PURPOSE: Although immune checkpoint inhibitors (ICIs) have improved outcomes in certain patients with cancer, they can also cause life-threatening immunotoxicities. Predicting immunotoxicity risks alongside response could provide a personalized risk-benefit profile, inform therapeutic decision making, and improve clinical trial cohort selection. We aimed to build a machine learning (ML) framework using routine electronic health record (EHR) data to predict hepatitis, colitis, pneumonitis, and 1-year overall survival. METHODS: Real-world EHR data of more than 2,200 patients treated with ICI through December 31, 2018, were used to develop predictive models. Using a prediction time point of ICI initiation, a 1-year prediction time window was applied to create binary labels for the four outcomes for each patient. Feature engineering involved aggregating laboratory measurements over appropriate time windows (60-365 days). Patients were randomly partitioned into training (80%) and test (20%) sets. Random forest classifiers were developed using a rigorous model development framework. RESULTS: The patient cohort had a median age of 63 years and was 61.8% male. Patients predominantly had melanoma (37.8%), lung cancer (27.3%), or genitourinary cancer (16.4%). They were treated with PD-1 (60.4%), PD-L1 (9.0%), and CTLA-4 (19.7%) ICIs. Our models demonstrate reasonably strong performance, with AUCs of 0.739, 0.729, 0.755, and 0.752 for the pneumonitis, hepatitis, colitis, and 1-year overall survival models, respectively. Each model relies on an outcome-specific feature set, though some features are shared among models. CONCLUSION: To our knowledge, this is the first ML solution that assesses individual ICI risk-benefit profiles based predominantly on routine structured EHR data. As such, use of our ML solution will not require additional data collection or documentation in the clinic.

Issue: Although there is consensus on the importance of including ethics in the medical school curriculum, there is wide variation in how this topic is taught. Recent literature also questions the effectiveness of current ethical teaching methods in changing student attitudes and future behavior. Furthermore, from the student perspective, there is a marked disconnect between the stated importance of and lack of effort in ethics courses. Evidence: Applying a student perspective of the hidden curriculum, as well as reviewing and applying insight from the available literature, we advocate for alignment of instructional design, content, and assessments. This article provides specific recommendations to increase student engagement in ethics courses and concludes by discussing whether a lack of engagement is attributable to intrinsic qualities of medical students in addition to pedagogical technique and educational setting and culture. Implications: This article has practical suggestions for medical educators to improve their ethics courses, leading to more well-rounded and thoughtful physicians.

Objective: To quantify the changes in prevalence and impact of dizziness and balance disorders in adults from 2008 to 2016.Study Design: Epidemiological survey analysis.Setting: United States.Methods: The balance modules of the adult 2008 and 2016 National Health Interview Surveys were examined, and persons reporting dizziness or balance problems identified. The prevalence of balance problems was determined and compared over time, adjusting for age and sex. Among those with balance problems, associated symptoms and self‐reported functional limitations were quantified and compared over time.Results: In 2016, 36.8 ± 1.0 million (15.5% ± 0.3%) adults reported a balance problem in the past year, versus 24.2 ± 0.7 million (11.% ± 0.3%) in 2008 (p < .001). After adjustment for age and sex, this percentage increase remained significant (odds ratio 1.435 [1.332‐1.546], p < .001). Among those with balance problems, significantly more patients reported specific issues with feeling: off‐balance (69.4% vs. 65.4%; p = .005), faint (48.5% vs. 40.3%; p < .001), or vertiginous (45.9% vs. 39.3%; p < .001) in 2016 than 2008. More adults experienced anxiety (29.4% vs. 19.4%; p < .001) and depression (16.3% vs. 12.9%; p = .002) with their balance problems in 2016 than in 2008. In 2016, adults with balance problems were limited in ability to drive motor vehicles (13.0%), exercise (14.4%), or walk downstairs (12.8%). These rates were not significantly different from 2008 (all p > .05).Conclusion: In this nationally representative analysis, we found a significantly increasing prevalence of balance problems and associated psychiatric symptom burden. This merits attention with respect to present and future health care resource allocation.