Beyond the override: Using evidence of previous drug tolerance to suppress drug allergy alerts; a retrospective study of opioid alerts.

OBJECTIVE: Despite the extensive literature exploring alert fatigue, most studies have focused on describing the phenomenon, but not on fixing it. The authors aimed to identify data useful to avert clinically irrelevant alerts to inform future research on clinical decision support (CDS) design. METHODS: We conducted a retrospective observational study of opioid drug allergy alert (DAA) overrides for the calendar year of 2019 at a large academic medical center, to identify data elements useful to find irrelevant alerts to be averted. RESULTS: Overall, 227,815 DAAs were fired in 2019, with an override rate of 91 % (n = 208196). Opioids represented nearly two-thirds of these overrides (n = 129063; 62 %) and were the drug class with the highest override rate (96 %). On average, 29 opioid DAAs were overridden per patient. While most opioid alerts (97.1 %) are fired for a possible match (the drug class of the allergen matches the drug class of the prescribed drug), they are overridden significantly less frequently for definite match (exact match between allergen and prescribed drug) (88 % vs. 95.9 %, p < 0.001). When comparing the triggering drug with previously administered drugs, override rates were equally high for both definite match (95.9 %), no match (95.5 %), and possible match (95.1 %). Likewise, when comparing to home medications, overrides were excessively high for possible match (96.3 %), no match (96 %), and definite match (94.4 %). CONCLUSION: We estimate that 74.5% of opioid DAAs (46.4% of all DAAs) at our institution could be relatively safely averted, since they either have a definite match for previous inpatient administrations suggesting drug tolerance or are fired as possible match with low risk of cross-sensitivity. Future research should focus on identifying other relevant data elements ideally with automated methods and use of emerging standards to empower CDS systems to suppress false-positive alerts while avoiding safety hazards.

Meaningful time-related aspects of alerts in Clinical Decision Support Systems. A unified framework.

Alerts are a common functionality of clinical decision support systems (CDSSs). Although they have proven to be useful in clinical practice, the alert burden can lead to alert fatigue and significantly reduce their usability and acceptance. Based on a literature review, we propose a unified framework consisting of a set of meaningful timestamps that allows the use of state-of-the-art measures for alert burden, such as alert dwell time, alert think time, and response time. In addition, it can be used to investigate other measures that could be relevant as regards dealing with this problem. Furthermore, we provide a case study concerning three different types of alerts to which the framework was successfully applied. We consider that our framework can easily be adapted to other CDSSs and that it could be useful for dealing with alert burden measurement thus contributing to its appropriate management.

Machine learning clinical prediction models for acute kidney injury: the impact of baseline creatinine on prediction efficacy.

BACKGROUND: There are many Machine Learning (ML) models which predict acute kidney injury (AKI) for hospitalised patients. While a primary goal of these models is to support clinical decision-making, the adoption of inconsistent methods of estimating baseline serum creatinine (sCr) may result in a poor understanding of these models' effectiveness in clinical practice. Until now, the performance of such models with different baselines has not been compared on a single dataset. Additionally, AKI prediction models are known to have a high rate of false positive (FP) events regardless of baseline methods. This warrants further exploration of FP events to provide insight into potential underlying reasons. OBJECTIVE: The first aim of this study was to assess the variance in performance of ML models using three methods of baseline sCr on a retrospective dataset. The second aim was to conduct an error analysis to gain insight into the underlying factors contributing to FP events. MATERIALS AND METHODS: The Intensive Care Unit (ICU) patients of the Medical Information Mart for Intensive Care (MIMIC)-IV dataset was used with the KDIGO (Kidney Disease Improving Global Outcome) definition to identify AKI episodes. Three different methods of estimating baseline sCr were defined as (1) the minimum sCr, (2) the Modification of Diet in Renal Disease (MDRD) equation and the minimum sCr and (3) the MDRD equation and the mean of preadmission sCr. For the first aim of this study, a suite of ML models was developed for each baseline and the performance of the models was assessed. An analysis of variance was performed to assess the significant difference between eXtreme Gradient Boosting (XGB) models across all baselines. To address the second aim, Explainable AI (XAI) methods were used to analyse the XGB errors with Baseline 3. RESULTS: Regarding the first aim, we observed variances in discriminative metrics and calibration errors of ML models when different baseline methods were adopted. Using Baseline 1 resulted in a 14% reduction in the f1 score for both Baseline 2 and Baseline 3. There was no significant difference observed in the results between Baseline 2 and Baseline 3. For the second aim, the FP cohort was analysed using the XAI methods which led to relabelling data with the mean of sCr in 180 to 0 days pre-ICU as the preferred sCr baseline method. The XGB model using this relabelled data achieved an AUC of 0.85, recall of 0.63, precision of 0.54 and f1 score of 0.58. The cohort size was 31,586 admissions, of which 5,473 (17.32%) had AKI. CONCLUSION: In the absence of a widely accepted method of baseline sCr, AKI prediction studies need to consider the impact of different baseline methods on the effectiveness of ML models and their potential implications in real-world implementations. The utilisation of XAI methods can be effective in providing insight into the occurrence of prediction errors. This can potentially augment the success rate of ML implementation in routine care.

