A national task force convened by the NKF-ASN recently recommended a new race-free creatinine equation for calculating eGFR. Although this equation is expected to be widely adopted, its broad effect on recommended clinical care across the eGFR spectrum and across different racial and ethnic groups is not known. The authors used nationally representative data from 44,360 participants in NHANES to quantify expected changes to recommended care. They found that nationwide implementation of the new creatinine-based eGFR equation may affect recommended care for hundreds of thousands of Black adults and millions of non-Black adults, including new CKD diagnoses and reversals, CKD stage reclassifications, and changes in kidney donation eligibility, nephrologist referral, and medication dosing.

Innovations in robotics, virtual and augmented reality, and artificial intelligence are being rapidly adopted as tools of "digital surgery". Despite its quickly emerging role, digital surgery is not well understood. A recent study defines the term itself, and then specifies ethical issues specific to the field. These include privacy and public trust, consent, and litigation.

Artificial intelligence (AI) tools for endoscopy are now entering clinical practice after demonstrating substantial improvements to polyp detection on colonoscopy. As this technology continues to mature, efforts to develop and validate a new frontier of possibilities—including diagnostic classification, risk stratification, and clinical outcomes assessment—are now underway. In npj Digital Medicine, scientists from Cosmo AI/Linkverse and collaborators report an extension to the first FDA-cleared AI tool for colonoscopy that goes beyond polyp detection to enable video-based diagnostic characterization.

In 1999, researchers introduced a Black race coefficient of 1.21 to the estimated glomerular filtration rate (eGFR) on the basis of the observation that participants who self-identified as Black had a 21% higher measured GFR after controlling for age, sex, and serum creatinine than those who did not self-identify as Black. Use of this coefficient mitigated underestimation bias among Black individuals and overestimation bias among non-Black individuals in the study population, but did not consider confounding from socioeconomic and structural factors, problematic views of race as biology, or potential implications of race-based differential treatment. The subsequent 2009 Chronic Kidney Disease Epidemiology Collaboration equation similarly derived a Black race coefficient of 1.16. In 2021, the recommendation by the joint National Kidney Foundation (NKF) and American Society of Nephrology (ASN) Task Force to adopt calculations of eGFR without race sent waves through the medical community. Consensus recognition that race-adjusted eGFR causes harm marked a positive and long-needed step forward for nephrology and the greater medical community.

Patients, transdisciplinary experts, and medical trainees contributed in key ways to this movement through grassroots organizing, outreach and education, scientific research and scholarly discourse, and policy and organized medicine. Yet, these perspectives are often excluded or erased from dominant narratives. In this article, writing as trainees ourselves, we recognize the foundational social science and activism upon which the movement to redress race-adjusted eGFR was built, explore motivations for trainee involvement, highlight trainee and patient contributions, and outline next steps to advance equity in kidney care and beyond.

With the increasing number of FDA-approved artificial intelligence (AI) systems, the financing of these technologies has become a primary gatekeeper to mass clinical adoption. Reimbursement models adapted for current payment schemes, including per-use rates, are feasible for early AI products. Alternative and complementary models may offer future payment options for value-based AI. A successful reimbursement strategy will align interests across stakeholders to guide value-based and cost-effective improvements to care.

Parkinson’s disease (PD) lacks sensitive, objective, and reliable measures for disease progression and response. This presents a challenge for clinical trials given the multifaceted and fluctuating nature of PD symptoms. Innovations in digital health and wearable sensors promise to more precisely measure aspects of patient function and well-being. Beyond research trials, digital biomarkers and clinical outcome assessments may someday support clinician-initiated or closed-loop treatment adjustments. A recent study from Verily Life Sciences presents results for a smartwatch-based motor exam intended to accelerate the development and evaluation of therapies for PD.

