Machine Learning for The Prediction of Ranked Applicants and Matriculants to an Internal Medicine Residency Program.

Phenomenon: Residency programs throughout the country each receive hundreds to thousands of applications every year. Holistic review of this many applications is challenging, and to-date, few tools exist to streamline or assist in the process for selecting candidates to interview and rank. Machine learning could assist programs in predicting which applicants are likely to be ranked, and among ranked applicants, which are likely to matriculate.Approach: In the present study, we used the machine learning algorithm Random Forest (RF) to differentiate between ranked and unranked applicants as well as matriculants and ranked non-matriculants to an internal medicine residency program in northern New England over a three-year period. In total, 5,067 ERAS applications were received during the 2016-17, 2017-18, and 2018-19 application cycles. Of these, 4,256 (84.0%) were unranked applicants, 754 (14.9%) were ranked non-matriculants, and 57 (1.12%) were ranked matriculants.Findings: For differentiating between ranked and unranked applicants, the RF algorithm achieved an area under the receiver operating characteristic (AUROC) curve of 0.925 (95% CI: 0.918-0.932) and area under the precision-recall curve (AUPRC) of 0.652 (0.611-0.685), while for differentiating between matriculants and ranked non-matriculants, the AUROC was 0.597 (95% CI: 0.516-0.680) and AUPRC was 0.114 (0.075-0.167). The ranks of matriculated applicants were significantly higher using the algorithmic rank list as compared with the actual rank list for the 2017-18 (median rank: 98 versus 204, p < .001) and 2018-19 cycles (74 versus 192, p = .006), but not the 2016-17 cycle (97 versus 144, p = .37).Insights: The RF algorithm predicted which applicants among the overall applicant pool were ranked with impressive accuracy and identified matriculants among ranked candidates with modest but better-than-random accuracy. This approach could assist residency programs with triaging applicants based on the likelihood of a candidate being ranked and/or matriculating.

Achieving educational mission and vision with an educational scorecard.

BACKGROUND: Achieving an academic section's educational mission and vision is difficult, particularly when individual faculty contributions may be self-directed and uncoordinated. Balanced scorecards have been used in other environments; however, a process for developing one focusing on the educational mission of an academic medical section has not previously been described. We aimed to develop and use an educational scorecard to help our academic clinical section achieve its educational mission and vision. METHODS: Six medical educators participated in a task force that developed, implemented, and evaluated an educational scorecard that incorporates four domains of educational value and six stakeholder perspectives. A modified Delphi process using 14 experts built expert consensus on the most valuable metrics. The task force then developed performance targets for each metric. RESULTS: Review of the scorecard at the sectional level resulted in both sectional and individual strategies which lead to a more balanced educational impact, including service structure changes and increased mentorship. Our section has used the scorecard and metrics to evaluate performance since 2014. CONCLUSION: An educational scorecard is a feasible way for academic groups to communicate educational goals, engage faculty, and provide objective information with which to base strategic decisions affecting their educational mission.

Identifying protective and risk factors for injurious falls in patients hospitalized for acute care: a retrospective case-control study.

BACKGROUND: Admitted patients who fall and injure themselves during an acute hospitalization incur increased costs, morbidity, and mortality, but little research has been conducted on identifying inpatients at high risk to injure themselves in a fall. Falls risk assessment tools have been unsuccessful due to their low positive predictive value when applied broadly to entire hospital populations. We aimed to identify variables associated with the risk of or protection against injurious fall in the inpatient setting. We also aimed to test the variables in the ABCs mnemonic (Age > 85, Bones-orthopedic conditions, anti-Coagulation and recent surgery) for correlation with injurious fall. METHODS: We performed a retrospective case-control study at an academic tertiary care center comparing admitted patients with injurious fall to admitted patients without fall. We collected data on the demographics, medical and fall history, outcomes, and discharge disposition of injured fallers and control patients. We performed multivariate analysis of potential risk factors for injurious fall with logistic regression to calculate adjusted odds ratios. RESULTS: We identified 117 injured fallers and 320 controls. There were no differences in age, anti-coagulation use or fragility fractures between cases and controls. In multivariate analysis, recent surgery (OR 0.46, p = 0.003) was protective; joint replacement (OR 5.58, P = 0.002), psychotropic agents (OR 2.23, p = 0.001), the male sex (OR 2.08, p = 0.003) and history of fall (OR 2.08, p = 0.02) were significantly associated with injurious fall. CONCLUSION: In this study, the variables in the ABCs parameters were among the variables not useful for identifying inpatients at risk of injuring themselves in a fall, while other non-ABCs variables demonstrated a significant association with injurious fall. Recent surgery was a protective factor, and practices around the care of surgical patients could be extrapolated to reduce the in-hospital fall rates.