SELID: Selective Event Labeling for Intrusion Detection Datasets.

A large volume of security events, generally collected by distributed monitoring sensors, overwhelms human analysts at security operations centers and raises an alert fatigue problem. Machine learning is expected to mitigate this problem by automatically distinguishing between true alerts, or attacks, and falsely reported ones. Machine learning models should first be trained on datasets having correct labels, but the labeling process itself requires considerable human resources. In this paper, we present a new selective sampling scheme for efficient data labeling via unsupervised clustering. The new scheme transforms the byte sequence of an event into a fixed-size vector through content-defined chunking and feature hashing. Then, a clustering algorithm is applied to the vectors, and only a few samples from each cluster are selected for manual labeling. The experimental results demonstrate that the new scheme can select only 2% of the data for labeling without degrading the F1-score of the machine learning model. Two datasets, a private dataset from a real security operations center and a public dataset from the Internet for experimental reproducibility, are used.

Alert-Grouping: Smart Personalization of Monitoring System Thresholds to Help Healthcare Teams Struggle with Alarm Fatigue in Intensive Care.

In Intensive Care Units (ICUs), patients are monitored using various devices that generate alerts when specific metrics, such as heart rate and oxygen saturation, exceed predetermined thresholds. However, these alerts can be inaccurate and lead to alert fatigue, resulting in errors and inaccurate diagnoses. We propose Alert grouping, a "Smart Personalization of Monitoring System Thresholds to Help Healthcare Teams Struggle Alarm Fatigue in Intensive Care" model. The alert grouping looks at patients at the individual and cluster levels, and healthcare-related constraints to assist medical and nursing teams in setting personalized alert thresholds of vital parameters. By simulating the function of ICU patient bed devices, we demonstrate that the proposed alert grouping model effectively reduces the number of alarms overall, improving the alert system's validity and reducing alarm fatigue. Implementing this personalized alert model in ICUs boosts medical and nursing teams' confidence in the alert system, leading to better care for ICU patients by significantly reducing alarm fatigue, thereby improving the quality of care for ICU patients.

Transitioning Cognitive Aids into Decision Support Platforms: Requirements and Design Guidelines.

Digital cognitive aids have the potential to serve as clinical decision support platforms, triggering alerts about process delays and recommending interventions. In this mixed-methods study, we examined how a digital checklist for pediatric trauma resuscitation could trigger decision support alerts and recommendations. We identified two criteria that cognitive aids must satisfy to support these alerts: (1) context information must be entered in a timely, accurate, and standardized manner, and (2) task status must be accurately documented. Using co-design sessions and near-live simulations, we created two checklist features to satisfy these criteria: a form for entering the pre-hospital information and a progress slider for documenting the progression of a multi-step task. We evaluated these two features in the wild, contributing guidelines for designing these features on cognitive aids to support alerts and recommendations in time- and safety-critical scenarios.

Habit and Automaticity in Medical Alert Override: Cohort Study.

