A polygenic risk score for the QT interval is an independent predictor of drug-induced QT prolongation.

Drug-induced QT prolongation (diLQTS), and subsequent risk of torsade de pointes, is a major concern with use of many medications, including for non-cardiac conditions. The possibility that genetic risk, in the form of polygenic risk scores (PGS), could be integrated into prediction of risk of diLQTS has great potential, although it is unknown how genetic risk is related to clinical risk factors as might be applied in clinical decision-making. In this study, we examined the PGS for QT interval in 2500 subjects exposed to a known QT-prolonging drug on prolongation of the QT interval over 500ms on subsequent ECG using electronic health record data. We found that the normalized QT PGS was higher in cases than controls (0.212±0.954 vs. -0.0270±1.003, P = 0.0002), with an unadjusted odds ratio of 1.34 (95%CI 1.17-1.53, P<0.001) for association with diLQTS. When included with age and clinical predictors of QT prolongation, we found that the PGS for QT interval provided independent risk prediction for diLQTS, in which the interaction for high-risk diagnosis or with certain high-risk medications (amiodarone, sotalol, and dofetilide) was not significant, indicating that genetic risk did not modify the effect of other risk factors on risk of diLQTS. We found that a high-risk cutoff (QT PGS ≥ 2 standard deviations above mean), but not a low-risk cutoff, was associated with risk of diLQTS after adjustment for clinical factors, and provided one method of integration based on the decision-tree framework. In conclusion, we found that PGS for QT interval is an independent predictor of diLQTS, but that in contrast to existing theories about repolarization reserve as a mechanism of increasing risk, the effect is independent of other clinical risk factors. More work is needed for external validation in clinical decision-making, as well as defining the mechanism through which genes that increase QT interval are associated with risk of diLQTS.

Development and validation of the Atri-Risk Conduction Index risk score to predict risk of atrial fibrillation after typical atrial flutter ablation.

BACKGROUND: Identification of patients at risk for atrial fibrillation (AF) after typical atrial flutter (tAFL) ablation is important to guide monitoring and treatment. OBJECTIVE: The purpose of this study was to create and validate a risk score to predict AF after tAFL ablation METHODS: We identified patients who underwent tAFL ablation with no AF history between 2017 and 2022 and randomly allocated to derivation and validation cohorts. We collected clinical variables and measured conduction parameters in sinus rhythm on an electrophysiology recording system (CardioLab, GE Healthcare). Univariate and multivariate logistic regressions (LogR) were used to evaluate association with AF development. RESULTS: A total of 242 consecutive patients (81% male; mean age 66 ± 11 years) were divided into derivation (n =142) and validation (n = 100) cohorts. Forty-two percent developed AF over median follow-up of 330 days. In multivariate LogR (derivation cohort), proximal to distal coronary sinus time (pCS-dCS) ≥70 ms (odds ratio [OR] 16.7; 95% confidence interval [CI] 5.6-49), pCS time ≥36 ms (OR 4.5; 95% CI 1.5-13), and CHADS2-VASc score ≥3 (OR 4.3; 95% CI 1.6-11.8) were independently associated with new AF during follow-up. The Atri-Risk Conduction Index (ARCI) score was created with 0 as minimal and 4 as high-risk using pCS-dCS ≥70 ms = 2 points; pCS ≥36 ms = 1 point; and CHADS2-VASc score ≥3 = 1 point. In the validation cohort, 0% of patients with ARCI score = 0 developed AF, whereas 89% of patients with ARCI score = 4 developed AF. CONCLUSION: We developed and validated a risk score using atrial conduction parameters and clinical risk factors to predict AF after tAFL ablation. It stratifies low-, moderate-, and high-risk patients and may be helpful in individualizing approaches to AF monitoring and anticoagulation.

Evaluating the prescribing and monitoring of medications associated with QTc-prolongation in the ambulatory care setting.

