Faculty Bibliography

IMPORTANCE/UNASSIGNED:Virtual patient-physician communications have increased since 2020 and negatively impacted primary care physician (PCP) well-being. Generative artificial intelligence (GenAI) drafts of patient messages could potentially reduce health care professional (HCP) workload and improve communication quality, but only if the drafts are considered useful. OBJECTIVES/UNASSIGNED:To assess PCPs' perceptions of GenAI drafts and to examine linguistic characteristics associated with equity and perceived empathy. DESIGN, SETTING, AND PARTICIPANTS/UNASSIGNED:This cross-sectional quality improvement study tested the hypothesis that PCPs' ratings of GenAI drafts (created using the electronic health record [EHR] standard prompts) would be equivalent to HCP-generated responses on 3 dimensions. The study was conducted at NYU Langone Health using private patient-HCP communications at 3 internal medicine practices piloting GenAI. EXPOSURES/UNASSIGNED:Randomly assigned patient messages coupled with either an HCP message or the draft GenAI response. MAIN OUTCOMES AND MEASURES/UNASSIGNED:PCPs rated responses' information content quality (eg, relevance), using a Likert scale, communication quality (eg, verbosity), using a Likert scale, and whether they would use the draft or start anew (usable vs unusable). Branching logic further probed for empathy, personalization, and professionalism of responses. Computational linguistics methods assessed content differences in HCP vs GenAI responses, focusing on equity and empathy. RESULTS/UNASSIGNED:A total of 16 PCPs (8 [50.0%] female) reviewed 344 messages (175 GenAI drafted; 169 HCP drafted). Both GenAI and HCP responses were rated favorably. GenAI responses were rated higher for communication style than HCP responses (mean [SD], 3.70 [1.15] vs 3.38 [1.20]; P = .01, U = 12 568.5) but were similar to HCPs on information content (mean [SD], 3.53 [1.26] vs 3.41 [1.27]; P = .37; U = 13 981.0) and usable draft proportion (mean [SD], 0.69 [0.48] vs 0.65 [0.47], P = .49, t = -0.6842). Usable GenAI responses were considered more empathetic than usable HCP responses (32 of 86 [37.2%] vs 13 of 79 [16.5%]; difference, 125.5%), possibly attributable to more subjective (mean [SD], 0.54 [0.16] vs 0.31 [0.23]; P < .001; difference, 74.2%) and positive (mean [SD] polarity, 0.21 [0.14] vs 0.13 [0.25]; P = .02; difference, 61.5%) language; they were also numerically longer (mean [SD] word count, 90.5 [32.0] vs 65.4 [62.6]; difference, 38.4%), but the difference was not statistically significant (P = .07) and more linguistically complex (mean [SD] score, 125.2 [47.8] vs 95.4 [58.8]; P = .002; difference, 31.2%). CONCLUSIONS/UNASSIGNED:In this cross-sectional study of PCP perceptions of an EHR-integrated GenAI chatbot, GenAI was found to communicate information better and with more empathy than HCPs, highlighting its potential to enhance patient-HCP communication. However, GenAI drafts were less readable than HCPs', a significant concern for patients with low health or English literacy.

Elevated manganese (Mn) accumulates in the brain and induces neurotoxicity. SLC30A10 is a Mn efflux transporter that controls body Mn levels. We previously reported that full-body Slc30a10 knockout mice: (1) recapitulate the body Mn retention phenotype of humans with loss-of-function SLC30A10 mutations; and (2) unexpectedly, develop hypothyroidism induced by Mn accumulation in the thyroid, which reduces intra-thyroid thyroxine. Subsequent analyses of National Health and Nutrition Examination Survey data identified an association between serum Mn and subclinical thyroid changes. The emergence of thyroid deficits as a feature of Mn toxicity suggests that changes in thyroid function may be an underappreciated, but critical, modulator of Mn-induced disease. To better understand the relationship between thyroid function and Mn toxicity, here we further defined the mechanism of Mn-induced hypothyroidism using mouse and rat models. Slc30a10 knockout mice exhibited a profound deficit in thyroid iodine levels that occurred contemporaneously with increases in thyroid Mn and preceded the onset of overt hypothyroidism. Wild-type Mn-exposed mice also exhibited increased thyroid Mn levels, an inverse correlation between thyroid Mn and iodine levels, and subclinical hypothyroidism. In contrast, thyroid iodine levels were unaltered in newly-generated Slc30a10 knockout rats despite an increase in thyroid Mn, and the knockout rats were euthyroid. Thus, Mn-induced thyroid dysfunction in genetic or Mn exposure-induced mouse models occurs due to a reduction in thyroid iodine subsequent to an increase in thyroid Mn. Moreover, rat and mouse thyroids have differential sensitivities to Mn, which may impact the manifestations of Mn-induced disease in these routinely-used animal models.