Due to its enormous capacity for benefit, harm, and cost, health care is among the most tightly regulated industries in the world. But with the rise of smartphones, an explosion of direct-to-consumer mobile health applications has challenged the role of centralized gatekeepers. As interest in health apps continue to climb, national regulatory bodies have turned their attention toward strategies to protect consumers from apps that mine and sell health data, recommend unsafe practices, or simply do not work as advertised. To characterize the current state and outlook of these efforts, Essén and colleagues map the nascent landscape of national health app policies and raise several considerations for cross-border collaboration. Strategies to increase transparency, organize app marketplaces, and monitor existing apps are needed to ensure that the global wave of new digital health tools fulfills its promise to improve health at scale.

As clinicians and scientists gather more data on the clinical trajectory of COVID-19 and the biology of its causative agent, the SARS-CoV-2 virus, novel strategies are needed to integrate these data to inform new therapies. A recent study by Howell et al. introduces a network model of viral-host interactions to produce explainable and testable predictions for treatment effects. Their model was consistent with experimental data and recommended treatments, and one of its predicted drug combinations was validated through in vitro assays. These findings support the utility of computational strategies for leveraging the vast literature on COVID-19 to generate insights for drug repurposing.

The vital signs—temperature, heart rate, respiratory rate, and blood pressure—are indispensable in clinical decision-making. These metrics are widely used to identify physiologic decline and prompt investigation or intervention. Vital sign monitoring is particularly important in acute care settings, where patients are at higher risk and may require additional vigilance. Conventional contact-based devices, while widespread and generally reliable, can be inconvenient or disruptive to patients, families, and staff. Non-contact, video-based methods present a more flexible and information-dense alternative that may enable creative improvements to patient care. Still, these approaches are susceptible to several sources of bias and require rigorous clinical validation. A recent study by Jorge et al. demonstrates that video-based monitoring can reliably capture heart rate and respiratory rate and overcome many potential sources of bias in post-operative settings. This presents real-world evaluation of a practical, noninvasive, and continuous monitoring technology that had previously only been tested in controlled settings.

In their recent study, Blumel and colleagues recommend that “the thumb-palm test be part of the standard physical examination” because “patients who do have a positive sign have a very high likelihood of harboring an ascending aneurysm.” Using Bayes theorem, we computed post-test probability to be 0.096% for instantaneous risk, implying more than 1000 false positives for every true positive. Applying the same calculations to autopsy-based studies, we computed post-test probability to be 1.0% for lifetime risk. For each patient who tests positive and later develops an ascending aortic aneurysm, there may be up to 100 others who never do in their lifetimes. We recommend that clinicians who use the thumb-palm test account for population base rates when interpreting positive results, and that researchers evaluating screening methods estimate positive predictive value in the populations for which they are recommended.

In recent years, a steady swell of biological image data has driven rapid progress in healthcare applications of computer vision and machine learning. To make sense of this data, scientists often rely on detailed annotations from domain experts for training artificial intelligence (AI) algorithms. The time-consuming and costly process of collecting annotations presents a sizable bottleneck for AI research and development. HALS (Human-Augmenting Labeling System) is a collaborative human-AI labeling workflow that uses an iterative “review-and-revise” model to improve the efficiency of this critical process in computational pathology.

Nearly half of US adults have hypertension, and three in four cases are not well-controlled. Due to structural barriers, underserved communities face greater burdens of disease, less consistent management, and worse outcomes. Mobile technology presents an opportunity to reduce financial, geographic, and workforce barriers, but little data currently support its use in populations with digital disparities. A recent article by Khoong et al. systematically reviews the literature to quantify outcomes for these populations and provide a roadmap toward more inclusive mobile health strategies.

Advances in medical machine learning are expected to help personalize care, improve outcomes, and reduce wasteful spending. In quantifying potential benefits, it is important to account for constraints arising from clinical workflows. Practice variation is known to influence the accuracy and generalizability of predictive models, but its effects on cost-effectiveness and utilization are less well-described. A simulation-based approach by Mišić and colleagues goes beyond simple performance metrics to evaluate how process variables may influence the impact and financial feasibility of clinical prediction algorithms.

