Faculty Bibliography

BACKGROUND:Because of the strong link between childhood obesity and adulthood obesity comorbidities, and the difficulty in decreasing body mass index (BMI) later in life, effective strategies are needed to address this condition in early childhood. The ability to predict obesity before age five could be a useful tool, allowing prevention strategies to focus on high risk children. The few existing prediction models for obesity in childhood have primarily employed data from longitudinal cohort studies, relying on difficult to collect data that are not readily available to all practitioners. Instead, we utilized real-world unaugmented electronic health record (EHR) data from the first two years of life to predict obesity status at age five, an approach not yet taken in pediatric obesity research. METHODS AND FINDINGS/RESULTS:We trained a variety of machine learning algorithms to perform both binary classification and regression. Following previous studies demonstrating different obesity determinants for boys and girls, we similarly developed separate models for both groups. In each of the separate models for boys and girls we found that weight for length z-score, BMI between 19 and 24 months, and the last BMI measure recorded before age two were the most important features for prediction. The best performing models were able to predict obesity with an Area Under the Receiver Operator Characteristic Curve (AUC) of 81.7% for girls and 76.1% for boys. CONCLUSIONS:We were able to predict obesity at age five using EHR data with an AUC comparable to cohort-based studies, reducing the need for investment in additional data collection. Our results suggest that machine learning approaches for predicting future childhood obesity using EHR data could improve the ability of clinicians and researchers to drive future policy, intervention design, and the decision-making process in a clinical setting.

Background/UNASSIGNED:Despite evidence that biological and genetic factors contribute strongly to obesity, many healthcare providers still attribute obesity more to controllable behavioral issues rather than factors outside a person's control. We evaluated whether medical school students' beliefs about obesity correlate with ability to effectively counsel patients with obesity. Methods/UNASSIGNED:Clerkship-year medical students at NYU School of Medicine completed an Objective Structured Clinical Experience (OSCE) that tests ability to effectively counsel standardized actor-patients with obesity. We surveyed these students to evaluate their beliefs about the causes of obesity and their attitudes towards people with obesity. We analyzed correlations between student beliefs, negative obesity attitudes, and OSCE performance. Results/UNASSIGNED: < 0.05). Conclusions/UNASSIGNED:Attribution of obesity to external factors correlated with greater ability to counsel patients with obesity, suggesting that educating providers on the biological causes of obesity could help reduce bias and improve provider care.

Increased investment in primary care is associated with lower healthcare costs and improved population health. The allocation of scarce resources should be driven by robust models that adequately describe primary care activities and spending within a health system, and allow comparisons within and across health systems. However, disparate definitions result in wide variations in estimates of spending on primary care. We propose a new model that allows for a dynamic assessment of primary care spending (PC Spend) within the context of a system's total healthcare budget. The model articulates varied definitions of primary care through a tiered structure which includes overall spending on primary care services, spending on services delivered by primary care professionals and spending delivered by providers that can be characterised by the '4Cs' (first contact, continuous, comprehensive and coordinated care). This unifying framework allows a more refined description of services to be included in any estimate of primary care spend and also supports measurement of primary care spending across nations of varying economic development, accommodating data limitations and international health system differences. It provides a goal for best accounting while also offering guidance, comparability and assessments of how primary care expenditures are associated with outcomes. Such a framework facilitates comparison through the creation of standard definitions and terms, and it also has the potential to foster new areas of research that facilitate robust policy analysis at the national and international levels.

INTRODUCTION:The United States Medical Licensing Examination (USMLE) Step 1 and Step 2 Clinical Knowledge (CK) are important for trainee medical knowledge assessment and licensure, medical school program assessment, and residency program applicant screening. Little is known about how USMLE performance varies between institutions. This observational study attempts to identify institutions with above-predicted USMLE performance, which may indicate educational programs successful at promoting students' medical knowledge. METHODS:Self-reported institution-level data was tabulated from publicly available US News and World Report and Association of American Medical Colleges publications for 131 US allopathic medical schools from 2012-2014. Bivariate and multiple linear regression were performed. The primary outcome was institutional mean USMLE Step 1 and Step 2 CK scores outside a 95% prediction interval (≥2 standard deviations above or below predicted) based on multiple regression accounting for students' prior academic performance. RESULTS:Eighty-nine US medical schools (54 public, 35 private) reported complete USMLE scores over the three-year study period, representing over 39,000 examinees. Institutional mean grade point average (GPA) and Medical College Admission Test score (MCAT) achieved an adjusted R2 of 72% for Step 1 (standardized βMCAT 0.7, βGPA 0.2) and 41% for Step 2 CK (standardized βMCAT 0.5, βGPA 0.3) in multiple regression. Using this regression model, 5 institutions were identified with above-predicted institutional USMLE performance, while 3 institutions had below-predicted performance. CONCLUSIONS:This exploratory study identified several US allopathic medical schools with significant above- or below-predicted USMLE performance. Although limited by self-reported data, the findings raise questions about inter-institutional USMLE performance parity, and thus, educational parity. Additional work is needed to determine the etiology and robustness of the observed performance differences.