Abstracts From the Proceedings of the 2015 Annual Meeting of the Clerkship Directors of Internal Medicine (CDIM).

Since its inception in 1989, Clerkship Directors in Internal Medicine (CDIM) has promoted excellence in medical student education. CDIM members move medical education forward by sharing innovations in curriculum and assessment and discoveries related to educating our students and administering our programs. The Alliance for Academic Internal Medicine, of which CDIM is a founding member, broadens the umbrella beyond student education to include five academically focused specialty organizations representing departments of medicine, teaching hospitals, and medical schools working together to advance learning, discovery, and caring. CDIM held its 2015 annual meeting at Academic Internal Medicine Week in Atlanta, Georgia. This year 36 innovation and research submissions were selected for either oral abstract or poster presentation. The quality of the presentations was outstanding this year and included many of the most important issues in medical education. The CDIM research committee selected the following seven abstracts as being of the highest quality, the most generalizable, and relevant to the readership of Teaching and Learning in Medicine. Two abstracts include information from the CDIM annual survey, which remains a rich source for answering questions about student education on a national level. Looking at trends in medical education, three of the seven selected abstracts mention entrustable professional activities. Three of the abstracts address how we assess student skill and provide them with appropriate feedback. These include two schools' approach to bringing milestones into the medical student realm, use of objective structured clinical exam for assessing clinical skill in clerkship, and what students want in terms of feedback. Four articles deal with curricular innovation. These include interprofessional education, high-value care, transitions of care, and internship preparation. We are pleased to share these abstracts, which represent the breadth and quality of thought of our CDIM members.

Does what we write matter? Determining the features of high- and low-quality summative written comments of students on the internal medicine clerkship using pile-sort and consensus analysis: a mixed-methods study.

BACKGROUND: Written comments by medical student supervisors provide written foundation for grade narratives and deans' letters and play an important role in student's professional development. Written comments are widely used but little has been published about the quality of written comments. We hypothesized that medical students share an understanding of qualities inherent to a high-quality and a low-quality narrative comment and we aimed to determine the features that define high- and low-quality comments. METHODS: Using the well-established anthropological pile-sort method, medical students sorted written comments into 'helpful' and 'unhelpful' piles, then were interviewed to determine how they evaluated comments. We used multidimensional scaling and cluster analysis to analyze data, revealing how written comments were sorted across student participants. We calculated the degree of shared knowledge to determine the level of internal validity in the data. We transcribed and coded data elicited during the structured interview to contextualize the student's answers. Length of comment was compared using one-way analysis of variance; valence and frequency comments were thought of as helpful were analyzed by chi-square. RESULTS: Analysis of written comments revealed four distinct clusters. Cluster A comments reinforced good behaviors or gave constructive criticism for how changes could be made. Cluster B comments exhorted students to continue non-specific behaviors already exhibited. Cluster C comments used grading rubric terms without giving student-specific examples. Cluster D comments used sentence fragments lacking verbs and punctuation. Student data exhibited a strong fit to the consensus model, demonstrating that medical students share a robust model of attributes of helpful and unhelpful comments. There was no correlation between valence of comment and perceived helpfulness. CONCLUSIONS: Students find comments demonstrating knowledge of the student and providing specific examples of appropriate behavior to be reinforced or inappropriate behavior to be eliminated helpful, and comments that are non-actionable and non-specific to be least helpful. Our research and analysis allow us to make recommendations helpful for faculty development around written feedback.

Investigating Medical Students' Navigation of Ethical Dilemmas: Understanding the Breakdown and How to Solve It.