BACKGROUND: Prior literature suggests that alert dismissal could be linked to physicians' habits and automaticity. The evidence for this perspective has been mainly observational data. This study uses log data from an electronic medical records system to empirically validate this perspective. OBJECTIVE: We seek to quantify the association between habit and alert dismissal in physicians. METHODS: We conducted a retrospective analysis using the log data comprising 66,049 alerts generated from hospitalized patients in a hospital from March 2017 to December 2018. We analyzed 1152 physicians exposed to a specific clinical support alert triggered in a hospital's electronic medical record system to estimate the extent to which the physicians' habit strength, which had been developed from habitual learning, impacted their propensity toward alert dismissal. We further examined the association between a physician's habit strength and their subsequent incidences of alert dismissal. Additionally, we recorded the time taken by the physician to respond to the alert and collected data on other clinical and environmental factors related to the alerts as covariates for the analysis. RESULTS: We found that a physician's prior dismissal of alerts leads to their increased habit strength to dismiss alerts. Furthermore, a physician's habit strength to dismiss alerts was found to be positively associated with incidences of subsequent alert dismissals after their initial alert dismissal. Alert dismissal due to habitual learning was also found to be pervasive across all physician ranks, from junior interns to senior attending specialists. Further, the dismissal of alerts had been observed to typically occur after a very short processing time. Our study found that 72.5% of alerts were dismissed in under 3 seconds after the alert appeared, and 13.2% of all alerts were dismissed in under 1 second after the alert appeared. We found empirical support that habitual dismissal is one of the key factors associated with alert dismissal. We also found that habitual dismissal of alerts is self-reinforcing, which suggests significant challenges in disrupting or changing alert dismissal habits once they are formed. CONCLUSIONS: Habitual tendencies are associated with the dismissal of alerts. This relationship is pervasive across all levels of physician rank and experience, and the effect is self-reinforcing.

Overall performance of a drug-drug interaction clinical decision support system: quantitative evaluation and end-user survey.

BACKGROUND: Clinical decision support systems are implemented in many hospitals to prevent medication errors and associated harm. They are however associated with a high burden of false positive alerts and alert fatigue. The aim of this study was to evaluate a drug-drug interaction (DDI) clinical decision support system in terms of its performance, uptake and user satisfaction and to identify barriers and opportunities for improvement. METHODS: A quantitative evaluation and end-user survey were performed in a large teaching hospital. First, very severe DDI alerts generated between 2019 and 2021 were evaluated retrospectively. Data collection comprised alert burden, override rates, the number of alert overrides reviewed by pharmacists and the resulting pharmacist recommendations as well as their acceptance rate. Second, an e-survey was carried out among prescribers to assess satisfaction, usefulness and relevance of DDI alerts as well as reasons for overriding. RESULTS: A total of 38,409 very severe DDI alerts were generated, of which 88.2% were overridden by the prescriber. In 3.2% of reviewed overrides, a recommendation by the pharmacist was provided, of which 79.2% was accepted. False positive alerts were caused by a too broad screening interval and lack of incorporation of patient-specific characteristics, such as QTc values. Co-prescribing of a non-vitamin K oral anticoagulant and a low molecular weight heparin accounted for 49.8% of alerts, of which 92.2% were overridden. In 88 (1.1%) of these overridden alerts, concurrent therapy was still present. Despite the high override rate, the e-survey revealed that the DDI clinical decision support system was found useful by prescribers. CONCLUSIONS: Identified barriers were the lack of DDI-specific screening intervals and inclusion of patient-specific characteristics, both leading to a high number of false positive alerts and risk for alert fatigue. Despite these barriers, the added value of the DDI clinical decision support system was recognized by prescribers. Hence, integration of DDI-specific screening intervals and patient-specific characteristics is warranted to improve the performance of the DDI software.

Creating Meaningful Alerts and Reducing Alert Fatigue: Strategies Implemented by Informatics Pharmacists to Optimize Dose Range Checking Alerts in a Multihospital Health System.

