Assessment of TB treatment adherence using computer-assisted self-interviewing.

OBJECTIVES: To evaluate the correlation and agreement between computer-assisted self-interviewing (CASI) and home-based unannounced pill counts (HUPC) for assessing anti-TB medication adherence (MA) and to examine the relationship between MA and treatment success.METHODS: Individual CASI-evaluated MA was compared three times with HUPC MA over the treatment course. The relationship between the two methodologies was determined using correlation coefficients (r) and intraclass correlation coefficients (ICC). The association between MA and efficacy was evaluated using odds ratios (ORs).RESULTS: According to CASI assessments, MA rates of 52 TB patients were 92.2%, 90.6%, and 87.5% at Week 4, 8 and 16-24, respectively, with a strong correlation (r > 0.76) and agreement (ICC > 0.88) with HUPC evaluations. CASI missed one-third of the non-adherent cases reported by HUPC based on patient adherence status. The treatment success rates of patients with >90% adherence, as measured by CASI and HUPC, did not differ significantly; however, >85% adherence was associated with higher treatment success (OR 45.1) than 90% adherence (OR 21.9).CONCLUSION: CASI results were comparable to those of HUPC. As it increased the likelihood of successful treatment, a threshold of >85% may be more appropriate than >90% for defining medication-adherent patients.

Allergic Adverse Drug Events After Alert Overrides in Hospitalized Patients.

OBJECTIVES: This study aimed to assess how often overridden drug allergy alerts (ODAAs) lead to allergic adverse drug events (All-ADEs) and to evaluate the frequency with which drug allergy alerts (DAAs) were overridden and the reasons, as well as appropriateness of these overrides. METHODS: A retrospective observational study of DAA generated between 2014 and 2016 was conducted. The corresponding DAA records were reviewed to determine the frequency of alert overrides. A chart review was performed on a subset of 194 ODAA (the first of every 3 chronologically ordered ODAA) to identify All-ADEs and to evaluate the override reasons and the appropriateness of these overrides. RESULTS: A total of 2044 DAAs were overridden (override rate of 44.8%). Most were triggered by a nonexact match (93.81%), when ordering nervous system (21.1%) and cardiovascular system (19.6%) drugs and were generated by physicians (72.7%). The main override reason was that the patient was already taking the drug or had previously tolerated the drug. Only 9.28% of ODAAs were inappropriately overridden. Six All-ADEs (3.09%) were identified and were due to anti-infective (1), antineoplastic (1), and iodinated-contrast (4) drug administration. Most All-ADEs were cutaneous and were mild. None was life-threatening or fatal. The All-ADEs rate was higher among inappropriately ODAA (15.79%, P = 0.013). CONCLUSIONS: Alert overrides are not exempt from clinical consequences, although few are associated with All-ADEs. It is necessary to identify the drugs involved in those reactions and to update allergy lists to generate only specific and important DAA and to avoid the negative consequences of overrides.

A randomized controlled trial to evaluate the use of a web-based application to manage medications during in vitro fertilization.

OBJECTIVE: To evaluate the use of a web-based application that assists in medication management during in vitro fertilization (IVF) treatment. DESIGN: Multicenter randomized controlled trial. SETTING: University hospitals. PATIENT(S): Women undergoing IVF. INTERVENTION(S): Subjects were recruited to assess quality of life during IVF and were randomly assigned to use either the OnTrack application to assist with medication management or conventional medication management. Surveys were administered at four time points. MAIN OUTCOME MEASURE(S): Medication surplus, incidence of medication errors, amount of patient-initiated communication, and patient satisfaction. RESULT(S): A total of 153 women participated. The average number of portal messages and telephone calls was similar between groups. Twelve patients in the control group (12/69, 17.4%) and 8 patients in the case group (8/72, 11.1%) made medication errors. There were similar amounts of medication surplus in the two groups. The estimated cost of medication waste was $2,578 ± $2,056 in the control group and $2,554 ± $1,855 in the case group. Patient satisfaction was similar between the two groups. CONCLUSION(S): Use of a web-based application did not decrease medication errors, medication surplus, or patient-initiated messages. Many patients had a medication surplus, which can be an area of cost reduction during IVF. CLINICAL TRIAL REGISTRATION NUMBER: Clinicaltrials.gov NCT03383848.

