A Correlation of a Medication-Focused Risk Score to Medication Errors at Discharge

Pharmacy transitions-of-care services at the time of hospital discharge are helpful in reducing medication errors. Validated risk tools are commonly used by pharmacists to identify patients at greatest benefit of these services. However, current tools lack assessment of medication-related risk factors and predict hospital readmissions rather than medication errors. To address this, a novel medication-focused risk tool (UCSD-Rx risk score) was created to help classify patients at a higher risk for medication errors. This study was split into 2 phases aimed to internally validate the risk score. Phase I of the study compared the predictability of 30-day unplanned readmissions between the UCSD-Rx risk score and a well-validated risk tool, the LACE+ index. To further specify our risk score for pharmacist use, phase II of the study analyzed the predictability of the risk score to medication errors at discharge. Phase I demonstrated similar classification performance of 30-day unplanned readmissions between the UCSD-Rx risk score (C-statistic, 0.66; 95% confidence interval [CI], 0.64-0.68; P < .0001) and the LACE+ index (C-statistic, 0.69; 95%CI, 0.67-0.71; P < .0001). In phase II, logistic regression showed an increasing UCSD-Rx risk score was predictive of individuals who would experience a medication error at discharge (odds ratio, 1.068; 95%CI, 1.005-1.136; P = .035). Results of this study demonstrate that the UCSD-Rx risk score is a promising tool targeted for pharmacist use to identify patients that may benefit most from transitions-of-care services prior to discharge.