RATIONALE: Little is known about the prescribing of medications with potential to cause QTc-prolongation in the ambulatory care settings. Understanding real-world prescribing of QTc-prolonging medications and actions taken to mitigate this risk will help guide strategies to optimize safety and appropriate prescribing among ambulatory patients. OBJECTIVE: To evaluate the frequency of clinician action taken to monitor and mitigate modifiable risk factors for QTc-prolongation when indicated. METHODS: This retrospective, cross-sectional study evaluated clinician action at the time of prescribing prespecified medications with potential to prolong QTc in adult patients in primary care. The index date was defined as the date the medication was ordered. Electronic health record (EHR) data were evaluated to assess patient, clinician and visit characteristics. Clinician action was determined if baseline or follow-up monitoring was ordered or if action was taken to mitigate modifiable risk factors (laboratory abnormalities or electrocardiogram [ECG] monitoring) within 48 h of prescribing a medication with QTc-prolonging risk. Descriptive statistics were used to describe current practice. RESULTS: A total of 399 prescriptions were prescribed to 386 patients, with a mean age of 51 ± 18 years, during March 2021 from a single-centre, multisite health system. Of these, 17 (4%) patients had a known history of QTc-prolongation, 170 (44%) did not have a documented history of QTc-prolongation and 199 (52%) had an unknown history (no ECG documented). Thirty-nine patients (10%) had at least one laboratory-related risk factor at the time of prescribing, specifically hypokalemia (16 patients), hypomagnesemia (8 patients) or hypocalcemia (19 patients). Of these 39 patients with laboratory risk factors, only 6 patients (15%) had their risk acknowledged or addressed by a clinician. Additionally, eight patients' most recent QTc was ≥500 ms and none had an ECG checked at the time the prescription was ordered. CONCLUSION: Despite national recommendations, medication monitoring and risk mitigation is infrequent when prescribing QTc-prolonging medications in the ambulatory care setting. These findings call for additional research to better understand this gap, including reasons for the gap and consequences on patient outcomes.

A Discussion of the Contemporary Prediction Models for Atrial Fibrillation.

Atrial Fibrillation is a complex disease state with many environmental and genetic risk factors. While there are environmental factors that have been shown to increase an individual's risk of atrial fibrillation, it has become clear that atrial fibrillation has a genetic component that influences why some patients are at a higher risk of developing atrial fibrillation compared to others. This review will first discuss the clinical diagnosis of atrial fibrillation and the corresponding rhythm atrial flutter. We will then discuss how a patients' risk of stroke can be assessed by using other clinical co-morbidities. We will then review the clinical risk factors that can be used to help predict an individual patient's risk of atrial fibrillation. Many of the clinical risk factors have been used to create several different risk scoring methods that will be reviewed. We will then discuss how genetics can be used to identify individuals who are at higher risk for developing atrial fibrillation. We will discuss genome-wide association studies and other sequencing high-throughput sequencing studies. Finally, we will touch on how genetic variants derived from a genome-wide association studies can be used to calculate an individual's polygenic risk score for atrial fibrillation. An atrial fibrillation polygenic risk score can be used to identify patients at higher risk of developing atrial fibrillation and may allow for a reduction in some of the complications associated with atrial fibrillation such as cerebrovascular accidents and the development of heart failure. Finally, there is a brief discussion of how artificial intelligence models can be used to predict which patients will develop atrial fibrillation.

Profound First-Degree Atrioventricular Block in a High-Level Basketball Athlete.

First-degree atrioventricular (AV) block (PR interval >200 ms) is commonly observed among screening electrocardiogram (ECG) in athletes. Profound first-degree AV block (PR interval >400 ms) and Mobitz type I (Wenckebach) second-degree AV block are generally uncommon and often require further workup on a case-by-case basis, particularly when there is concern for a structural cardiac abnormality. In this case, we present an example of an asymptomatic profound first-degree AV block with Mobitz type I (Wenckebach) second-degree AV block. Transthoracic echocardiogram and stress echocardiogram were unremarkable and the patient was cleared to participate in sports without any restriction. Physicians managing athletes should be aware of ECG features that require additional evaluation and cardiology consultation.

PRIME score for prediction of permanent pacemaker implantation after transcatheter aortic valve replacement.

OBJECTIVES: We sought to produce a simple scoring system that can be applied at clinical visits before transcatheter aortic valve replacement (TAVR) to stratify the risk of permanent pacemaker (PPM) after the procedure. BACKGROUND: Atrioventricular block is a known complication of TAVR. Current models for predicting the risk of PPM after TAVR are not designed to be applied clinically to assist with preprocedural planning. METHODS: Patients undergoing TAVR at the University of Colorado were split into a training cohort for the development of a predictive model, and a testing cohort for model validation. Stepwise and binary logistic regressions were performed on the training cohort to produce a predictive model. Beta coefficients from the binary logistic regression were used to create a simple scoring system for predicting the need for PPM implantation. Scores were then applied to the validation cohort to assess predictive accuracy. RESULTS: Patients undergoing TAVR from 2013 to 2019 were analyzed: with 483 included in the training cohort and 123 included in the validation cohort. The need for a pacemaker was associated with five preprocedure variables in the training cohort: PR interval > 200 ms, Right bundle branch block, valve-In-valve procedure, prior Myocardial infarction, and self-Expandable valve. The PRIME score was developed using these clinical features, and was highly accurate for predicting PPM in both the training and model validation cohorts (area under the curve 0.804 and 0.830 in the model training and validation cohorts, respectively). CONCLUSIONS: The PRIME score is a simple and accurate preprocedural tool for predicting the need for PPM implantation after TAVR.

A machine learning evaluation of patient characteristics associated with prescribing of guideline-directed medical therapy for heart failure.

Introduction/background: Patients with heart failure and reduced ejection fraction (HFrEF) are consistently underprescribed guideline-directed medications. Although many barriers to prescribing are known, identification of these barriers has relied on traditional a priori hypotheses or qualitative methods. Machine learning can overcome many limitations of traditional methods to capture complex relationships in data and lead to a more comprehensive understanding of the underpinnings driving underprescribing. Here, we used machine learning methods and routinely available electronic health record data to identify predictors of prescribing. Methods: We evaluated the predictive performance of machine learning algorithms to predict prescription of four types of medications for adults with HFrEF: angiotensin converting enzyme inhibitor/angiotensin receptor blocker (ACE/ARB), angiotensin receptor-neprilysin inhibitor (ARNI), evidence-based beta blocker (BB), or mineralocorticoid receptor antagonist (MRA). The models with the best predictive performance were used to identify the top 20 characteristics associated with prescribing each medication type. Shapley values were used to provide insight into the importance and direction of the predictor relationships with medication prescribing. Results: For 3,832 patients meeting the inclusion criteria, 70% were prescribed an ACE/ARB, 8% an ARNI, 75% a BB, and 40% an MRA. The best-predicting model for each medication type was a random forest (area under the curve: 0.788-0.821; Brier score: 0.063-0.185). Across all medications, top predictors of prescribing included prescription of other evidence-based medications and younger age. Unique to prescribing an ARNI, the top predictors included lack of diagnoses of chronic kidney disease, chronic obstructive pulmonary disease, or hypotension, as well as being in a relationship, nontobacco use, and alcohol use. Discussion/conclusions: We identified multiple predictors of prescribing for HFrEF medications that are being used to strategically design interventions to address barriers to prescribing and to inform further investigations. The machine learning approach used in this study to identify predictors of suboptimal prescribing can also be used by other health systems to identify and address locally relevant gaps and solutions to prescribing.

Clinician challenges to evidence-based prescribing for heart failure and reduced ejection fraction: A qualitative evaluation.

BACKGROUND: Reasons for suboptimal prescribing for heart failure with reduced ejection fraction (HFrEF) have been identified, but it is unclear if they remain relevant with recent advances in healthcare delivery and technologies. This study aimed to identify and understand current clinician-perceived challenges to prescribing guideline-directed HFrEF medications. METHODS: We conducted content analysis methodology, including interviews and member-checking focus groups with primary care and cardiology clinicians. Interview guides were informed by the Cabana Framework. RESULTS: We conducted interviews with 33 clinicians (13 cardiology specialists, 22 physicians) and member checking with 10 of these. We identified four levels of challenges from the clinician perspective. Clinician level challenges included misconceptions about guideline recommendations, clinician assumptions (e.g., drug cost or affordability), and clinical inertia. Patient-clinician level challenges included misalignment of priorities and insufficient communication. Clinician-clinician level challenges were primarily between generalists and specialists, including lack of role clarity, competing priorities of providing focused versus holistic care, and contrasting confidence regarding safety of newer drugs. Policy and system/organisation level challenges included insufficient access to timely/reliable patient data, and unintended care gaps for medications without financially incentivized metrics. CONCLUSION: This study presents current challenges faced by cardiology and primary care which can be used to strategically design interventions to improve guideline-directed care for HFrEF. The findings support the persistence of many challenges and also sheds light on new challenges. New challenges identified include conflicting perspectives between generalists and specialists, hesitancy to prescribe newer medications due to safety concerns, and unintended consequences related to value-based reimbursement metrics for select medications.