Performance of an Electronic Decision Support System as a Therapeutic Intervention During a Multicenter PICU Clinical Trial: Heart and Lung Failure-Pediatric Insulin Titration Trial (HALF-PINT).

BACKGROUND: The use of electronic clinical decision support (CDS) systems for pediatric critical care trials is rare. We sought to describe in detail the use of a CDS tool (Children's Hospital Euglycemia for Kids Spreadsheet [CHECKS]), for the management of hyperglycemia during the 32 multicenter Heart And Lung Failure-Pediatric Insulin Titration trial. RESEARCH QUESTION: In critically ill pediatric patients who were treated with CHECKS, how was user compliance associated with outcomes; and what patient and clinician factors might account for the observed differences in CHECKS compliance? STUDY DESIGN AND METHODS: During an observational retrospective study of compliance with a CDS tool used during a prospective randomized controlled trial, we compared patients with high and low CHECKS compliance. We investigated the association between compliance and blood glucose metrics. We describe CHECKS and use a computer interface analysis framework (the user, function, representation, and task analysis framework) to categorize user interactions. We discuss implications for future randomized controlled trials. RESULTS: Over a 4.5-year period, 658 of 698 children were treated with the CHECKS protocol for ≥24 hours with a median of 119 recommendations per patient. Compliance per patient was high (median, 99.5%), with only 30 patients having low compliance (<90%). Patients with low compliance were from 16 of 32 sites, younger (P = .02), and less likely to be on inotropic support (P = .04). They were more likely to be have been assigned randomly to the lower blood glucose target (80% vs 48%; P < .001) and to have spent a shorter time (53% vs 75%; P < .001) at the blood glucose target. Overrides (classified by the user, function, representation, and task analysis framework), were largely (89%) due to the user with patient factors contributing 29% of the time. INTERPRETATION: The use of CHECKS for the Heart And Lung Failure-Pediatric Insulin Titration trial resulted in a highly reproducible and explicit method for the management of hyperglycemia in critically ill children across varied environments. CDS systems represent an important mechanism for conducting explicit complex pediatric critical care trials. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT01565941, registered March 29 2012; https://clinicaltrials.gov/ct2/show/NCT01565941?term=HALF-PINT&draw=2&rank=1.

Developing a Consistent, Reproducible Botulinum Toxin Type A Dosing Method for Upper Limb Tremor by Kinematic Analysis.

Botulinum toxin type A (BoNT-A) injection patterns customized to each patient's unique tremor characteristics produce better efficacy and lower adverse effects compared to the fixed-muscle-fixed-dose approach for Essential Tremor (ET) and Parkinson's disease (PD) tremor therapy. This article outlined how a kinematic-based dosing method to standardize and customize BoNT-A injections for tremors was developed. Seven ET and eight PD participants with significant tremor reduction and minimal perceived weakness using optimized BoNT-A injections determined by clinical and kinematic guidance were retrospectively selected to develop the kinematic-based dosing method. BoNT-A dosages allocated per joint were paired to baseline tremor amplitudes per joint. The final kinematic-based dosing method was prospectively utilized to validate BoNT-A injection pattern selection without clinical/visual assessments in 31 ET and 47 PD participants with debilitating arm tremors (totaling 122 unique tremor patterns). Whole-arm kinematic tremor analysis was performed at baseline and 6-weeks post-injection. Correlation and linear regression analyses between baseline tremor amplitudes and the change in tremor amplitude 6-weeks post-injection, with BoNT-A dosages per joint, were performed. Injection patterns determined using clinical assessment and interpretation of kinematics produced significant associations between baseline tremor amplitudes and optimized BoNT-A dosages in all joints. The change in elbow tremor was only significantly associated with the elbow total dose as the change in the wrist and shoulder tremor amplitudes were not significantly associated with the wrist and shoulder dosages from the selected 15 ET and PD participants. Using the kinematic-based dosing method, significant associations between baseline tremor amplitudes and the change (6-weeks post-first treatment) in tremor at each joint with BoNT-A dosages for all joints was observed in all 78 ET and PD participants. The kinematic-based dosing method provided consistency in dose selection and subsequent tremor reduction and can be used to standardize tremor assessments for whole-arm tremor treatment planning.