PURPOSE: Medical students receive a varying amount of training in medical ethics and are expected to navigate clinical ethical dilemmas innately. There is little literature on attempts to navigate ethical dilemmas experienced during early clinical experiences and whether current curricula prepare students for these dilemmas. This study explores the different ethical dilemmas experienced by medical students on their third-year clerkships and analyzes the factors, sources, and resolutions proposed by them. METHODS: From 2016 to 2018, third-year medical students completed a written assignment to describe, analyze, and reflect on a clinical situation in which they experienced an ethical dilemma. They identified specific ethical dilemmas present, potential preventative and aftermath solutions, and reflected on their professional development from their experience. The research team utilized applied thematic analysis to identify themes and patterns in the data. A thematic matrix was utilized to examine similarities and differences across medical students. RESULTS: Of the 162 reflections, 144 (88.9%) students indicated an ethical dilemma that included issues related to autonomy and beneficence. Of these, 116 (71.6%) students found the two ethical principles in direct conflict. Students identified three common sources of this conflict: lack of communication; unclear understanding of clinical policies regarding family authority and psychiatric capacity; and medical negligence. Lastly, students suggested different solutions for dealing with and preventing this conflict. CONCLUSION: Our findings suggest that an overwhelming number of students face ethical challenges when confronted with medical situations that raise conflicts between autonomy and beneficence. Their recommended solutions reveal an appeal among students to have tools and strategies in place to ease the need to make difficult decisions. Medical students might be better served by learning about the complexities of ethical decision-making and the likelihood of experiencing moral distress when they feel an inability to implement what they envision as the best solution.

Is It All About the Form? Norm- vs Criterion-Referenced Ratings and Faculty Inter-Rater Reliability.

Background: Little research to date has examined the quality of data obtained from resident performance evaluations. This study sought to address this need and compared inter-rater reliability obtained from norm-referenced and criterion-referenced evaluation scaling approaches for faculty completing resident performance evaluations. Methods: Resident performance evaluation data were examined from 2 institutions (3 programs, 2 internal medicine and 1 surgery; 426 residents in total), with 4 evaluation forms: 2 criterion-referenced (1 with an additional norm-referenced item) and 2 norm-referenced. Faculty inter-rater reliability was calculated with intraclass correlation coefficients (ICCs) (1,10) for each competency area within the form. ICCs were transformed to z-scores, and 95% CIs were computed. Reliabilities for each evaluation form and competency, averages within competency, and averages within scaling type were examined. Results: Inter-rater reliability averages were higher for all competencies that used criterion-referenced scaling relative to those that used norm-referenced scaling. Aggregate scores of all independent categories (competencies and the items assessing overall competence) for criterion-referenced scaling demonstrated higher reliability (z=1.37, CI 1.26-1.48) than norm-referenced scaling (z=0.88, CI 0.77-0.99). Moreover, examination of the distributions of composite scores (average of all competencies and raters for each individual being rated) suggested that the criterion-referenced evaluations better represented the performance continuum. Conclusion: Criterion-referenced evaluation approaches appear to provide superior inter-rater reliability relative to norm-referenced evaluation scaling approaches. Although more research is needed to identify resident evaluation best practices, using criterion-referenced scaling may provide more valid data than norm-referenced scaling.

Maximizing Career Advancement During the COVID-19 Pandemic: Recommendations for Postgraduate Training Programs.

The ongoing novel Coronavirus disease 2019 (COVID-19) pandemic has created many threats as well as opportunities for the career development of physicians-in-training. Institutional responses to the demand for patient care reduced the time many residents have to pursue clinical electives, scholarship projects, and other experiences meant to clarify and advance their personal and professional goals. Moreover, many academic medical centers experienced profound fiscal losses that require thoughtful revisions to budgets and curricula. In this article, the authors recommend strategies for residency programs to mitigate these losses and capitalize on growth in virtual education, scholarship opportunities, and relationships arising from the pandemic. Drawing from career development guidelines from the National Career Development Association and existing literature about factors associated with positive career outcomes, the authors suggest leadership roles, curricula, and events that training programs can quickly and inexpensively implement to help residents grow as professionals, achieve personal training goals, produce scholarship, and attain future employment. To help trainees manage their careers, the authors recommend structured mentorship and education in career pathways and the preparation of job application materials. To foster attainment of specific knowledge and cultivate lifelong learning, the authors recommend leveraging existing resources to provide time, funding, academic coaching, and skills training for scholarship projects. To promote development of effective work relationships and community, the authors recommend appointment of a faculty champion for career advancement, scholarship showcases, attendance at virtual journal clubs, and networking through social media outlets. These recommendations for supporting career advancement may apply to early career faculty development as well as undergraduate and postgraduate medical education beyond the pandemic era. Outcomes studies will be needed to evaluate the impact of these recommendations.

Macronuclear genome sequence of the ciliate Tetrahymena thermophila, a model eukaryote.