The corona virus disease (COVID-19) pandemic disrupted daily life worldwide, and its impact on child well-being remains a major concern. Neighborhood characteristics affect child well-being, but how these associations were affected by the pandemic is not well understood. We analyzed data from 1039 children enrolled in the Environmental influences on Child Health Outcomes Program whose well-being was assessed using the Patient-Reported Outcomes Measurement Information System Global Health questionnaire and linked these data to American Community Survey (ACS) data to evaluate the impacts of neighborhood characteristics on child well-being before and during the pandemic. We estimated the associations between more than 400 ACS variables and child well-being t-scores stratified by race/ethnicity (non-Hispanic white vs. all other races and ethnicities) and the timing of outcome data assessment (pre-vs. during the pandemic). Network graphs were used to visualize the associations between ACS variables and child well-being t-scores. The number of ACS variables associated with well-being t-scores decreased during the pandemic period. Comparing non-Hispanic white with other racial/ethnic groups during the pandemic, different ACS variables were associated with child well-being. Multiple ACS variables representing census tract-level housing conditions and neighborhood racial composition were associated with lower well-being t-scores among non-Hispanic white children during the pandemic, while higher percentage of Hispanic residents and higher percentage of adults working as essential workers in census tracts were associated with lower well-being t-scores among non-white children during the same study period. Our study provides insights into the associations between neighborhood characteristics and child well-being, and how the COVID-19 pandemic affected this relationship.

Patients who speak languages other than English are frequently excluded from research. This exclusion exacerbates inequities, biases results, and may violate federal regulations and research ethics. Language justice is the right to communicate in an individual's preferred language to address power imbalances and promote equity. To promote language justice in research, we propose a method to translate and culturally-adapt multifaceted research materials into multiple languages simultaneously. Our method involves a multistep approach, including professional translation, review by bilingual expert panels to refine and reach consensus, and piloting or cognitive interviews with patients and families. Key differences from other translation approaches (eg, the World Health Organization) include omitting back-translation, given its limited utility in identifying translation challenges, and limiting expert panelist and piloting-participant numbers for feasibility. We detail a step-by-step approach to operationalizing this method and outline key considerations learned after utilizing this method to translate materials into 8 languages other than English for an ongoing multicenter pediatric research study on family safety-reporting. Materials included family brochures, surveys, and intervention materials. This approach took ∼6 months overall at a cost of <$2000 per language (not including study personnel costs). Key themes across the project included (1) tailor scope to timeline, budget, and resources, (2) thoughtfully design English source materials, (3) identify and apply guiding principles throughout the translation and editing process, and (4) carefully review content and formatting to account for nuances across multiple languages. This method balances feasibility and rigor in translating participant-facing materials into multiple languages simultaneously, advancing language justice in research.

The influence of the social environment on health behaviors is well documented. In recent years, there is mounting evidence of the health benefits of a plant-based eating pattern, yet little is known about how the social environment impacts the adoption of a plant-based eating pattern, specifically. In this convergent parallel mixed-methods study, we analyzed quantitative survey data and qualitative focus group data to assess how social support impacted participants of a lifestyle medicine intervention focused on the adoption of a plant-predominant eating pattern. Regression analysis of survey data showed a positive association between positive social support and healthy plant-based eating, while no association was found between negative social support and healthy plant-based eating. Focus groups yielded further insights into how positive aspects of social relationships with family and friends facilitated the adoption of plant-predominant eating among participants. Qualitative findings also showed the ways in which negative social support hindered progress to adopt a plant-predominant eating pattern including not eating the same foods as participants, being judgmental about new dietary behaviors, and encouraging participants to eat non-plant-based foods. Taken together, social support appears to be an important factor for individuals adopting a plant-predominant eating pattern. Future research is needed to explore mechanisms to enhance positive social support while mitigating negative aspects of social relationships for individuals participating in similar lifestyle medicine interventions that emphasize on plant-predominant eating.

KEY POINTS:Longitudinal untargeted metabolomics. Children with CKD have a circulating metabolome that changes over time. BACKGROUND:Understanding plasma metabolome patterns in relation to changing kidney function in pediatric CKD is important for continued research for identifying novel biomarkers, characterizing biochemical pathophysiology, and developing targeted interventions. There are a limited number of studies of longitudinal metabolomics and virtually none in pediatric CKD. METHODS:< 0.05. RESULTS:< 0.05 and associations with UPCR in LME modeling at FDR <0.05. CONCLUSIONS:We characterized longitudinal plasma metabolomic patterns associated with eGFR and UPCR in a large pediatric CKD population. Many of these metabolite signals have been associated with CKD progression, etiology, and proteinuria in previous CKD Biomarkers Consortium studies. There were also novel metabolite associations with eGFR and proteinuria detected.

