem for the U.S. health care system. An automated algorithm was developed to target this problem and proven to predict elderly patients at greater risk of re-hospitalization based on their medication regimens. Objective: Create an automated algorithm for predicting elderly patients' medication-related risks for re-hospitalization (study 1), optimize the algorithm by improving the sensitivity of its medication criteria (study 2), and determine its usefulness within different patient populations (study 3). Materials and methods: Outcome and Assessment Information Set (OASIS) and medication data were reused from a previous, manual study of 911 patients from 15 Medicare-certified home health care agencies. Medication data was converted to standardized drug codes using APIs managed by the National Library of Medicine (NLM), and then integrated in an automated algorithm that calculations patients' high risk medication regime scores (HRMRs). A comparison of results between the automated and manual processes was conducted to determine HRMR score match rates (study 1). Odds Ratio analyses, literature reviews and clinical judgments were used to adjust the scoring of patients' High Risk Medication Regimens (HRMRs). Receiver Operating Characteristic (ROC) analysis evaluated whether these adjustments improved the predictive strength of the algorithm (study 2). Unsupervised clustering was used to determine patient population subgroups. HRMR scores were then applied to these subgroups, and ROC & FDR analysis evaluated whether the predictive strength of the algorithm increased for a specific patient population subgroup (study 3). Results: HRMR scores are composed of polypharmacy (number of drugs), potentially inappropriate medications (PIM) (drugs risky to the elderly), and Medication Regimen Complexity Index (MRCI) (complex dose forms, instructions or administration). The automated algorithm produced polypharmacy, PIM and MRCI scores that matched with 99, 87, 99 percent of the scores, respectively, from the manual analysis (study 1). Strongest ROC results for the HRMR components were .68 for polypharmacy when excluding supplements; and .60 for PIM and .69 for MRCI using the original HRMR criteria (study 2). Subgroups consisting of males who have adult children as primary caregivers show stronger AUC curves than the entire population. (study 3). Conclusion: The automated algorithm can predict elderly patients at risk of hospital readmissions and is improved by a modification to its polypharmacy criteria. A hypothesis for future study includes that the algorithm is more predictive in the subgroup of males who have adult children as their caregiver.enInformaticsThesis or Dissertation