The ciliate Tetrahymena thermophila is a model organism for molecular and cellular biology. Like other ciliates, this species has separate germline and soma functions that are embodied by distinct nuclei within a single cell. The germline-like micronucleus (MIC) has its genome held in reserve for sexual reproduction. The soma-like macronucleus (MAC), which possesses a genome processed from that of the MIC, is the center of gene expression and does not directly contribute DNA to sexual progeny. We report here the shotgun sequencing, assembly, and analysis of the MAC genome of T. thermophila, which is approximately 104 Mb in length and composed of approximately 225 chromosomes. Overall, the gene set is robust, with more than 27,000 predicted protein-coding genes, 15,000 of which have strong matches to genes in other organisms. The functional diversity encoded by these genes is substantial and reflects the complexity of processes required for a free-living, predatory, single-celled organism. This is highlighted by the abundance of lineage-specific duplications of genes with predicted roles in sensing and responding to environmental conditions (e.g., kinases), using diverse resources (e.g., proteases and transporters), and generating structural complexity (e.g., kinesins and dyneins). In contrast to the other lineages of alveolates (apicomplexans and dinoflagellates), no compelling evidence could be found for plastid-derived genes in the genome. UGA, the only T. thermophila stop codon, is used in some genes to encode selenocysteine, thus making this organism the first known with the potential to translate all 64 codons in nuclear genes into amino acids. We present genomic evidence supporting the hypothesis that the excision of DNA from the MIC to generate the MAC specifically targets foreign DNA as a form of genome self-defense. The combination of the genome sequence, the functional diversity encoded therein, and the presence of some pathways missing from other model organisms makes T. thermophila an ideal model for functional genomic studies to address biological, biomedical, and biotechnological questions of fundamental importance.

Fatal aortic dissection due to a fulminant variety of isolated aortitis.

Aortitis is typically a chronic, progressive disease manifestation associated with large vessel vasculitidies, most notably giant cell, Takayasu arteritis, and a newly described entity, isolated aortitis. The aortitis may lead to aneurysm formation and symptoms associated with branch vessel occlusion in these diseases, but aortic dissection is rare and usually a late complication of smoldering, incompletely treated disease. We present a case of aortitis in a previously healthy 39-year-old man who succumbed to aortic dissection hours after the onset of symptoms. No aneurysm or fibrosis was found on postmortem examination. The inflammation was characterized by disruption of the media with patchy transmural chronic and focally acute inflammatory infiltrate. We review case reports of other individuals with aortitis, who initially or very early in their course presented with aortic dissection in the absence of known rheumatic disease and most without evidence of aneurysm formation. We believe that this represents a process characterized by an aggressive vasculitis of the aorta with its own clinical features, a fulminant variety of isolated aortitis.

Decision Analysis and Cost-effectiveness Analysis.

Healthcare decision-making can be complex, often requiring decision makers to weigh serious trade-offs, consider patients' values, and incorporate evidence in the face of uncertainty. Medical decisions are made implicitly by clinicians and other decision-makers on a daily basis. Decisions based largely on personal experience are subject to many biases. Decision analysis and cost-effectiveness analysis are systematic approaches used to support decision-making under conditions of uncertainty that involve important trade-offs. These mathematical tools can provide patients, physicians and policy makers with a useful approach to complex medical decision making.

What Do I Do When Something Goes Wrong? Teaching Medical Students to Identify, Understand, and Engage in Reporting Medical Errors.

PROBLEM: Identifying and processing medical errors are overlooked components of undergraduate medical education. Organizations and leaders advocate teaching medical students about patient safety and medical error, yet few feasible examples demonstrate how this teaching should occur. To provide students with familiarity in identifying, reporting, and analyzing medical errors, the authors developed the interactive patient safety reporting curriculum (PSRC), requiring clinical students to engage intellectually and emotionally with personally experienced events in which the safety of one of their patients was compromised. APPROACH: In 2015, the authors incorporated the PSRC into the third-year internal medicine clerkship. Students completed a structured written report, analyzing a patient safety incident they experienced. The report focused on severity of outcome, root cause(s) analysis, system-based prevention, and personal reflection. The report was bookended by 2 interactive, case-based sessions led by faculty with expertise in patient safety, quality improvement, and medical errors. OUTCOMES: Students accurately analyzed the severity of the outcome, and their reports directly led to 2 formal root cause analyses and 4 system-based improvements. NEXT STEPS: The time- and resource-efficient PSRC allows students to apply patient safety knowledge to a medical error they experienced in a way that can directly affect care delivery. This model-interactive learning sessions coupled with engaging in a personally experienced case-can be implemented in various settings. Educators seeking to use student-experienced events for learning should not discount the emotional effects of those events on medical students.

Metabolic acidosis after sodium thiosulfate infusion and the role of hydrogen sulfide.

