Dataset Information

Development and Evaluation of a Clinical Decision Support System to Improve Medication Safety.

ABSTRACT: BACKGROUND:Clinical decision support systems (CDSSs) are a good strategy for preventing medication errors and reducing the incidence and severity of adverse drug events (ADEs). However, these systems are not very effective and are subject to multiple limitations that prevent their implementation in clinical practice. OBJECTIVES:The objective of this study was to evaluate the effectiveness of an advanced CDSS, HIGEA, which generates alerts based on predefined clinical rules to identify patients at risk of an ADE. METHODS:A multidisciplinary team defined the system and the clinical rules focusing on medication errors commonly encountered in clinical practice. Four intervention programs were defined: (1) dose adjustment in renal impairment; (2) adjustment of anticoagulation/antiplatelet therapy; (3) detection of biochemical/hematologic toxicities; and (4) therapeutic drug monitoring. We performed a 6-month observational prospective study to analyze the effectiveness of these clinical rules by calculating the positive predictive value (PPV). RESULTS:The team defined 211 clinical rules. During the study period, HIGEA generated 1,086 alerts (8.9 alerts per working day), which were reviewed by pharmacists. Fifty-one percent (554/1,086) of alerts generated an intervention to prevent a possible ADE; of these, 66% (368/554) required a documented modification to therapy owing to a real prescription error intercepted. The intervention program that induced the highest number of modifications to therapy was the dose adjustment in renal impairment program (PPV?=?0.51), followed by the adjustment of anticoagulation/antiplatelet therapy program (PPV?=?0.24). The percentage of accepted interventions was similar in surgical units (68%), medical units (67%), and critical care units (63%). CONCLUSION:Our study offers evidence that HIGEA is highly effective in preventing potential ADEs at the prescription stage.

SUBMITTER: Ibanez-Garcia S

PROVIDER: S-EPMC6637024 | biostudies-literature | 2019 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development and Evaluation of a Clinical Decision Support System to Improve Medication Safety.

Ibáñez-Garcia Sara S Rodriguez-Gonzalez Carmen C Escudero-Vilaplana Vicente V Martin-Barbero Maria Luisa ML Marzal-Alfaro Belén B De la Rosa-Triviño Jose Luis JL Iglesias-Peinado Irene I Herranz-Alonso Ana A Sanjurjo Saez Maria M

Applied clinical informatics 20190501 3

<h4>Background</h4>Clinical decision support systems (CDSSs) are a good strategy for preventing medication errors and reducing the incidence and severity of adverse drug events (ADEs). However, these systems are not very effective and are subject to multiple limitations that prevent their implementation in clinical practice.<h4>Objectives</h4>The objective of this study was to evaluate the effectiveness of an advanced CDSS, HIGEA, which generates alerts based on predefined clinical rules to iden ...[more]

PMID: 31315138

Similar Datasets

Project description:BackgroundThe quality and safety of prescribing in general practice is important, Clinical decision support (CDS) systems can be used which present alerts to health professionals when prescribing in order to identify patients at risk of potentially hazardous prescribing. It is known that such computerised alerts may improve the safety of prescribing in hospitals but their implementation and sustainable use in general practice is less well understood. We aimed to understand the factors that influenced the successful implementation and sustained use in primary care of a CDS system.MethodsParticipants were purposively recruited from Clinical Commissioning Groups (CCGs) and general practices in the North West and East Midlands regions of England and from the CDS developers. We conducted face-to-face and telephone-based semi-structured qualitative interviews with staff stakeholders. A selection of participants was interviewed longitudinally to explore the further sustainability 1-2 years after implementation of the CDS system. The analysis, informed by Normalisation Process Theory (NPT), was thematic, iterative and conducted alongside data collection.ResultsThirty-nine interviews were conducted either individually or in groups, with 33 stakeholders, including 11 follow-up interviews. Eight themes were interpreted in alignment with the four NPT constructs: Coherence (The purpose of the CDS: Enhancing medication safety and improving cost effectiveness; Relationship of users to the technology; Engagement and communication between different stakeholders); Cognitive Participation (Management of the profile of alerts); Collective Action (Prescribing in general practice, patient and population characteristics and engagement with patients; Knowledge);and Reflexive Monitoring (Sustaining the use of the CDS through maintenance and customisation; Learning and behaviour change. Participants saw that the CDS could have a role in enhancing medication safety and in the quality of care. Engagement through communication and support for local primary care providers and management leaders was considered important for successful implementation. Management of prescribing alert profiles for general practices was a dynamic process evolving over time. At regional management levels, work was required to adapt, and modify the system to optimise its use in practice and fulfil local priorities. Contextual factors, including patient and population characteristics, could impact upon the decision-making processes of prescribers influencing the response to alerts. The CDS could operate as a knowledge base allowing prescribers access to evidence-based information that they otherwise would not have.ConclusionsThis qualitative evaluation utilised NPT to understand the implementation, use and sustainability of a widely deployed CDS system offering prescribing alerts in general practice. The system was understood as having a role in medication safety in providing relevant patient specific information to prescribers in a timely manner. Engagement between stakeholders was considered important for the intervention in ensuring prescribers continued to utilise its functionality. Sustained implementation might be enhanced by careful profile management of the suite of alerts in the system. Our findings suggest that the use and sustainability of the CDS was related to prescribers' perceptions of the relevance of alerts. Shared understanding of the purpose of the CDS between CCGS and general practices particularly in balancing cost saving and safety messages could be beneficial.

Project description:BackgroundComputerized physician order entry (CPOE) systems are incorporated into clinical decision support systems (CDSSs) to reduce medication errors and improve patient safety. Automatic alerts generated from CDSSs can directly assist physicians in making useful clinical decisions and can help shape prescribing behavior. Multiple studies reported that approximately 90%-96% of alerts are overridden by physicians, which raises questions about the effectiveness of CDSSs. There is intense interest in developing sophisticated methods to combat alert fatigue, but there is no consensus on the optimal approaches so far.ObjectiveOur objective was to develop machine learning prediction models to predict physicians' responses in order to reduce alert fatigue from disease medication-related CDSSs.MethodsWe collected data from a disease medication-related CDSS from a university teaching hospital in Taiwan. We considered prescriptions that triggered alerts in the CDSS between August 2018 and May 2019. Machine learning models, such as artificial neural network (ANN), random forest (RF), naïve Bayes (NB), gradient boosting (GB), and support vector machine (SVM), were used to develop prediction models. The data were randomly split into training (80%) and testing (20%) datasets.ResultsA total of 6453 prescriptions were used in our model. The ANN machine learning prediction model demonstrated excellent discrimination (area under the receiver operating characteristic curve [AUROC] 0.94; accuracy 0.85), whereas the RF, NB, GB, and SVM models had AUROCs of 0.93, 0.91, 0.91, and 0.80, respectively. The sensitivity and specificity of the ANN model were 0.87 and 0.83, respectively.ConclusionsIn this study, ANN showed substantially better performance in predicting individual physician responses to an alert from a disease medication-related CDSS, as compared to the other models. To our knowledge, this is the first study to use machine learning models to predict physician responses to alerts; furthermore, it can help to develop sophisticated CDSSs in real-world clinical settings.

Dataset Information

Development and Evaluation of a Clinical Decision Support System to Improve Medication Safety.

Publications

Development and Evaluation of a Clinical Decision Support System to Improve Medication Safety.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets