Kevin Josey

ORCID: 0000-0003-2490-6272
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Air Quality and Health Impacts
  • Climate Change and Health Impacts
  • Statistical Methods in Clinical Trials
  • Advanced Causal Inference Techniques
  • Diabetes Management and Research
  • Statistical Methods and Inference
  • Global Health Care Issues
  • Diabetes Treatment and Management
  • Statistical Methods and Bayesian Inference
  • Health Systems, Economic Evaluations, Quality of Life
  • Diabetes and associated disorders
  • Air Quality Monitoring and Forecasting
  • Bayesian Modeling and Causal Inference
  • COVID-19 impact on air quality
  • Coronary Interventions and Diagnostics
  • Dietary Effects on Health
  • Bariatric Surgery and Outcomes
  • Computational Drug Discovery Methods
  • Acute Myocardial Infarction Research
  • Health disparities and outcomes
  • Diabetes, Cardiovascular Risks, and Lipoproteins
  • Thermoregulation and physiological responses
  • Peripheral Artery Disease Management
  • Advanced Bandit Algorithms Research
  • Meta-analysis and systematic reviews

Office of Patient Care Services
2025

University of Colorado Anschutz Medical Campus
2020-2025

Colorado School of Public Health
2020-2024

Harvard University
2021-2024

University of Colorado Boulder
2023

VA Eastern Colorado Health Care System
2016-2022

University of Colorado Denver
2021-2022

United States Department of Veterans Affairs
2022

Denver VA Medical Center
2017

Univerzitetski Klinički Centar Srbije
2016

High levels of PM 2.5 during the 2020 wildfires in western United States led to an excess COVID-19 cases and deaths.

10.1126/sciadv.abi8789 article EN cc-by-nc Science Advances 2021-08-13

Black Americans are exposed to higher annual levels of air pollution containing fine particulate matter (particles with an aerodynamic diameter ≤2.5 μm [PM

10.1056/nejmsa2300523 article EN New England Journal of Medicine 2023-03-24

Many patients with chronic obstructive pulmonary disease (COPD) receive hypnotic prescriptions to mitigate insomnia symptoms. Although clinical practice guidelines advise short-term use, often these medications long-term. As COPD may be more susceptible adverse effects of hypnotics, it is critical that we better understand the incidence and potential influences this practice. To characterize predictors guideline-discordant long-term receipt among COPD. Using nationwide Veterans Health...

10.1513/annalsats.202407-798oc article EN Annals of the American Thoracic Society 2025-02-12

Little epidemiologic research has focused on pollution-related risks in medically vulnerable or marginalized groups. Using a nationwide 50% random sample of 2008-2016 Medicare Part D-eligible fee-for-service participants the United States, we identified cohort with high-risk conditions for cardiovascular and thromboembolic events (CTEs) linked individuals seasonal average zip-code-level concentrations fine particulate matter (particulate an aerodynamic diameter ≤ 2.5 μm (PM2.5)). We assessed...

10.1093/aje/kwad089 article EN American Journal of Epidemiology 2023-04-17

To assess the association between ambient heat and all-cause cause-specific emergency department (ED) visits acute hospitalizations among Medicare beneficiaries in conterminous United States.

10.1016/j.ajem.2024.04.010 article EN cc-by-nc-nd The American Journal of Emergency Medicine 2024-04-11

Evidence to guide type 2 diabetes treatment individualization is limited. We evaluated heterogeneous effects (HTE) of intensive glycemic control in patients on major adverse cardiovascular events (MACE) the Action Control Cardiovascular Risk Diabetes Study (ACCORD) and Veterans Affairs Trial (VADT).Causal forests machine learning analysis was performed using pooled individual data from two randomized trials (n = 12,042) identify HTE versus standard MACE with diabetes. used variable...

10.1186/s12933-022-01496-7 article EN cc-by Cardiovascular Diabetology 2022-04-27

Two important considerations in clinical research studies are proper evaluations of internal and external validity. While randomized trials can overcome several threats to validity, they may be prone poor Conversely, large prospective observational sampled from a broadly generalizable population externally valid, yet susceptible particularly confounding. Thus, methods that address confounding enhance transportability study results across populations essential for internally valid causal...

10.1002/sim.9523 article EN publisher-specific-oa Statistics in Medicine 2022-07-18

We show how entropy balancing can be used for transporting experimental treatment effects from a trial population onto target population. This method is doubly robust in the sense that if either outcome model or probability of participation correctly specified, then estimate average effect consistent. Furthermore, we only require sample moments modifiers drawn to consistently effect. compared finite-sample performance with several alternative methods between populations. Entropy techniques...