Background: Among the many clinical decision support (CDS) mechanisms available in electronic health record (EHR) systems, dose range checking (DRC) is one of the most impactful safeguard tools integrated within most computerized provider order entry (CPOE) workflows. Unfortunately, improper configurations and lack of resources to maintain and monitor CDS systems can hinder and even disrupt daily clinical operations. Objective: This article seeks to highlight the impact that informatics pharmacists can make by implementing different strategies to decrease nuisance alerts and create clinically meaningful DRC alerts that guide clinicians in their practice. Methods: Following the activation of the DRC application for 3623 medication groupers (ie, generic drugs and all their dosage form variations), informatics pharmacists implemented strategies to monitor DRC alert output and decrease the number of inappropriate alerts. Such strategies included weekly monitoring of alerts, modification of order sentences (including dose, route, and age/weight filters), update to the rule triggering the alerts, and modifications of the preference settings. Results: From July to September 2018, an average of 70 DRC tables were reviewed by informatics pharmacists, reducing the number of overridden DRC alerts to 4796 in the first week of September-a 63% decrease in a 3-month period. Conclusions: By reducing the number of DRC nuisance alerts and improving the clinical content of DRC alerts, informatics pharmacists can contribute to lowering alert fatigue and improving providers' trust in CDS alerts.

Adverse Drug Events Caused by Drugs Contraindicated for Coadministration Reported in the Japanese Adverse Drug Event Report Database and Recognized by Reporters.

The "INTERACTIONS" section of package inserts aims to provide alert-type warnings in clinical practice; however, these also include many drug-drug interactions that occur rarely. Moreover, considering that drug-drug interaction alert systems were created based on package inserts, repeated alerts can lead to alert fatigue. Although investigations have been conducted to determine prescriptions that induce drug-drug interactions, no studies have focused explicitly on the adverse events induced by drug-drug interactions. We, therefore, sought to investigate the true occurrence of adverse events caused by drug pair contraindications for coadministration in routine clinical practice. Toward this, we created a list of drug combinations that were designated as "contraindications for coadministration" and extracted the cases of adverse drug events from the Japanese Adverse Drug Event Report database that occurred due to combined drug usage. We then calculated the reporters' recognition rate of the drug-drug interactions. Out of the 2121 investigated drug pairs, drug-drug interactions were reported in 43 pairs, 23 of which included an injected drug and many included catecholamines. Warfarin potassium and miconazole (19 reports), azathioprine and febuxostat (11 reports), and warfarin potassium and iguratimod (six reports) were among the 20 most-commonly reported oral medication pairs that were contraindicated for coadministration, for which recognition rates of drug-drug interactions were high. Although these results indicate that only a few drug pair contraindications for coadministration were associated with adverse drug events (43 pairs out of 2121 pairs), it remains necessary to translate these findings into clinical practice.

Applying A/B Testing to Clinical Decision Support: Rapid Randomized Controlled Trials.

BACKGROUND: Clinical decision support (CDS) is a valuable feature of electronic health records (EHRs) designed to improve quality and safety. However, due to the complexities of system design and inconsistent results, CDS tools may inadvertently increase alert fatigue and contribute to physician burnout. A/B testing, or rapid-cycle randomized tests, is a useful method that can be applied to the EHR in order to rapidly understand and iteratively improve design choices embedded within CDS tools. OBJECTIVE: This paper describes how rapid randomized controlled trials (RCTs) embedded within EHRs can be used to quickly ascertain the superiority of potential CDS design changes to improve their usability, reduce alert fatigue, and promote quality of care. METHODS: A multistep process combining tools from user-centered design, A/B testing, and implementation science was used to understand, ideate, prototype, test, analyze, and improve each candidate CDS. CDS engagement metrics (alert views, acceptance rates) were used to evaluate which CDS version is superior. RESULTS: To demonstrate the impact of the process, 2 experiments are highlighted. First, after multiple rounds of usability testing, a revised CDS influenza alert was tested against usual care CDS in a rapid (~6 weeks) RCT. The new alert text resulted in minimal impact on reducing firings per patients per day, but this failure triggered another round of review that identified key technical improvements (ie, removal of dismissal button and firings in procedural areas) that led to a dramatic decrease in firings per patient per day (23.1 to 7.3). In the second experiment, the process was used to test 3 versions (financial, quality, regulatory) of text supporting tobacco cessation alerts as well as 3 supporting images. Based on 3 rounds of RCTs, there was no significant difference in acceptance rates based on the framing of the messages or addition of images. CONCLUSIONS: These experiments support the potential for this new process to rapidly develop, deploy, and rigorously evaluate CDS within an EHR. We also identified important considerations in applying these methods. This approach may be an important tool for improving the impact of and experience with CDS. TRIAL REGISTRATION: Flu alert trial: ClinicalTrials.gov NCT03415425; https://clinicaltrials.gov/ct2/show/NCT03415425. Tobacco alert trial: ClinicalTrials.gov NCT03714191; https://clinicaltrials.gov/ct2/show/NCT03714191.