Evaluation of Quantitative Decision-Making for Rhythm Management of Atrial Fibrillation Using Tabular Q-Learning.

Background Rhythm management is a complex decision for patients with atrial fibrillation (AF). Although clinical trials have identified subsets of patients who might benefit from a given rhythm-management strategy, for individual patients it is not always clear which strategy is expected to have the greatest mortality benefit or durability. Methods and Results In this investigation 52 547 patients with a new atrial fibrillation diagnosis between 2010 and 2020 were retrospectively identified. We applied a type of artificial intelligence called tabular Q-learning to identify the optimal initial rhythm-management strategy, based on a composite outcome of mortality, change in treatment, and sustainability of the given treatment, termed the reward function. We first applied an unsupervised learning algorithm using a variational autoencoder with K-means clustering to cluster atrial fibrillation patients into 8 distinct phenotypes. We then fit a Q-learning algorithm to predict the best outcome for each cluster. Although rate-control strategy was most frequently selected by treating providers, the outcome was superior for rhythm-control strategies across all clusters. Subjects in whom provider-selected treatment matched the Q-table recommendation had fewer total deaths (4 [8.5%] versus 473 [22.4%], odds ratio=0.32, P=0.02) and a greater reward (P=4.8×10-6). We then demonstrated application of dynamic learning by updating the Q-table prospectively using batch gradient descent, in which the optimal strategy in some clusters changed from cardioversion to ablation. Conclusions Tabular Q-learning provides a dynamic and interpretable approach to apply artificial intelligence to clinical decision-making for atrial fibrillation. Further work is needed to examine application of Q-learning prospectively in clinical patients.

Qualitative Evaluation of an Artificial Intelligence-Based Clinical Decision Support System to Guide Rhythm Management of Atrial Fibrillation: Survey Study.

BACKGROUND: Despite the numerous studies evaluating various rhythm control strategies for atrial fibrillation (AF), determination of the optimal strategy in a single patient is often based on trial and error, with no one-size-fits-all approach based on international guidelines/recommendations. The decision, therefore, remains personal and lends itself well to help from a clinical decision support system, specifically one guided by artificial intelligence (AI). QRhythm utilizes a 2-stage machine learning (ML) model to identify the optimal rhythm management strategy in a given patient based on a set of clinical factors, in which the model first uses supervised learning to predict the actions of an expert clinician and identifies the best strategy through reinforcement learning to obtain the best clinical outcome-a composite of symptomatic recurrence, hospitalization, and stroke. OBJECTIVE: We qualitatively evaluated a novel, AI-based, clinical decision support system (CDSS) for AF rhythm management, called QRhythm, which uses both supervised and reinforcement learning to recommend either a rate control or one of 3 types of rhythm control strategies-external cardioversion, antiarrhythmic medication, or ablation-based on individual patient characteristics. METHODS: Thirty-three clinicians, including cardiology attendings and fellows and internal medicine attendings and residents, performed an assessment of QRhythm, followed by a survey to assess relative comfort with automated CDSS in rhythm management and to examine areas for future development. RESULTS: The 33 providers were surveyed with training levels ranging from resident to fellow to attending. Of the characteristics of the app surveyed, safety was most important to providers, with an average importance rating of 4.7 out of 5 (SD 0.72). This priority was followed by clinical integrity (a desire for the advice provided to make clinical sense; importance rating 4.5, SD 0.9), backward interpretability (transparency in the population used to create the algorithm; importance rating 4.3, SD 0.65), transparency of the algorithm (reasoning underlying the decisions made; importance rating 4.3, SD 0.88), and provider autonomy (the ability to challenge the decisions made by the model; importance rating 3.85, SD 0.83). Providers who used the app ranked the integrity of recommendations as their highest concern with ongoing clinical use of the model, followed by efficacy of the application and patient data security. Trust in the app varied; 1 (17%) provider responded that they somewhat disagreed with the statement, "I trust the recommendations provided by the QRhythm app," 2 (33%) providers responded with neutrality to the statement, and 3 (50%) somewhat agreed with the statement. CONCLUSIONS: Safety of ML applications was the highest priority of the providers surveyed, and trust of such models remains varied. Widespread clinical acceptance of ML in health care is dependent on how much providers trust the algorithms. Building this trust involves ensuring transparency and interpretability of the model.