[Adverse Drug Effects - how to detect, how to avoid?] / Arzneimittelnebenwirkungen ­ wie erkennen, wie vermeiden?

There is no drug therapy without risk for toxicity. The patient must tolerate some toxic or idiosyncratic adverse events (e. g. hand foot syndrome). After life threatening adverse effects, one needs to screen (hyperkalemia). The sick-day-rule should be communicated with the informed patient as for example stopping SGLT2-inhibitors (loss of appetite, hypotension). The list of prescriptions should regularly criticized for dangerous or superfluous medicines (deprescribing).

The effect of medication related clinical decision support at the time of physician order entry.

Background In advanced clinical decision support systems, patient characteristics and laboratory values are included in the algorithms that generate alerts. These alerts have a higher specificity than basic medication surveillance alerts. The alerts of advanced clinical decision support systems can be shown directly to the prescriber during order entry, without the risk of generating an overload of irrelevant alerts. We implemented five advanced algorithms that are shown directly to the prescriber. These algorithms are for gastrointestinal prophylaxis, folic or folinic acid prescribed with orally or subcutaneously administered methotrexate, vitamin D prescribed with bisphosphonates, hyponatremia and measuring plasma levels for vancomycin and gentamicin. Objective We evaluated the effect of the implementation of the algorithms. Setting We performed prospective intervention studies with a historical group for comparison in both inpatients and outpatients at a teaching hospital in the Netherlands. Methods We compared the time period after implementation of the algorithm with the time period before implementation, using data from the hospital information system Epic. Difference in guideline adherence were analyzed using Chi square tests. Main outcome measure The outcome measures were the number of alerts, the acceptance rate of the advice in the alert, and for the algorithm measuring plasma levels for vancomycin and gentamicin the time to the correct dose. Results For all algorithms, the implementation resulted in a significant increase in guideline adherence, varying from 11 to 36%. The acceptance rate varied from 14% for hyponatremia to 90% for methotrexate. For gastrointestinal prophylaxis the acceptance rate was 4.4% for basic drug-drug interaction alerts when no gastrointestinal prophylaxis was prescribed and increased to 44.7% after implementation of the advanced algorithm. This algorithm substantially decreased the number of alerts from 812 before implementation to 217 after implementation. After implementation of the algorithm for measuring plasma levels for vancomycin and gentamicin, the proportion of patients receiving the correct dose after 48 h increased from 73 to 84% (p = 0.03). Conclusion Implementation of advanced algorithms that take patient characteristics into account and are shown directly to the physician during order entry, result in an increased guideline adherence.

Software de contenido farmacocinético para la monitorización terapéutica de fármacos: Protocolo de revisión exploratoria / Pharmacokinetic software for therapeutic drug monitoring: A scoping review protoco