Barriers and facilitators to the adoption of electronic clinical decision support systems: a qualitative interview study with UK general practitioners.

BACKGROUND: Well-established electronic data capture in UK general practice means that algorithms, developed on patient data, can be used for automated clinical decision support systems (CDSSs). These can predict patient risk, help with prescribing safety, improve diagnosis and prompt clinicians to record extra data. However, there is persistent evidence of low uptake of CDSSs in the clinic. We interviewed UK General Practitioners (GPs) to understand what features of CDSSs, and the contexts of their use, facilitate or present barriers to their use. METHODS: We interviewed 11 practicing GPs in London and South England using a semi-structured interview schedule and discussed a hypothetical CDSS that could detect early signs of dementia. We applied thematic analysis to the anonymised interview transcripts. RESULTS: We identified three overarching themes: trust in individual CDSSs; usability of individual CDSSs; and usability of CDSSs in the broader practice context, to which nine subthemes contributed. Trust was affected by CDSS provenance, perceived threat to autonomy and clear management guidance. Usability was influenced by sensitivity to the patient context, CDSS flexibility, ease of control, and non-intrusiveness. CDSSs were more likely to be used by GPs if they did not contribute to alert proliferation and subsequent fatigue, or if GPs were provided with training in their use. CONCLUSIONS: Building on these findings we make a number of recommendations for CDSS developers to consider when bringing a new CDSS into GP patient records systems. These include co-producing CDSS with GPs to improve fit within clinic workflow and wider practice systems, ensuring a high level of accuracy and a clear clinical pathway, and providing CDSS training for practice staff. These recommendations may reduce the proliferation of unhelpful alerts that can result in important decision-support being ignored.

Impact of Electronic Medical Record Alerts on Emergency Physician Workflow and Medical Management.

BACKGROUND: Electronic medical record (EMR) alerts are automated messages that notify the physician of important information. However, little is known about how EMR alerts affect the workflow and decision-making of emergency physicians (EPs). STUDY OBJECTIVES: This study aimed to quantify the number of EMR alerts EPs receive, the time required to resolve alerts, the types of alerts EPs receive, and the impact of alerts on patient management. METHODS: We performed a prospective observational study at a tertiary care ED with 130,000 visits annually. Research assistants observed EPs on shift from May to December 2018. They recorded the number of EMR alerts received, time spent addressing the alerts, the types of alerts received, and queried the EP to determine if the alert impacted patient management. RESULTS: Seven residents and six attending physicians were observed on a total of 17 shifts and 153 patient encounters; 78% (119) of patient encounters involved alerts. These 119 patients triggered 530 EMR alerts. EPs spent a mean of 7.06 s addressing each alert and addressed 3.46 alerts per total patient seen. In total, EPs spent approximately 24 s per patient resolving alerts. Only 12 alerts (2.26%) changed clinical management. CONCLUSION: EPs frequently receive EMR alerts, however, most alerts were not perceived to impact patient care. These alerts contribute to the high volume of interruptions EPs must contend with in the clinical environment of the ED.

Reducing Alert Fatigue by Sharing Low-Level Alerts With Patients and Enhancing Collaborative Decision Making Using Blockchain Technology: Scoping Review and Proposed Framework (MedAlert).

