Faculty Bibliography

In studies with time-to-event outcomes, multiple, inter-correlated, and time-varying covariates are commonly observed. It is of great interest to model their joint effects by allowing a flexible functional form and to delineate their relative contributions to survival risk. A class of semiparametric transformation (ST) models offers flexible specifications of the intensity function and can be a general framework to accommodate nonlinear covariate effects. In this paper, we propose a partial-linear single-index (PLSI) transformation model that reduces the dimensionality of multiple covariates into a single index and provides interpretable estimates of the covariate effects. We develop an iterative algorithm using the regression spline technique to model the nonparametric single-index function for possibly nonlinear joint effects, followed by nonparametric maximum likelihood estimation. We also propose a nonparametric testing procedure to formally examine the linearity of covariate effects. We conduct Monte Carlo simulation studies to compare the PLSI transformation model with the standard ST model and apply it to NYU Langone Health de-identified electronic health record data on COVID-19 hospitalized patients' mortality and a Veteran's Administration lung cancer trial.

Cisgender women and transgender men are less likely to be assessed for PrEP eligibility, prescribed PrEP, or retained in PrEP care. Thus, this pilot PrEP educational intervention was tailored for healthcare providers (HCPs) in obstetrics/gynecology who provide care to cisgender women and transgender men in an academically-affiliated, public hospital women's health clinic. The three-lecture educational curriculum designed for HCPs focused on PrEP eligibility and counseling, formulations and adherence, and prescription and payment assistance programs. Pre- and post-intervention surveys assessed HCP knowledge and barriers to PrEP counseling and prescription. Among n = 49 participants (mean age = 32.8 years; 85.7% cisgender women, mean years practicing = 4.2 years) pre-intervention, 8.7% had prior PrEP training and 61.2% felt very/somewhat uncomfortable prescribing PrEP. Post-intervention, knowledge of PrEP contraindications, eligibility, follow-up care, and assistance programs all increased. HCPs identified key barriers to PrEP care including lack of a dedicated PrEP navigator, culturally and linguistically appropriate patient materials on PrEP resources/costs, and PrEP-related content integrated into EHRs. Ongoing PrEP educational sessions can provide opportunities to practice PrEP counseling, including information on financial assistance. At the institutional level, incorporating PrEP screening in routine clinical practice via EMR prompts, facilitating PrEP medication monitoring, and enhancing telehealth for follow-up care could enhance PrEP prescription.

INTRODUCTION/BACKGROUND:ChatGPT has shown the ability to answer clinical questions in general medicine but may be constrained by the specialized nature of kidney transplantation. Thus, it is important to explore how ChatGPT can be used in kidney transplantation and how its knowledge compares to human respondents. METHODS:We prompted ChatGPT versions 3.5, 4, and 4 Visual (4 V) with 12 multiple-choice questions related to six kidney transplant cases from 2013 to 2015 American Society of Nephrology (ASN) fellowship program quizzes. We compared the performance of ChatGPT with US nephrology fellowship program directors, nephrology fellows, and the audience of the ASN's annual Kidney Week meeting. RESULTS:Overall, ChatGPT 4 V correctly answered 10 out of 12 questions, showing a performance level comparable to nephrology fellows (group majority correctly answered 9 of 12 questions) and training program directors (11 of 12). This surpassed ChatGPT 4 (7 of 12 correct) and 3.5 (5 of 12). All three ChatGPT versions failed to correctly answer questions where the consensus among human respondents was low. CONCLUSION/CONCLUSIONS:Each iterative version of ChatGPT performed better than the prior version, with version 4 V achieving performance on par with nephrology fellows and training program directors. While it shows promise in understanding and answering kidney transplantation questions, ChatGPT should be seen as a complementary tool to human expertise rather than a replacement.

OBJECTIVE:Both diabetes and smoking significantly increase the risk of cardiovascular disease (CVD). Understanding whether a diagnosis of diabetes can be leveraged to promote smoking cessation is a gap in the literature. METHODS:We used data from the US National Health Interview Survey, 2006 to 2018, to investigate the relationship between self-report of diagnosis of diabetes and subsequent smoking abstinence among 142,884 respondents who reported regular smoking at baseline. Effect sizes were presented as hazard ratios (HRs) derived from multivariable Cox regression models adjusted for potential confounders using diabetes as a time-dependent covariate. Subgroup-specific estimates were obtained using interaction terms between diabetes and variables of interest. RESULTS:A self-reported diagnosis of diabetes was associated with smoking abstinence (HR: 1.21; 95% CI: 1.16 to 1.27). The strength of the association varied based on race (P for interaction: 0.004), where it was strongest in African Americans (HR: 1.44; 95% CI: 1.29 to 1.60); income (P for interaction <0.001), where it was strongest in those with a yearly income less than $35,000 (HR: 1.45; 95% CI: 1.36 to 1.53); and educational attainment (P for interaction <0.001), where it was strongest in those who did not attend college (HR: 1.48; 95% CI: 1.40 to 1.57). CONCLUSION/CONCLUSIONS:Among adults who smoke, a diagnosis of diabetes is significantly associated with subsequent smoking abstinence. The association is strongest in socially disadvantaged demographics, including African Americans, low-income individuals, and those who did not attend college.