Limitations of Unipolar Signals in Guiding Successful Outflow Tract Premature Ventricular Contraction Ablation.

BACKGROUND: Unipolar electrograms (UniEGMs) are commonly used to annotate earliest local activation of focal arrhythmias. However, their utility in guiding premature ventricular contractions (PVCs) ablation may be limited when the PVC source is less superficial. OBJECTIVES: The authors sought to compare bipolar electrograms (BiEGMs) vs UniEGMs in guiding successful ablation of right ventricular outflow tract (RVOT) vs intramural outflow tract (OT) PVCs. The authors hypothesized that: 1) earliest bipolar local activation time (LATBi) would better guide mapping and ablation, vs UniEGM dV/dt (LATUni) or QS morphology; and 2) LAT differences using bipolar vs unipolar EGMs (ΔLATBi-Uni) would be greater for intramural OT than RVOT PVCs. METHODS: Consecutive patients undergoing successful PVC ablation 2017 to2020 requiring only RVOT or RVOT+left ventricular OT (RVOT+LVOT) ablation were retrospectively analyzed. BiEGMs and UniEGMs at successful ablation sites were compared. RESULTS: Of 70 patients, 50 required RVOT-only, and 20 required RVOT+LVOT ablation for acute and long-term PVC suppression. Mean ΔLATBi-Uni was lower for RVOT vs RVOT+LVOT groups (9.3 ± 6.4 ms vs 17.4 ± 9.9 ms; P < 0.01). QS UniEGM was seen in 78% of RVOT, compared with 53% of RVOT+LVOT patients (P < 0.016). RVOT+LVOT sites most frequently included the posteroseptal RVOT and adjacent LVOT (73%), and 43% lacked a QS unipolar EGM. ΔLATBi-Uni ≥15 ms best distinguished sites in which RVOT-only vs RVOT+LVOT ablation achieved acute PVC suppression (area under the curve: 0.77). CONCLUSIONS: Earliest BiEGM activation guides successful ablation of OT PVCs better than UniEGM-guided analysis, especially when an intramural PVC source is present.

Assessment of a Mobile Health iPhone App for Semiautomated Self-management of Chronic Recurrent Medical Conditions Using an N-of-1 Trial Framework: Feasibility Pilot Study.

BACKGROUND: Management of chronic recurrent medical conditions (CRMCs), such as migraine headaches, chronic pain, and anxiety/depression, remains a major challenge for modern providers. Our team has developed an edge-based, semiautomated mobile health (mHealth) technology called iMTracker that employs the N-of-1 trial approach to allow self-management of CRMCs. OBJECTIVE: This study examines the patterns of adoption, identifies CRMCs that users selected for self-application, and explores barriers to use of the iMTracker app. METHODS: This is a feasibility pilot study with internet-based recruitment that ran from May 15, 2019, to December 23, 2020. We recruited 180 patients to pilot test the iMTracker app for user-selected CRMCs for a 3-month period. Patients were administered surveys before and after the study. RESULTS: We found reasonable usage rates: a total of 73/103 (70.9%) patients who were not lost to follow-up reported the full 3-month use of the app. Most users chose to use the iMTracker app to self-manage chronic pain (other than headaches; 80/212, 37.7%), followed by headaches in 36/212 (17.0%) and mental health (anxiety and depression) in 27/212 (12.8%). The recurrence rate of CRMCs was at least weekly in over 93% (169/180) of patients, with 36.1% (65/180) of CRMCs recurring multiple times in a day, 41.7% (75/180) daily, and 16.1% (29/180) weekly. We found that the main barriers to use were the design and technical function of the app, but that use of the app resulted in an improvement in confidence in the efficiency and safety/privacy of this approach. CONCLUSIONS: The iMTracker app provides a feasible platform for the N-of-1 trial approach to self-management of CRMCs, although internet-based recruitment provided limited follow-up, suggesting that in-person evaluation may be needed. The rate of CRMC recurrence was high enough to allow the N-of-1 trial assessment for most traits.