Objetivo: Los nomogramas, ecuaciones y software de contenido farmacocinético se consideran las principales herramientas disponibles para la monitorización farmacocinética clínica. Debido a su gran aplicabilidad en numerososgrupos de fármacos, el empleo de software se encuentra ampliamente extendido en la práctica clínica. Generalmente, el objetivo principal de los estudiosque incluyen el uso de estos software no es la descripción de los mismos, porlo que la información disponible es escasa y, además, no se dispone de unarevisión que aúne toda la información disponible referente a este tipo de softwareEl objetivo de este estudio será sintetizar la evidencia disponible sobre losdistintos software de aplicación en la monitorización farmacocinética parafacilitar a los usuarios su identificación, evaluación y selección.Método: Se realizará una revisión exploratoria de la literatura cuyo protocolo se describe en este artículo, de acuerdo con las recomendacionesPRISMA para la elaboración de revisiones exploratorias y publicaciónde protocolos. Se realizará una búsqueda bibliográfica en las bases dedatos Medline, Embase, OpenAire y Bielefeld Academic Search Engine.Se incluirán en el estudio aquellos software detectados de los que se disponga de la siguiente información: nombre del software, desarrollador/comercializador, tipo de análisis farmacocinético y fármacos incluidos.Resultados: En este estudio se espera realizar una síntesis de lascaracterísticas más relevantes en la práctica clínica de los software decontenido farmacocinético disponibles en el mercado. Se realizará una síntesis narrativa crítica de las fuentes de información utilizadas. Además,se llevará a cabo, si es posible, una comparación de los mismos parafacilitar la evaluación y selección por parte de los usuarios. (AU)

Objective: Nomograms, equations and pharmacokinetic software areconsidered the main tools available for therapeutic drug monitoring. Dueto its great applicability to various groups of drugs, the use of software iswidely extended in clinical practice. The main goals of the studies usingthis type of software do not normally include the description of its features,therefore, the information about its characteristic is scarce. Moreover, noreview of the literature has been published that brings together all theinformation available about these software. The present study aimed tosynthesize the available evidence regarding software applied to therapeutic drug monitoring to facilitate its identification, evaluation and selectionby users.Method: This article describe a scoping review protocol, developedfollowing the PRISMA-P and PRISMA-ScR guidelines. An electronic literature search was performed in MEDLINE, EMBASE, OpenAire and BASE(Bielefeld Academic Search Engine) databases. Only those software forwhich the following information was available were included: name of thesoftware, developer/marketer, type of pharmacokinetic analysis allowed,and drugs included in the analysis.Results: In this study we will synthesized the most relevant characteristics for the clinical practice of the pharmacokinetic software available.A critical appraisal of the sources if information will be included. Also,if it is possible, a comparison of the available tools will be carried out in order to facilitate the evaluation and selection of pharmacokineticsoftware. (AU)

Computer-based simulation to reduce EHR-related chemotherapy ordering errors.

BACKGROUND: The electronic health record (EHR) is a contributor to serious patient harm occurring within a sociotechnical system. Chemotherapy ordering is a high-risk task due to the complex nature of ordering workflows and potential detrimental effects if wrong chemotherapeutic doses are administered. Many chemotherapy ordering errors cannot be mitigated through systems-based changes due to the limited extent to which individual institutions are able to customize proprietary EHR software. We hypothesized that simulation-based training could improve providers' ability to identify and mitigate common chemotherapy ordering errors. METHODS: Pediatric hematology/oncology providers voluntarily participated in simulations using an EHR testing ("Playground") environment. The number of safety risks identified and mitigated by each provider at baseline was recorded. Risks were reviewed one-on-one after initial simulations and at a group "lunch-and-learn" session. At three-month follow-up, repeat simulations assessed for improvements in error identification and mitigation, and providers were surveyed about prevention of real-life safety events. RESULTS: The 8 participating providers identified and mitigated an average of 5.5 out of 10 safety risks during the initial simulation, compared 7.4 safety risks at the follow up simulation (p=0.030). Two of the providers (25%) reported preventing at least one real-world patient safety event in the clinical setting as a result of the initial training session. CONCLUSIONS: Simulation-based training may reduce providers' susceptibility to chemotherapy ordering safety vulnerabilities within the EHR. This approach may be used when systems-based EHR improvements are not feasible due to limited ability to customize local instances of proprietary EHR software.