On June 1, 2020, NYC Health + Hospitals, in partnership with the NYC Department of Health and Mental Hygiene, other city agencies, and a large network of community partners, launched the New York City Test & Trace (T2) COVID-19 response program to identify and isolate cases, reduce transmission through contact tracing, and provide support to residents during isolation or quarantine periods. In this paper, we describe lessons learned with respect to planning and implementation of case notification and contact tracing. Our findings are based on extensive document review and analysis of 74 key informant interviews with T2 leadership and frontline staff, cases, and contacts conducted between January and September 2022. Interviews elicited respondent background, history of program development, program leadership and structure, goals of the program, program evolution, staffing, data systems, elements of community engagement, trust with community, program reach, timeliness, equity, general barriers and challenges, general facilitators and best practices, and recommendations/improvement for the program. Facilitators and barriers revealed in the interviews primarily revolved around hiring and managing staff, data and technology, and quality of interactions with the public. Based on these facilitators and barriers, we identify suggestions to support effective planning and response for future case notification and contact tracing programs, including recommendations for planning during latent periods, case management and data systems, and processes for outreach to cases and contacts.

BACKGROUND/UNASSIGNED:Fewer minoritized patients with end-stage kidney disease (ESKD) receive kidney transplantation (KT); efforts to mitigate disparities have thus far failed. Pinpointing the specific stage(s) within the transplant care continuum (being informed of KT options, joining the waiting list, to receiving KT) where disparities emerge among each minoritized population is pivotal for achieving equity. We therefore quantified racial and ethnic disparities across the KT care continuum. METHODS/UNASSIGNED:We conducted a retrospective cohort study (2015-2020), with follow-up through 12/10/2021. Patients with incident dialysis were identified using the US national registry data. The exposure was race and ethnicity (Asian, Black, Hispanic, and White). We used adjusted modified Poisson regression to quantify the adjusted prevalence ratio (aPR) of being informed of KT, and cause-specific hazards models to calculate adjusted hazard ratios (aHR) of listing, and transplantation after listing. FINDINGS/UNASSIGNED:Among 637,951 adults initiating dialysis, the mean age (SD) was 63.8 (14.6), 41.8% were female, 5.4% were Asian, 26.3% were Black, 16.6% were Hispanic, and 51.7% were White (median follow-up in years [IQR]:1.92 [0.97-3.39]). Black and Hispanic patients were modestly more likely to be informed of KT (Black: aPR = 1.02, 95% confidence interval [CI]:1.01-1.02; Hispanic: aPR = 1.03, 95% CI: 1.02-1.03) relative to White patients. Asian patients were more likely to be listed (aHR = 1.18, 95% CI: 1.15-1.21) but less likely to receive KT (aHR = 0.56, 95% CI: 0.54-0.58). Both Black and Hispanic patients were less likely to be listed (Black: aHR = 0.87, 95% CI: 0.85-0.88; Hispanic: aHR = 0.85, 95% CI: 0.85-0.88) and receive KT (Black: aHR = 0.61, 95% CI: 0.60-0.63; Hispanic: aHR = 0.64, 95% CI: 0.63-0.66). INTERPRETATION/UNASSIGNED:Improved characterization of the barriers in KT access specific to each racial and ethnic group, and the interventions to address these distinct challenges throughout the KT care continuum are needed; our findings identify specific stages most in need of mitigation. FUNDING/UNASSIGNED:National Institutes of Health.

RATIONALE & OBJECTIVE/UNASSIGNED:The prevalence of chronic kidney disease (CKD) is known to increase with age; however, creatinine may be a less reliable filtration marker in older adults. Few studies have investigated the prevalence and progression of CKD using different filtration markers for estimating glomerular filtration rate (GFR). STUDY DESIGN/UNASSIGNED:A prospective observational cohort study. SETTING & PARTICIPANTS/UNASSIGNED:6,393 White and African American participants aged 65-100 years from the Atherosclerosis Risk in Communities Study (ARIC) at Visit 5, followed longitudinally at Visits 6 and 7. EXPOSURE AND OUTCOME/UNASSIGNED:The eGFR was estimated either by creatinine (eGFRcr), cystatin C (eGFRcys), creatinine and cystatin C (eGFRcr-cys), or using creatinine, cystatin C, and β-2-microglobulin (eGFRcr-cys-b2m). CKD progression was defined as 30% decline in eGFR at follow-up visits. ANALYTICAL APPROACH/UNASSIGNED:Logistic regression models, adjusted for sex, race and study center, diabetes, blood pressure, body mass index, prevalent cardiovascular disease, and heart failure. RESULTS/UNASSIGNED:when using eGFRcys (33%) compared with eGFRcr-cys (12%) or eGFRcr-cys-b2m (18%). The proportion with 30% eGFR decline was lowest with eGFRcr and highest with eGFRcys, with greater incidence in older age groups for all markers. LIMITATIONS/UNASSIGNED:No direct measurement of GFR. Not all participants survived or attended subsequent follow-up visits. CONCLUSIONS/UNASSIGNED:The prevalence and progression of CKD increase with age, but estimates vary with the filtration marker used. The eGFRcr gave the lowest estimate of CKD at 15% for people aged 65-69 years at Visit 5 while eGFRcys gave the highest estimates of CKD at 26% for that same population.