Nebivolol and incident cardiovascular events in hypertensive patients compared with nonvasodilatory beta blockers.

OBJECTIVE: Nonvasodilatory beta blockers are associated with inferior cardiovascular event reduction compared with other antihypertensive classes, and there is uncertainty about first-line use of beta blockers for hypertension in guidelines. The third generation vasodilatory beta blocker nebivolol has unique beneficial effects on central and peripheral vasculature. Our objective was to compare longitudinal cardiovascular outcomes of hypertensive patients taking nebivolol with those taking the nonvasodilatory beta blockers metoprolol and atenolol. METHODS: We performed a retrospective cohort analysis of hypertensive adults in the University of Colorado health system, without preceding diagnosis of cardiovascular or cerebrovascular disease. The primary outcome was composite incident heart failure, stroke, myocardial infarction, angina, or coronary revascularization. Mahalanobis 1:2 distance matching and Cox proportional hazards regression was used. Matching and regression variables included baseline demographics, socioeconomic factors, medical insurance type, prescribing provider type, cardiovascular risk factors, Charlson comorbidity index, other medications, and follow-up duration. RESULTS: After matching, patients were predominantly women (54%, 3085 of 5705) and non-Hispanic Caucasian (79%, 4534 of 5705), with median age of 58. In matched Cox regression analysis, nebivolol was associated with 17% reduction in incident cardiovascular events compared with all nonvasodilatory beta blockers [hazard ratio 0.83, 95% confidence interval (CI) 0.74-0.94, P  = 0.004], and 24% reduction compared with metoprolol (hazard ratio 0.76, CI 0.66-0.87, P = 0.0001). CONCLUSION: The vasodilatory beta blocker nebivolol was associated with reduced incident cardiovascular events compared with nonvasodilatory beta blockers. Additional study of other beta blockers is necessary to determine if this is a vasodilatory beta blocker class effect or is specific to nebivolol.http://links.lww.com/HJH/B916.

Increased incidence of cavotricuspid isthmus atrial flutter following slow pathway ablation.

PURPOSE: The incidence of atrial flutter following radiofrequency ablation of supraventricular tachycardias is poorly understood. Ablation of atrioventricular nodal reentry tachycardia may place patients at risk of flutter because ablation of the slow pathway is in close proximity to the cavotricuspid isthmus. This study aims to evaluate the risk of atrial flutter following ablation of atrioventricular nodal reentry tachycardia relative to ablation of other supraventricular tachycardias. METHODS: A single-center retrospective analysis was completed for all supraventricular tachycardia ablations performed between July 2006 and July 2016. Patient and procedural details were collected for 544 patients who underwent atrioventricular nodal reentry tachycardia ablation (n = 342), atrioventricular reentry tachycardia ablation (n = 125), or atrial tachycardia ablation (n = 60). Follow-up for flutter after ablation of their incident arrhythmia was assessed. RESULTS: Patients who underwent atrioventricular nodal reentry tachycardia ablation were more likely to develop CTI-dependent flutter than patients who underwent ablation of other supraventricular tachycardias (4.97% vs. 0%; p = 0.002). Compared with patients who did not develop flutter, patients who developed flutter after atrioventricular nodal reentry tachycardia ablation were more likely to have undergone ablation of atypical atrioventricular nodal reentry tachycardia (11.8% vs. 2.15%; p = 0.016). CONCLUSIONS: We identified an association between atrioventricular nodal reentry tachycardia ablation and development of CTI-dependent atrial flutter. This finding may have implications for the management and follow-up after atrioventricular nodal reentry tachycardia ablation.

Machine Learning Methodologies for Prediction of Rhythm-Control Strategy in Patients Diagnosed With Atrial Fibrillation: Observational, Retrospective, Case-Control Study.