Human-Based Errors Involving Smart Infusion Pumps: A Catalog of Error Types and Prevention Strategies.

Over 4000 preventable injuries due to medication errors occur each year in any given hospital. Smart pumps have been widely introduced as one means to prevent these errors. Although smart pumps have been implemented to prevent errors, they fail to prevent specific types of errors in the medication administration process and may introduce new errors themselves. As a result, unique prevention strategies have been implemented by providers. No catalog of smart pump error types and prevention strategies currently exists. The aim of this study is to review and catalog the types of human-based errors related to smart pump use identified in the literature and to summarize the associated error-prevention strategies. We searched MEDLINE, PubMed, PubMed Central, and Cumulative Index to Nursing and Allied Health Literature (CINAHL) for literature pertaining to human-based errors associated with smart pumps. Studies related to smart pump implementation, other types of pumps, and mechanical failures were excluded. Final selections were mapped for error types and associated prevention strategies. A total of 1177 articles were initially identified, and 105 articles were included in the final review. Extraction of error types and prevention strategies resulted in the identification of 18 error types and ten prevention strategies. Through a comprehensive literature review, we compiled a catalog of smart pump-related errors and associated prevention strategies. Strategies were mapped to error types to provide an initial framework for others to use as a resource in their error reviews and improvement work. Future research should assess the application of the resources provided by this review.

Bolus pharmacokinetics: moving beyond mass-based dosing to guide drug administration.

Despite the common approach of bolus drug dosing using a patient's mass, a more tailored approach would be to use empirically derived pharmacokinetic models. Previously, this could only be possible though the use of computer simulation using programs which are rarely available in clinical practice. Through mathematical manipulations and approximations, a simplified set of equations is demonstrated that can identify a bolus dose required to achieve a specified target effect site concentration. The proposed solution is compared against simulations of a wide variety of pharmacokinetic models. This set of equations provides a near-identical solution to the simulation approach. A boundary condition is established to ensure the derived equations have an acceptable error. This approach may allow for more precise administration of medications with the use of point of care technology and potentially allows for pharmacokinetic dosing in artificial intelligence problems.

CT fluoroscopy-guided transforaminal and intra-articular facet steroid injections for the treatment of cervical radiculopathy: injectate distribution patterns and association with clinical outcome.

OBJECTIVES: To investigate injectate dispersal patterns and their association with therapeutic efficacy during a transforaminal (TFSI) or an intra-articular facet steroid injection (IFSI) to treat cervical radiculopathy. METHODS: This retrospective study examined the post-intervention cervical spine CT of 56 patients randomized to receive one CT fluoroscopy-guided IFSI (29 patients; 10 (34.5%) males; mean age 45.0 years; SD 8.8 years; range 26-61 years) or TFSI (27 patients; 13 (48.2%) males; mean age 51.1 years; SD 11.2 years; range 29-72 years) (December 2010 to August 2013). The presence of contrast within the intra-articular facet, juxta-articular facet, retrodural, epidural, and foraminal and extraforaminal spaces during IFSI, and within the extraforaminal, foraminal, and epidural spaces during TFSI was assessed. Descriptive data are presented as frequencies. The association between injectate dispersal patterns and therapeutic efficacy, 4-week post-intervention, was assessed with ANCOVA models. RESULTS: During IFSI, the injectate predominantly spread to the retrodural (62%; 18/29) or juxta-articular (21%; 6/29) space. During TFSI, the injectate predominantly spread to the extraforaminal/foraminal spaces (41%; 11/27) or to the extraforaminal/foraminal/epidural spaces (33%; 9/27). Injectate presence in the juxta-articular (p = .007) or extraforaminal (p < .001) space was a predictor of therapeutic efficacy but not in the foraminal (p = .54), epidural (p = .89), or retrodural (p = .75) space. CONCLUSIONS: TFSI and IFSI led to preferential extraforaminal and retrodural injectate spread, respectively. Targeting the extraforaminal or juxta-articular facet space improved the clinical efficacy of steroid injections when treating cervical radiculopathy. KEY POINTS: • During intra-articular facet injection, the injectate spreads from the facet joint to the retrodural space and rarely reaches the epidural and/or foraminal spaces. • Epidural spread of the injectate during an anterolateral transforaminal steroid injection is the least effective for pain relief in patients with cervical radiculopathy. • Injection techniques targeting the extraforaminal or juxta-articular facet space are safer than transforaminal injections and effectively relieve pain in patients with cervical radiculopathy.