10.1002/sim.9031 article EN Statistics in Medicine 2021-05-20

Abstract A common goal in observational research is to estimate marginal causal effects the presence of confounding variables. One solution this problem use covariate distribution weight outcomes such that data appear randomized. The propensity score a natural quantity arises setting. Propensity weights have desirable asymptotic properties, but they often fail adequately balance finite samples. Empirical balancing methods pose as an appealing alternative by exactly sample moments...

10.1111/sjos.12457 article EN Scandinavian Journal of Statistics 2020-04-07

<h3>Importance</h3> Fruit and vegetable vouchers have been implemented by cities counties across the US to increase fruit intake thereby improve overall nutritional quality. <h3>Objective</h3> To determine whether why use of are associated with varied different populations environments. <h3>Design, Setting, Participants</h3> In a population-based pre-post cohort study 671 adult participants low income before during (6 months after initiation) participation in 6-month program, were...

10.1001/jamanetworkopen.2021.1757 article EN cc-by-nc-nd JAMA Network Open 2021-03-22

To evaluate the synergistic effects created by fine particulate matter (PM2.5) and corticosteroid use on hospitalisation mortality in older adults at high risk for cardiovascular thromboembolic events (CTEs). A retrospective cohort study using a US nationwide administrative healthcare claims database. 50% random sample of participants with high-risk conditions CTE from 2008-2016 Medicare Fee-for-Service population. Corticosteroid therapy seasonal-average PM2.5. Incidences myocardial...

10.1136/bmjopen-2023-072810 article EN cc-by-nc BMJ Open 2023-09-01

Therapeutic inertia threatens the potential long-term benefits of achieving early glycemic control after type 2 diabetes diagnosis. We evaluated temporal trends in second-line medication initiation among individuals initially treated with metformin.

10.2337/dc21-2492 article EN Diabetes Care 2022-03-28

Abstract Background Multiple studies from countries with relatively lower PM 2.5 level demonstrated that acute and chronic exposure even at than recommended level, e.g., 9 μg/m 3 in the US increased risk of cardiovascular (CV) events. However, limited using individual data exist a wider range levels to illustrate shape exposure-response curve throughout including &gt; 20 2·5 concentrations. Taiwan its policies reduced over time provide opportunities dose response curves how reductions...

10.1101/2024.05.08.24306967 preprint EN medRxiv (Cold Spring Harbor Laboratory) 2024-05-09

A fundamental task in causal inference is estimating the effect of a distribution shift treatment variable. We refer to this problem as shift-response function (SRF) estimation. Existing neural network methods for lack theoretical guarantees and practical implementations SRF In paper, we introduce Targeted Regularization Exposure Shifts with Neural Networks (TRESNET), method estimate SRFs robustness efficiency guarantees. Our contributions are twofold. First, propose targeted regularization...

10.1145/3637528.3671761 article EN Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining 2024-08-24

Exposure measurement error is a ubiquitous but often overlooked challenge in causal inference with observational data. Existing methods accounting for exposure largely rely on restrictive parametric assumptions, while emerging data-adaptive estimation approaches allow less assumptions at the cost of flexibility, as they are typically tailored towards rigidly-defined statistical quantities. There remains critical need assumption-lean that both flexible and possess desirable theoretical...

10.48550/arxiv.2410.12590 preprint EN arXiv (Cornell University) 2024-10-16

Unequal exposure to air pollution by race and socioeconomic status is well-documented in the U.S. However, there has been relatively little research on inequities collection of PM2.5 data, creating a critical gap understanding which neighborhood exposures are represented these datasets. In this study we use multilevel models with random intercepts county state, stratified urbanicity investigate association between six key environmental justice (EJ) attributes (%AIAN, %Asian %Black,...

10.48550/arxiv.2410.18692 preprint EN arXiv (Cornell University) 2024-10-24

Current guidelines recommend that patients with non-ST elevation acute coronary syndromes (NSTEACS) receive dual antiplatelet therapy (DAPT) early in hospitalisations. However, observational studies suggest this rarely occurs. We evaluated site-specific variation and clinical outcomes associated DAPT among undergoing angiography for NSTEACS.In analysis, we identified NSTEACS Veterans Affairs hospitals from 2008 to 2016 assessed characteristics site (administration <24 hours of admission)....

10.1136/heartjnl-2018-314553 article EN Heart 2019-05-15
Coming Soon ...