BACKGROUND: The identification of an appropriate rhythm management strategy for patients diagnosed with atrial fibrillation (AF) remains a major challenge for providers. Although clinical trials have identified subgroups of patients in whom a rate- or rhythm-control strategy might be indicated to improve outcomes, the wide range of presentations and risk factors among patients presenting with AF makes such approaches challenging. The strength of electronic health records is the ability to build in logic to guide management decisions, such that the system can automatically identify patients in whom a rhythm-control strategy is more likely and can promote efficient referrals to specialists. However, like any clinical decision support tool, there is a balance between interpretability and accurate prediction. OBJECTIVE: This study aims to create an electronic health record-based prediction tool to guide patient referral to specialists for rhythm-control management by comparing different machine learning algorithms. METHODS: We compared machine learning models of increasing complexity and used up to 50,845 variables to predict the rhythm-control strategy in 42,022 patients within the University of Colorado Health system at the time of AF diagnosis. Models were evaluated on the basis of their classification accuracy, defined by the F1 score and other metrics, and interpretability, captured by inspection of the relative importance of each predictor. RESULTS: We found that age was by far the strongest single predictor of a rhythm-control strategy but that greater accuracy could be achieved with more complex models incorporating neural networks and more predictors for each participant. We determined that the impact of better prediction models was notable primarily in the rate of inappropriate referrals for rhythm-control, in which more complex models provided an average of 20% fewer inappropriate referrals than simpler, more interpretable models. CONCLUSIONS: We conclude that any health care system seeking to incorporate algorithms to guide rhythm management for patients with AF will need to address this trade-off between prediction accuracy and model interpretability.

Prediction of Drug-Induced Long QT Syndrome Using Machine Learning Applied to Harmonized Electronic Health Record Data.

BACKGROUND: Drug-induced QT prolongation is a potentially preventable cause of morbidity and mortality, however there are no widespread clinical tools utilized to predict which individuals are at greatest risk. Machine learning (ML) algorithms may provide a method for identifying these individuals, and could be automated to directly alert providers in real time. OBJECTIVE: This study applies ML techniques to electronic health record (EHR) data to identify an integrated risk-prediction model that can be deployed to predict risk of drug-induced QT prolongation. METHODS: We examined harmonized data from the UCHealth EHR and identified inpatients who had received a medication known to prolong the QT interval. Using a binary outcome of the development of a QTc interval >500 ms within 24 hours of medication initiation or no ECG with a QTc interval >500 ms, we compared multiple machine learning methods by classification accuracy and performed calibration and rescaling of the final model. RESULTS: We identified 35,639 inpatients who received a known QT-prolonging medication and an ECG performed within 24 hours of administration. Of those, 4,558 patients developed a QTc > 500 ms and 31,081 patients did not. A deep neural network with random oversampling of controls was found to provide superior classification accuracy (F1 score 0.404; AUC 0.71) for the development of a long QT interval compared with other methods. The optimal cutpoint for prediction was determined and was reasonably accurate (sensitivity 71%; specificity 73%). CONCLUSIONS: We found that deep neural networks applied to EHR data provide reasonable prediction of which individuals are most susceptible to drug-induced QT prolongation. Future studies are needed to validate this model in novel EHRs and within the physician order entry system to assess the ability to improve patient safety.

Assessing Prescriber Behavior with a Clinical Decision Support Tool to Prevent Drug-Induced Long QT Syndrome.

OBJECTIVE: Clinical decision support (CDS) alerts built into the electronic health record (EHR) have the potential to reduce the risk of drug-induced long QT syndrome (diLQTS) in susceptible patients. However, the degree to which providers incorporate this information into prescription behavior and the impact on patient outcomes is often unknown. METHODS: We examined provider response data over a period from October 8, 2016 until November 8, 2018 for a CDS alert deployed within the EHR from a 13-hospital integrated health care system that fires when a patient with a QTc ≥ 500 ms within the past 14 days is prescribed a known QT-prolonging medication. We used multivariate generalized estimating equations to analyze the impact of therapeutic alternatives, relative risk of diLQTS for specific medications, and patient characteristics on provider response to the CDS and overall patient mortality. RESULTS: The CDS alert fired 15,002 times for 7,510 patients for which the most common response (51.0%) was to override the alert and order the culprit medication. In multivariate models, we found that patient age, relative risk of diLQTS, and presence of alternative agents were significant predictors of adherence to the CDS alerts and that nonadherence itself was a predictor of mortality. Risk of diLQTS and presence of an alternative agent are major factors in provider adherence to a CDS to prevent diLQTS; however, provider nonadherence was associated with a decreased risk of mortality. CONCLUSION: Surrogate endpoints, such as provider adherence, can be useful measures of CDS value but attention to hard outcomes, such as mortality, is likely needed.