KEY POINTS:Regardless of race and ethnicity, older adults with kidney failure residing in or receiving care at dialysis facilities located in high-segregation neighborhoods were at a 1.63-fold and 1.53-fold higher risk of dementia diagnosis, respectively. Older adults with kidney failure residing in minority-predominant high-segregation neighborhoods had a 2.19-fold higher risk of dementia diagnosis compared with White individuals in White-predominant neighborhoods. BACKGROUND:a form of structural racism recently identified as a mechanism in numerous other health disparities. METHODS:We identified 901,065 older adults (aged ≥55 years) with kidney failure from 2003 to 2019 using the United States Renal Data System. We quantified dementia risk across tertiles of residential neighborhood segregation score using cause-specific hazard models, adjusting for individual- and neighborhood-level factors. We included an interaction term to quantify the differential effect of segregation on dementia diagnosis by race and ethnicity. RESULTS: CONCLUSIONS:Residing in or receiving care at dialysis facilities located in high-segregation neighborhoods was associated with a higher risk of dementia diagnosis among older individuals with kidney failure, particularly minoritized individuals.

KEY POINTS:Integrated analysis of proteome and metabolome identifies modules associated with CKD progression and kidney failure. Ephrin transmembrane proteins and podocyte-expressed CRIM1 and NPNT emerged as central components and warrant experimental and clinical investigation. BACKGROUND:Proteins and metabolites play crucial roles in various biological functions and are frequently interconnected through enzymatic or transport processes. METHODS:=569). RESULTS:). CONCLUSIONS:This study demonstrates that integration of the proteome and metabolome can identify functions of pathophysiologic importance in kidney disease.

IMPORTANCE/UNASSIGNED:Heart failure (HF) and frailty frequently coexist and may share a common pathobiology, although the underlying mechanisms remain unclear. Understanding these mechanisms may provide guidance for preventing and treating both conditions. OBJECTIVE/UNASSIGNED:To identify shared pathways between incident HF and frailty in late life using large-scale proteomics. DESIGN, SETTING, AND PARTICIPANTS/UNASSIGNED:In this cohort study, 4877 aptamers (Somascan v4) were measured among participants in the community-based longitudinal Atherosclerosis Risk In Communities (ARIC) cohort study at visit 3 (V3; 1993-1995; n = 10 638) and at visit 5 (V5; 2011-2013; n = 3908). Analyses were externally replicated among 3189 participants in the Cardiovascular Health Study (CHS). Data analysis was conducted from February 2022 to June 2023. EXPOSURES/UNASSIGNED:Protein aptamers, measured at study V3 and V5. MAIN OUTCOMES AND MEASURES/UNASSIGNED:Outcomes assessed included incident HF hospitalization after V3 and after V5, prevalent frailty at V5, and incident frailty between V5 and visit 6 (V6; 2016-2017; n = 4131). Frailty was assessed using the Fried criteria. Analyses were adjusted for age, gender, race, field center, hypertension, diabetes, smoking status, body mass index, estimated glomerular filtration rate, prevalent coronary heart disease, prevalent atrial fibrillation, and history of myocardial infarction. Mendelian randomization (MR) analysis was performed to assess potential causal effects of candidate proteins on HF and frailty. RESULTS/UNASSIGNED:A total of 4877 protein aptamers were measured among 10 638 participants at V3 (mean [SD] age, 60 [6] years; 4886 [46%] men). Overall, 286 proteins were associated with incident HF after V3 (822 events; P < 1.0 × 10-5), 83 of which were also associated with incident after V5 (336 events; P < 1.7 × 10-4). Among HF-free participants at V5 (n = 3908; mean [SD] age, 75 [5] years; 1861 [42%] men), 48 of 83 HF-associated proteins were associated with prevalent frailty (223 cases; P < 6.0 × 10-4), 18 of which were also associated with incident frailty at V6 (152 cases; P < 1.0 × 10-3). These proteins enriched fibrosis and inflammation pathways and demonstrated stronger associations with incident HF with preserved ejection fraction (HFpEF) than HF with reduced ejection fraction. All 18 proteins were associated with both prevalent frailty and incident HF in CHS. MR identified potential causal effects of several proteins on frailty and HF. CONCLUSIONS AND RELEVANCE/UNASSIGNED:In this study, the proteins associated with risk of HF and frailty enrich for pathways related to inflammation and fibrosis as well as risk of HFpEF. Several of these proteins could potentially contribute to the shared pathophysiology of frailty and HF.