La implementación de la administración de medicamentos con código de barras y las bombas de infusión inteligentes es sólo el comienzo del camino seguro para prevenir los errores de administración / Implementing barcode medication administration and smart infusion pumps is just the beginning of the safety journey to prevent administration errors

INTRODUCCIÓN: La tecnología sanitaria se ha convertido en la solución más aceptada para reducir los eventos adversos provocados por los medicamentos, minimizando los posibles errores humanos. La introducción de la tecnología puede mejorar la seguridad y permitir una mayor eficiencia en la clínica. Sin embargo, no elimina todos los tipos de error y puede crear otros nuevos. La administración de medicamentos con código de barras y la utilización de bombas de infusión inteligentes son dos estrategias que pueden emplearse durante la administración de medicamentos para evitar errores antes de que estos lleguen al paciente. OBJETIVO: En este artículo se han revisado diferentes tipos de errores relativos a la administración de medicamentos con código de barras y las bombas de infusión inteligentes, y se ha examinado la forma en la que se producían dichos errores al emplear la tecnología. También se exponen las recomendaciones encaminadas a evitar este tipo de errores. CONCLUSIÓN: Los hospitales deben comprender la tecnología, su funcionamiento y los errores que pretende evitar, así como analizar de qué manera cambiará los procesos clínicos. Es esencial que la dirección del hospital establezca las métricas necesarias y las monitorice regularmente para garantizar el uso óptimo de estas tecnologías. También es importante identificar y evitar desviaciones en los procesos que puedan eliminar o disminuir los beneficios de seguridad para los que fue diseñada. De igual forma, es necesario recopilar periódicamente las opiniones del profesional que la utiliza para detectar los posibles problemas que pudieran surgir. Sin embargo, la dirección debe ser consciente de que incluso con la implementación completa de la tecnología pueden surgir errores a la hora de administrar la medicación

INTRODUCTION: Healthcare-related technology has been widely accepted as a key patient safety solution to reduce adverse drug events by decreasing the risk of human error. The introduction of technology can enhance safety and support workflow; however, it does not eliminate all error types and may create new ones. Barcode medication adminis-tration and smart infusion pumps are two technologies utilized during medication administration to prevent medication errors before they reach the patient. OBJECTIVE: This article reviewed different error types with barcode medi-cation administration and smart infusion pumps and examined how these errors were able to occur while using the technology. Recommendations for preventing these types of errors were also discussed. CONCLUSION: Hospitals must understand the technology, how it is desig-ned to work, which errors it is intended to prevent, as well as understand how it will change staff workflow. It is essential that metrics are set by hospital leadership and regularly monitored to ensure optimal use of these technologies. It is also important to identify and avoid workarounds which eliminate or diminish the safety benefits that the technology was designed to achieve. Front line staff feedback should be gathered on a periodic basis to understand any struggles with utilizing the technology. Leaders must also understand that even with full implementation of technology, medication errors may still occur

Augmented Intelligence Dermatology: Deep Neural Networks Empower Medical Professionals in Diagnosing Skin Cancer and Predicting Treatment Options for 134 Skin Disorders.

