A medication error is an inadvertent failure in the drug therapy process that can cause serious harm to patients by increasing morbidity and mortality and are associated with significant economic costs to the healthcare system. Medication reconciliation is the most cost-effective intervention and can result in a 66% reduction in medication errors. To improve patient safety, we developed a machine learning-based tool that prioritizes patients at risk of medication errors upon admission to the hospital to ensure that they undergo medication reconciliation by clinical pharmacists. The data were collected from the electronic health records of patients admitted to Reims University Hospital who underwent medication reconciliation between 2017 and 2023. The data from 7200 patients were used to train four machine learning-based models based on 52 variables in the development dataset. These models were used to prioritize admitted patients according to their likelihood of being exposed to a medication error. Our models, particularly the voting classifier model, demonstrated good performance (a recall of 0.75, precision of 0.65, F1 score of 0.70, AUROC of 0.74 and AUCPR of 0.75). In a retrospective evaluation simulating real-life use, the voting classifier model successfully identified 45% of the total patients selected who were found to have at least one unintended discrepancy compared to 21% when using the existing tool. The positive experimental results of this tool showed a superior improvement of 113% over the existing tool by targeting patients at risk of medication errors upon admission to Reims University Hospital.
13 December 2024 in International Journal of Clinical Pharmacy
We developed a clinical named entity recognition model to predict clinical relevance of pharmacist interventions (PIs) by identifying and labelling expressions from unstructured comments of PIs. Three labels, drug, kidney and dosage, had a great inter-annotator agreement (>60%) and could be used as reference labelization. These labels also showed a high precision (>70%) and a variable recall (50-90 %).
During the dispensing process of medical orders (MOs), community pharmacists (CPs) can manage drug-related problems (DRPs) by performing pharmacist interventions (PIs). There is little evidence that the PI rate is higher with MOs from hospitals (MOHs) than ambulatory (MOAs) settings, and their impact on the patient and community pharmacy is unknown. The primary objective of this study was to compare the MOH and MOA PI rates. The secondary objective was to describe PIs and their clinical and organizational impacts on patient and community pharmacy workflow. A total of 120 CPs participated in a prospective study. Each CP included 10 MOH and 10 MOA between January and June 2020. DRP and PI description and clinical and organizational impacts between MOH and MOA were assessed and compared. We analyzed 2325 MOs. PIs were significantly more frequent in MOH than in MOA (9.7% versus 4.7%; p < 0.001). The most reported PI was the difficulty of contacting hospital prescribers (n = 45; 52.2%). MOHs were associated with a longer dispensing process time and a greater impact on patient pathway and community pharmacy workflow than MOAs. Lack of communication between hospital and primary care settings partly explains the results. Implementation of clinical pharmacy activities at patient discharge could alleviate these impacts.
Purpose: The high-dose methotrexate (HD-MTX) regimen (>1 g/m 2 ) is the standard of care for preventing and managing central nervous system (CNS) invasion in diffuse large B-cell lymphoma (DLBCL). Delayed elimination of HD-MTX can lead to severe toxicity. Albumin is a suggested biological marker to predict delayed elimination. This study aims to evaluate whether serum albumin can be used as a predictive factor. Methods: This retrospective bicentric study includes DLBCL patients, receiving HD-MTX from 06.30.2016 to 12.31.2023 using CHIMI0v5.9. Data collected included demographic, biological, clinical and therapeutic. Eligible patients were over 18 years, with DLBCL diagnosis, available serum albumin before HD-MTX, with therapeutic drug monitoring of plasma methotrexate available. Univariate and multivariate analysis compared patients with and without delayed elimination. Results: A total of 68 patients were included, with 15 suffering delayed elimination. Univariate analysis showed that male and Body Surface Area ≥ 2 m² (BSA) patients were more likely to have delayed clearance (respectively p = 0.01; p = 0.05). The association with male sex was confirmed in multivariate analysis (p = 0.01). Most cases of hypoalbuminemia were classified as grade 1 or 2. Patients with hypoalbuminemia received lower doses of MTX. Serum albumin level was not significantly associated with delayed elimination (p = 0.8). Conclusions: Overall, mainly biometric factors are associated with delayed elimination, highlighting the problem of prescribing habits based on body surface area. Albuminemia's role remains debated, and this study does not provide a conclusion, as patients with hypoalbuminemia received lower HD-MTX doses. Studies should account factors related to inflammation, nutrition and specific biomarkers (prealbumin).