Sodium thiosulfate (STS), first-line treatment for calcific uremic arteriolopathy, causes a mild asymptomatic acidosis in many patients. However, severe, life-threatening acidosis out of proportion with the expected acid load of STS may occur, potentially due to metabolism of STS to hydrogen sulfide.

Clinical Prediction Rule for Patient Outcome after In-Hospital CPR: A New Model, Using Characteristics Present at Hospital Admission, to Identify Patients Unlikely to Benefit from CPR after In-Hospital Cardiac Arrest.

BACKGROUND: Physicians and patients frequently overestimate likelihood of survival after in-hospital cardiopulmonary resuscitation. Discussions and decisions around resuscitation after in-hospital cardiopulmonary arrest often take place without adequate or accurate information. METHODS: We conducted a retrospective chart review of 470 instances of resuscitation after in-hospital cardiopulmonary arrest. Individuals were randomly assigned to a derivation cohort and a validation cohort. Logistic Regression and Linear Discriminant Analysis were used to perform multivariate analysis of the data. The resultant best performing rule was converted to a weighted integer tool, and thresholds of survival and nonsurvival were determined with an attempt to optimize sensitivity and specificity for survival. RESULTS: A 10-feature rule, using thresholds for survival and nonsurvival, was created; the sensitivity of the rule on the validation cohort was 42.7% and specificity was 82.4%. In the Dartmouth Score (DS), the features of age (greater than 70 years of age), history of cancer, previous cardiovascular accident, and presence of coma, hypotension, abnormal PaO2, and abnormal bicarbonate were identified as the best predictors of nonsurvival. Angina, dementia, and chronic respiratory insufficiency were selected as protective features. CONCLUSIONS: Utilizing information easily obtainable on admission, our clinical prediction tool, the DS, provides physicians individualized information about their patients' probability of survival after in-hospital cardiopulmonary arrest. The DS may become a useful addition to medical expertise and clinical judgment in evaluating and communicating an individual's probability of survival after in-hospital cardiopulmonary arrest after it is validated by other cohorts.

Educating adolescents about acne vulgaris: a comparison of written handouts with audiovisual computerized presentations.

OBJECTIVE: To compare the efficacy of written handouts with that of audiovisual computerized presentations in educating adolescents about acne vulgaris. DESIGN: Randomized study. SETTING: A private dermatology office or 1 of 3 general pediatric clinics in New Haven. PARTICIPANTS: One hundred one adolescent patients, aged 13 to 17 years. Intervention All participants completed a brief enrollment questionnaire to gauge baseline knowledge of acne vulgaris. Subjects were then randomized to either receive a written handout or watch an audiovisual computerized presentation. Immediately following the intervention, and again at 1 month, patients were asked to complete identical questionnaires to assess change in knowledge. MAIN OUTCOME MEASURES: Change in knowledge about acne vulgaris, as indicated by performance on preintervention, postintervention, and 1-month follow-up questionnaires. RESULTS: Baseline questionnaires were completed by 21 patients in the pilot study and 80 patients in the revised study; 17 (81%) and 77 (96%) completed the respective studies. In both the pilot (P = .64) and revised (P = .63) studies, there was no significant difference between intervention groups in terms of baseline knowledge or gain in knowledge. Immediately postintervention, both groups showed significant improvement from baseline (P < .001 in the revised study and P < .01 in the pilot study). At the 1-month follow-up, patients in the pilot study randomized to receive the computerized presentation still showed significant gain in knowledge from baseline (P < .05), while those in the handout group did not. Meanwhile, both intervention groups in the revised study continued to show significant gain in knowledge from baseline at 1 month (P < .001). Conclusion Both written handouts and audiovisual computerized presentations about acne vulgaris confer significant and equivalent benefits in terms of short- and long-term knowledge gains among adolescent patients with acne.

Nevus anelasticus, papular elastorrhexis, and eruptive collagenoma: clinically similar entities with focal absence of elastic fibers in childhood.

Focal absence of elastic fibers is a defining characteristic of three rare, clinically similar conditions: nevus anelasticus, papular elastorrhexis, and eruptive collagenoma. We present a 16-year-old with nevus anelasticus treated with intralesional steroid injections, resulting in erythematous atrophic depressions. We also review the English language literature on nevus anelasticus, papular elastorrhexis, and eruptive collagenoma. There may be sufficient clinical and histopathologic similarity to indicate that these three entities represent a single disease or disease spectrum. We believe that this entity should be referred to as papular elastorrhexis.