Medical centers nationwide are considering race-free equations to improve the reporting of eGFR. Numerous alternatives have been proposed and implemented—including removal of the coefficient for Black race—with substantial implications for guideline-recommended care. Despite the abundance of race-free alternatives, it is unclear how their eGFR distributions compare with that of eGFRcr and eGFRcr-cys. Population-level differences estimated from a representative national sample may indicate shifts in health care access and utilization to consider alongside accuracy comparisons on the basis of mGFR.

In response to our study on the implications of removing race from eGFR, Ms. van der Burgh and colleagues describe advantages of eGFRcys, including potentially improved prediction of cardiovascular events and mortality in patients with CKD. We agree that eGFRcys should be evaluated alongside other race-free options. We have reproduced our analyses using 2 waves of the National Health and Nutrition Examination Survey (1999-2002) containing serum cystatin C measurements, with the caveat that these data represent fewer participants and an earlier time period than examined in our study.

Images of dark skin are routinely underrepresented in medical education and research. This lack of representation may influence diagnostic ability among clinicians and health outcomes for patients of color. Using over 40 million responses from the weekly multiple-choice New England Journal of Medicine Image Challenge, we aimed to investigate the differences in representation and image-based diagnostic accuracy by skin type. 

We present an approach for predicting clinically-relevant molecular phenotypes from whole-slide histopathology images using human-interpretable image features (HIFs). Our method leverages >1.6 million annotations from board-certified pathologists across >5700 samples to train deep learning models for cell and tissue classification that can exhaustively map whole-slide images at two and four micron-resolution. Cell- and tissue-type model outputs are combined into 607 HIFs that quantify specific and biologically-relevant characteristics across five cancer types. We demonstrate that these HIFs correlate with well-known markers of the tumor microenvironment and can predict diverse molecular signatures (AUROC 0.601-0.864), including expression of four immune checkpoint proteins and homologous recombination deficiency, with performance comparable to 'black-box' methods. Our HIF-based approach provides a comprehensive, quantitative, and interpretable window into the composition and spatial architecture of the tumor microenvironment.

Although many experts agree that we should reconsider the use of race in equations for estimated glomerular filtration rate (eGFR) and in medicine more generally, precisely how eGFR equations should remove race remains unclear. We documented race-free alternatives with respect to validation, overall and within-group accuracy, availability of assays and equation parameters, representation of Black patients in development data, and use of race. The potential implications for millions of patients necessitate a thorough consideration of these factors in the search for a better equation.

Over the past year, medical centers across the US have removed race adjustment from estimated glomerular filtration rate from serum creatinine (eGFRcr), with many now reporting the “White/other” value for all patients. These changes follow calls to reconsider the use of race in estimating kidney function and in medicine broadly. We analyzed potential changes in recommended care using eGFRcr with and without race among Black individuals in the US. Removal of race adjustment may increase CKD diagnoses among Black adults and enhance access to specialist care, medical nutrition therapy, kidney disease education, and kidney transplantation, while potentially excluding kidney donors and prompting drug contraindications or dose reductions for individuals reclassified to advanced stages of CKD. This potential for benefits and harms must be interpreted in light of persistent disparities in care, documented biases of eGFRcr with race removed, and the historical misuse of race as a biological variable to further racism.

The analysis of exRNA samples can be challenging: they are vulnerable to contamination and artifacts from different isolation techniques, present in lower concentrations than cellular RNA, and occasionally of exogenous origin. To address these challenges, we present exceRpt, the exRNA-processing toolkit of the NIH Extracellular RNA Communication Consortium (ERCC). exceRpt is structured as a cascade of filters and quantifications prioritized based on one's confidence in a given set of annotated RNAs. It generates quality control reports and abundance estimates for RNA biotypes. It is also capable of characterizing mappings to exogenous genomes, which, in turn, can be used to generate phylogenetic trees. exceRpt has been used to uniformly process all ∼3,500 exRNA-seq datasets in the public exRNA Atlas and is available from genboree.org and github.gersteinlab.org/exceRpt.