BACKGROUND: Clinical decision support (CDS) is a tool that helps clinicians in decision making by generating clinical alerts to supplement their previous knowledge and experience. However, CDS generates a high volume of irrelevant alerts, resulting in alert fatigue among clinicians. Alert fatigue is the mental state of alerts consuming too much time and mental energy, which often results in relevant alerts being overridden unjustifiably, along with clinically irrelevant ones. Consequently, clinicians become less responsive to important alerts, which opens the door to medication errors. OBJECTIVE: This study aims to explore how a blockchain-based solution can reduce alert fatigue through collaborative alert sharing in the health sector, thus improving overall health care quality for both patients and clinicians. METHODS: We have designed a 4-step approach to answer this research question. First, we identified five potential challenges based on the published literature through a scoping review. Second, a framework is designed to reduce alert fatigue by addressing the identified challenges with different digital components. Third, an evaluation is made by comparing MedAlert with other proposed solutions. Finally, the limitations and future work are also discussed. RESULTS: Of the 341 academic papers collected, 8 were selected and analyzed. MedAlert securely distributes low-level (nonlife-threatening) clinical alerts to patients, enabling a collaborative clinical decision. Among the solutions in our framework, Hyperledger (private permissioned blockchain) and BankID (federated digital identity management) have been selected to overcome challenges such as data integrity, user identity, and privacy issues. CONCLUSIONS: MedAlert can reduce alert fatigue by attracting the attention of patients and clinicians, instead of solely reducing the total number of alerts. MedAlert offers other advantages, such as ensuring a higher degree of patient privacy and faster transaction times compared with other frameworks. This framework may not be suitable for elderly patients who are not technology savvy or in-patients. Future work in validating this framework based on real health care scenarios is needed to provide the performance evaluations of MedAlert and thus gain support for the better development of this idea.

Development, Implementation, and Evaluation of a Personalized Machine Learning Algorithm for Clinical Decision Support: Case Study With Shingles Vaccination.

BACKGROUND: Although clinical decision support (CDS) alerts are effective reminders of best practices, their effectiveness is blunted by clinicians who fail to respond to an overabundance of inappropriate alerts. An electronic health record (EHR)-integrated machine learning (ML) algorithm is a potentially powerful tool to increase the signal-to-noise ratio of CDS alerts and positively impact the clinician's interaction with these alerts in general. OBJECTIVE: This study aimed to describe the development and implementation of an ML-based signal-to-noise optimization system (SmartCDS) to increase the signal of alerts by decreasing the volume of low-value herpes zoster (shingles) vaccination alerts. METHODS: We built and deployed SmartCDS, which builds personalized user activity profiles to suppress shingles vaccination alerts unlikely to yield a clinician's interaction. We extracted all records of shingles alerts from January 2017 to March 2019 from our EHR system, including 327,737 encounters, 780 providers, and 144,438 patients. RESULTS: During the 6 weeks of pilot deployment, the SmartCDS system suppressed an average of 43.67% (15,425/35,315) potential shingles alerts (appointments) and maintained stable counts of weekly shingles vaccination orders (326.3 with system active vs 331.3 in the control group; P=.38) and weekly user-alert interactions (1118.3 with system active vs 1166.3 in the control group; P=.20). CONCLUSIONS: All key statistics remained stable while the system was turned on. Although the results are promising, the characteristics of the system can be subject to future data shifts, which require automated logging and monitoring. We demonstrated that an automated, ML-based method and data architecture to suppress alerts are feasible without detriment to overall order rates. This work is the first alert suppression ML-based model deployed in practice and serves as foundational work in encounter-level customization of alert display to maximize effectiveness.

A Late Attempt to Involve End Users in the Design of Medication-Related Alerts: Survey Study.