Background: The growing opioid overdose epidemic has grappled the nation with the CDC now reporting that drug overdose deaths have become the most common cause of death for young people. Medical education has historically ignored substance use disorders, and though they generally require all medical students to learn basic life support, they have not taught how to respond to opioid overdoses. Further, medical education is moving towards modalities which utilize adult learning theory. One such modality are online modules. However, there are few studies comparing their outcomes with traditional lectures. Previously, the authors compared in-person and online training of medical students to respond to opioid overdoses using naloxone in a non-randomized controlled setting, which showed no meaningful differences in knowledge, attitudes, and preparedness outcomes for students. In this paper, the authors attempt to use a randomized controlled trial to compare the two educational modalities at a second urban medical school.
Objective(s): The author's primary objective was to demonstrate non-inferiority of online compared to in-person training for knowledge. Our secondary objective were to show non-inferiority of online compared to in-person training attitudes, and preparedness.
Method(s): Our study received IRB exemption as an education intervention. As a part of a transition to clinical clerkships curriculum used for second year medical students, second year medical students in an urban medical school were randomized into training sessions by the office of medical education without foreknowledge of the planned study. Students taking the online training were provided with a link to online modules with pre- and post-tests and video based lectures. Students randomized to the in-person training group took a pre-test just prior to receiving an oral lecture, and then immediately completed a post-test. Paired student's t-tests were used to compare measurements for each group in knowledge, attitudes, and preparedness, and Cohen's D was used to measure the effect size of the change. We calculated 99% confidence intervals for each measure and utilized a margin of non-inferiority of 5%.
Result(s): The in-person group demonstrated a statistically significant increase in knowledge, a non-statistically significant decrease in self-reported preparedness, and a small non-statistically significant increase in attitudes, see Table 1. The online group demonstrated a statistically significant increase in knowledge and self-reported preparedness, without a statistically significant change in attitudes, see Table 1. 99% CIs were [-0.20, 1.09] for knowledge, [6.51, 10.93] for preparedness, and [-2.32, 1.59] for attitudes, see Figure 1.
Conclusion(s): Online training for opioid overdose prevention training provided non-inferior outcomes for knowledge, preparedness, and attitudes. This study supports the use of online opioid overdose prevention training as a non-inferior alternative to in-person training

Reports have correlated the use of estrogen for the treatment of menopausal symptoms with beneficial effects on the cardiovascular system. Molecular, biochemical, preclinical, and clinical studies have furnished a wealth of evidence in support of this outcome of estrogen action. The prospective randomized Women's Health Initiative (WHI) and the Early Versus Late Intervention Trial (ELITE) showed that starting menopausal hormone treatment (MHT) within 5 to 10 years of menopause is fundamental to the success of estrogen's cardioprotection in post-menopausal women without adverse effects. Age stratification of the WHI data has shown that starting hormone treatment within the first decade after menopause is both safe and effective, and the long-term WHI follow-up studies are supportive of cardioprotection. This is especially true in estrogen-treated women who underwent surgical menopause. A critique of the WHI and other relevant studies is presented, supporting that the timely use of estrogens protects against age- and hormone-related cardiovascular complications. Salutary long-term hormone treatment for menopausal symptoms and prevention of complications has been widely reported, but there are no prospective trials defining the correct length to continue MHT. At present, women undergoing premature menopause receive estrogen treatment (ET) until evidence of hormone-related complications intervenes. Normal women started on MHT who receive treatment for decades without hormone-related complications have been reported, and the WHI follow-up studies are promising of long-term post-treatment cardioprotection. A prevention-based holistic approach is proposed for timely and continuing MHT/ET administration as part of the general management of the menopausal woman. But this should be undertaken only with scheduled, annual patient visits including evaluations of cardiovascular status. Because of the continued occurrence of reproductive cancers well into older ages, these visits should include genital and breast cancer screening.