Although deep learning algorithms have demonstrated expert-level performance, previous efforts were mostly binary classifications of limited disorders. We trained an algorithm with 220,680 images of 174 disorders and validated it using Edinburgh (1,300 images; 10 disorders) and SNU datasets (2,201 images; 134 disorders). The algorithm could accurately predict malignancy, suggest primary treatment options, render multi-class classification among 134 disorders, and improve the performance of medical professionals. The area under the curves for malignancy detection were 0.928 ± 0.002 (Edinburgh) and 0.937 ± 0.004 (SNU). The area under the curves of primary treatment suggestion (SNU) were 0.828 ± 0.012, 0.885 ± 0.006, 0.885 ± 0.006, and 0.918 ± 0.006 for steroids, antibiotics, antivirals, and antifungals, respectively. For multi-class classification, the mean top-1 and top-5 accuracies were 56.7 ± 1.6% and 92.0 ± 1.1% (Edinburgh) and 44.8 ± 1.2% and 78.1 ± 0.3% (SNU), respectively. With the assistance of our algorithm, the sensitivity and specificity of 47 clinicians (21 dermatologists and 26 dermatology residents) for malignancy prediction (SNU; 240 images) were improved by 12.1% (P < 0.0001) and 1.1% (P < 0.0001), respectively. The malignancy prediction sensitivity of 23 non-medical professionals was significantly increased by 83.8% (P < 0.0001). The top-1 and top-3 accuracies of four doctors in the multi-class classification of 134 diseases (SNU; 2,201 images) were increased by 7.0% (P = 0.045) and 10.1% (P = 0.0020), respectively. The results suggest that our algorithm may serve as augmented intelligence that can empower medical professionals in diagnostic dermatology.

High-priority drug-drug interaction clinical decision support overrides in a newly implemented commercial computerized provider order-entry system: Override appropriateness and adverse drug events.

OBJECTIVE: The study sought to determine frequency and appropriateness of overrides of high-priority drug-drug interaction (DDI) alerts and whether adverse drug events (ADEs) were associated with overrides in a newly implemented electronic health record. MATERIALS AND METHODS: We conducted a retrospective study of overridden high-priority DDI alerts occurring from April 1, 2016, to March 31, 2017, from inpatient and outpatient settings at an academic health center. We studied highest-severity DDIs that were previously designated as "hard stops" and additional high-priority DDIs identified from clinical experience and literature review. All highest-severity alert overrides (n = 193) plus a stratified random sample of additional overrides (n = 371) were evaluated for override appropriateness, using predetermined criteria. Charts were reviewed to identify ADEs for overrides that resulted in medication administration. A chi-square test was used to compare ADE rate by override appropriateness. RESULTS: Of 16 011 alerts presented to providers, 15 318 (95.7%) were overridden, including 193 (87.3%) of the highest-severity DDIs and 15 125 (95.8%) of additional DDIs. Override appropriateness was 45.4% overall, 0.5% for highest-severity DDIs and 68.7% for additional DDIs. For alerts that resulted in medication administration (n = 423, 75.0%), 29 ADEs were identified (6.9%, 5.1 per 100 overrides). The rate of ADEs was higher with inappropriate vs appropriate overrides (9.4% vs 4.3%; P = .038). CONCLUSIONS: The override rate was nearly 90% for even the highest-severity DDI alerts, indicating that stronger suggestions should be made for these alerts, while other alerts should be evaluated for potential suppression.

Computer-Based versus Paper-Based Insulin Infusion Algorithms in Diabetic Ketoacidosis.