BACKGROUND: When users of electronic medical records (EMRs) are presented with large numbers of irrelevant computerized alerts, they experience alert fatigue, begin to ignore alert information, and override alerts without processing or heeding alert recommendations. Anecdotally, doctors at our study site were dissatisfied with the medication-related alerts being generated, both in terms of volume being experienced and clinical relevance. OBJECTIVE: This study aimed to involve end users in the redesign of medication-related alerts in a hospital EMR, 4 years post implementation. METHODS: This work was undertaken at a private not-for-profit teaching hospital in Sydney, Australia. Since EMR implementation in 2015, the organization elected to implement all medication-related alert types available in the system for prescribers: allergy and intolerance alerts, therapeutic duplication alerts, pregnancy alerts, and drug-drug interaction alerts. The EMR included no medication administration alerts for nurses. To obtain feedback on current alerts and suggestions for redesign, a Web-based survey was distributed to all doctors and nurses at the site via hospital mailing lists. RESULTS: Despite a general dissatisfaction with alerts, very few end users completed the survey. In total, only 3.37% (36/1066) of doctors and 14.5% (60/411) of nurses took part. Approximately 90% (30/33) of doctors who responded held the view that too many alerts were triggered in the EMR. Doctors suggested that most alerts be removed and that alerts be more specific and less sensitive. In contrast, 97% (58/60) of the nurse respondents indicated that they would like to receive medication administration alerts in the EMR. Most nurses indicated that they would like to receive all the alert types available at all severity levels. CONCLUSIONS: Attempting to engage with end users several years post implementation was challenging. Involving users so late in the implementation process may lead to clinicians viewing the provision of feedback to be futile. Seeking user feedback on usefulness, volume, and design of alerts is extremely valuable; however, we suggest this is undertaken early, preferably before system implementation.

The Effect of Patient-Specific Drug-Drug Interaction Alerting on the Frequency of Alerts: A Pilot Study.

Background: False-positive drug-drug interaction alerts are frequent and result in alert fatigue that can result in prescribers bypassing important alerts. Development of a method to present patient-appropriate alerts is needed to help restore alert relevance. Objective: The purpose of this study was to assess the potential for patient-specific drug-drug interaction (DDI) alerts to reduce alert burden. Methods: This project was conducted at a tertiary care medical center. Seven of the most frequently encountered DDI alerts were chosen for developing patient-specific, algorithm-based DDI alerts. For each of the DDI pairs, 2 algorithms featuring different values for modifying factors were made. DDI alerts from the 7 drug pairs were collected over 30 days. Outcome measures included the number of DDI alerts generated before and after patient-specific algorithm application to the same patients over the same time period. Results: A total of 14 algorithms were generated, and each was evaluated by comparing the number of alerts generated by our existing, customized clinical decision support (CDS) software and the patient-specific algorithms. The CDS DDI alerting software generated an average of 185.3 alerts per drug pair over the 30-day study period. Patient-specific algorithms reduced the number of alerts resulting from the algorithms by 11.3% to 93.5%. Conclusion and Relevance: Patient-specific DDI alerting is an innovative and effective approach to reduce the number of DDI alerts, may potentially increase the appropriateness of alerts, and may decrease the potential for alert fatigue.

Artificial Intelligence Technologies for Coping with Alarm Fatigue in Hospital Environments Because of Sensory Overload: Algorithm Development and Validation.

BACKGROUND: Informed estimates claim that 80% to 99% of alarms set off in hospital units are false or clinically insignificant, representing a cacophony of sounds that do not present a real danger to patients. These false alarms can lead to an alert overload that causes a health care provider to miss important events that could be harmful or even life-threatening. As health care units become more dependent on monitoring devices for patient care purposes, the alarm fatigue issue has to be addressed as a major concern for the health care team as well as to enhance patient safety. OBJECTIVE: The main goal of this paper was to propose a feasible solution for the alarm fatigue problem by using an automatic reasoning mechanism to decide how to notify members of the health care team. The aim was to reduce the number of notifications sent by determining whether or not to group a set of alarms that occur over a short period of time to deliver them together, without compromising patient safety. METHODS: This paper describes: (1) a model for supporting reasoning algorithms that decide how to notify caregivers to avoid alarm fatigue; (2) an architecture for health systems that support patient monitoring and notification capabilities; and (3) a reasoning algorithm that specifies how to notify caregivers by deciding whether to aggregate a group of alarms to avoid alarm fatigue. RESULTS: Experiments were used to demonstrate that providing a reasoning system can reduce the notifications received by the caregivers by up to 99.3% (582/586) of the total alarms generated. Our experiments were evaluated through the use of a dataset comprising patient monitoring data and vital signs recorded during 32 surgical cases where patients underwent anesthesia at the Royal Adelaide Hospital. We present the results of our algorithm by using graphs we generated using the R language, where we show whether the algorithm decided to deliver an alarm immediately or after a delay. CONCLUSIONS: The experimental results strongly suggest that this reasoning algorithm is a useful strategy for avoiding alarm fatigue. Although we evaluated our algorithm in an experimental environment, we tried to reproduce the context of a clinical environment by using real-world patient data. Our future work is to reproduce the evaluation study based on more realistic clinical conditions by increasing the number of patients, monitoring parameters, and types of alarm.

The effect of ICU-tailored drug-drug interaction alerts on medication prescribing and monitoring: protocol for a cluster randomized stepped-wedge trial.

BACKGROUND: Drug-drug interactions (DDIs) can cause patient harm. Between 46 and 90% of patients admitted to the Intensive Care Unit (ICU) are exposed to potential DDIs (pDDIs). This rate is twice as high as patients on general wards. Clinical decision support systems (CDSSs) have shown their potential to prevent pDDIs. However, the literature shows that there is considerable room for improvement of CDSSs, in particular by increasing the clinical relevance of the pDDI alerts they generate and thereby reducing alert fatigue. However, consensus on which pDDIs are clinically relevant in the ICU setting is lacking. The primary aim of this study is to evaluate the effect of alerts based on only clinically relevant interactions for the ICU setting on the prevention of pDDIs among Dutch ICUs. METHODS: To define the clinically relevant pDDIs, we will follow a rigorous two-step Delphi procedure in which a national expert panel will assess which pDDIs are perceived clinically relevant for the Dutch ICU setting. The intervention is the CDSS that generates alerts based on the clinically relevant pDDIs. The intervention will be evaluated in a stepped-wedge trial. A total of 12 Dutch adult ICUs using the same patient data management system, in which the CDSS will operate, were invited to participate in the trial. Of the 12 ICUs, 9 agreed to participate and will be enrolled in the trial. Our primary outcome measure is the incidence of clinically relevant pDDIs per 1000 medication administrations. DISCUSSION: This study will identify pDDIs relevant for the ICU setting. It will also enhance our understanding of the effectiveness of alerts confined to clinically relevant pDDIs. Both of these contributions can facilitate the successful implementation of CDSSs in the ICU and in other domains as well. TRIAL REGISTRATION: Nederlands Trial register Identifier: NL6762 . Registered November 26, 2018.

Smart pumps improve medication safety but increase alert burden in neonatal care.

BACKGROUND: Smart pumps have been widely adopted but there is limited evidence to understand and support their use in pediatric populations. Our objective was to assess whether smart pumps are effective at reducing medication errors in the neonatal population and determine whether they are a source of alert burden and alert fatigue in an intensive care environment. METHODS: Using smart pump records, over 370,000 infusion starts for continuously infused medications used in neonates and infants hospitalized in a level IV NICU from 2014 to 2016 were evaluated. Attempts to exceed preset soft and hard maximum limits, percent variance from those limits, and pump alert frequency, patterns and salience were evaluated. RESULTS: Smart pumps prevented 160 attempts to exceed the hard maximum limit for doses that were as high as 7-29 times the maximum dose and resulted in the reprogramming or cancellation of 2093 infusions after soft maximum alerts. While the overall alert burden from smart pumps for continuous infusions was not high, alerts clustered around specific patients and medications, and a small portion (17%) of infusions generated the majority of alerts. Soft maximum alerts were often overridden (79%), consistent with low alert salience. CONCLUSIONS: Smart pumps have the ability to improve neonatal medication safety when compliance with dose error reducing software is high. Numerous attempts to administer high doses were intercepted by dosing alerts. Clustered alerts may generate a high alert burden and limit safety benefit by desensitizing providers to alerts. Future efforts should address ways to improve alert salience.