BACKGROUND: Paper-based and computer-based insulin infusion algorithms facilitate appropriate glycemic therapy. The data comparing these algorithms in the management of diabetic ketoacidosis in the intensive care unit (ICU) setting are limited. We aimed to determine the differences in time to diabetic ketoacidosis resolution and incidence of hypoglycemia between computer and paper-based insulin infusion. METHODS: Single-institution retrospective review of patients admitted to the ICU with diabetic ketoacidosis between 4/1/2015 and 7/20/2018. Our institution introduced computer-based insulin infusion (Glucommander) to the intensive care unit on 3/28/2016. Patients were grouped into either paper-based group (preintervention) or a computer-based group (postintervention). Summary and univariate analyses were performed. RESULTS: A total of 620 patients (paper-based=247; computer-based=373) with a median (IQR) age of 40 (26-56) years were included; 46% were male. Patients in the computer-based group were significantly older (p=0.003); otherwise, there were no significant differences in gender, race, body mass index and HbA1c. The mean (±SD) time to diabetic ketoacidosis resolution in the computer-based group was significantly lower than the paper-based group (p=0.02). The number of patients in the paper-based group who developed severe hypoglycemia (<50 mg/dl) was significantly higher {8% vs 1%; p<0.0001}. CONCLUSION: Our analyses demonstrate statistically significant decreases in time to DKA resolution and hypoglycemic events in DKA patients who were managed using a computer-based insulin infusion algorithm providing a more effective and safer option when compared to paper-based insulin infusion.

Drug repurposing to improve treatment of rheumatic autoimmune inflammatory diseases.

The past century has been characterized by intensive efforts, within both academia and the pharmaceutical industry, to introduce new treatments to individuals with rheumatic autoimmune inflammatory diseases (RAIDs), often by 'borrowing' treatments already employed in one RAID or previously used in an entirely different disease, a concept known as drug repurposing. However, despite sharing some clinical manifestations and immune dysregulation, disease pathogenesis and phenotype vary greatly among RAIDs, and limited understanding of their aetiology has made repurposing drugs for RAIDs challenging. Nevertheless, the past century has been characterized by different 'waves' of repurposing. Early drug repurposing occurred in academia and was based on serendipitous observations or perceived disease similarity, often driven by the availability and popularity of drug classes. Since the 1990s, most biologic therapies have been developed for one or several RAIDs and then tested among the others, with varying levels of success. The past two decades have seen data-driven repurposing characterized by signature-based approaches that rely on molecular biology and genomics. Additionally, many data-driven strategies employ computational modelling and machine learning to integrate multiple sources of data. Together, these repurposing periods have led to advances in the treatment for many RAIDs.

New artificial intelligence prediction model using serial prothrombin time international normalized ratio measurements in atrial fibrillation patients on vitamin K antagonists: GARFIELD-AF.

AIMS: Most clinical risk stratification models are based on measurement at a single time-point rather than serial measurements. Artificial intelligence (AI) is able to predict one-dimensional outcomes from multi-dimensional datasets. Using data from Global Anticoagulant Registry in the Field (GARFIELD)-AF registry, a new AI model was developed for predicting clinical outcomes in atrial fibrillation (AF) patients up to 1 year based on sequential measures of prothrombin time international normalized ratio (PT-INR) within 30 days of enrolment. METHODS AND RESULTS: Patients with newly diagnosed AF who were treated with vitamin K antagonists (VKAs) and had at least three measurements of PT-INR taken over the first 30 days after prescription were analysed. The AI model was constructed with multilayer neural network including long short-term memory and one-dimensional convolution layers. The neural network was trained using PT-INR measurements within days 0-30 after starting treatment and clinical outcomes over days 31-365 in a derivation cohort (cohorts 1-3; n = 3185). Accuracy of the AI model at predicting major bleed, stroke/systemic embolism (SE), and death was assessed in a validation cohort (cohorts 4-5; n = 1523). The model's c-statistic for predicting major bleed, stroke/SE, and all-cause death was 0.75, 0.70, and 0.61, respectively. CONCLUSIONS: Using serial PT-INR values collected within 1 month after starting VKA, the new AI model performed better than time in therapeutic range at predicting clinical outcomes occurring up to 12 months thereafter. Serial PT-INR values contain important information that can be analysed by computer to help predict adverse clinical outcomes.