About

Michelle L. Bell is the Mary E. Pinchot Professor of Environmental Health and serves as Senior Associate Dean of Research and Director of Doctoral Studies at Yale School of the Environment. Her areas of expertise include climate science and policy, extreme weather and climate change, environmental justice, environmental ethics, human health and well-being, environmental resources and systems, air pollution, and urban environmental issues. Her educational background includes a Ph.D. in Environmental Engineering from Johns Hopkins University, a Master of Science in Environmental Management and Economics from Johns Hopkins University, a Master of Science in Environmental Engineering and Science from Stanford University, a Master of Science in Philosophy of Epistemology, Ethics, and Mind from the University of Edinburgh, and a B.S. in Environmental Engineering Science with a minor in Music from MIT. She has contributed extensively to research on atmospheric environment, health impacts of air pollution, and climate-related health risks, with numerous publications and a focus on advancing understanding of environmental health issues.

Research topics

• Medicine
• Environmental health
• Geography
• Demography
• Environmental science
• Gerontology
• Mathematics
• Surgery
• Statistics
• Sociology
• Internal medicine
• Meteorology
• Archaeology
• Chemistry
• Economic geography
• Nursing
• Biology
• Economics
• Environmental protection
• Ecology
• Family medicine
• Environmental engineering
• Library science
• Socioeconomics

Selected publications

• The Creation of a Weight of Evidence Scoring Database for Risk Factors for Adverse Impacts to Birth Outcomes Using Expert Elicitation

  Birth Defects Research · 2026-01-01

  articleOpen access

  BACKGROUND: Adverse birth outcomes are important public health measures and account for a substantial public health burden. There is considerable diversity of these health endpoints, as well as in the many factors suspected or recognized to increase their risk. This diversity renders it challenging to synthesize the totality of the evidence by carrying out a systematic review. METHODS: We undertook an expert elicitation to characterize the state of the literature on this topic, the interconnections amongst risk factors, and identify those that are a priority for future research. A panel of 30 scientists and physicians with expertise in birth outcome epidemiology were solicited to identify a series of health outcomes, and associated risk factors. RESULTS: This resulted in a listing of 15 birth outcomes, and 247 possible risk factors in total, with varying numbers of risk factors for each health outcome. Each panel member was asked to score the weight of evidence (WOE) of each risk factor and birth outcome combination (n = 1127) on a scale of 1 (no evidence) to 5 (strong evidence) for which the expert had a working knowledge of the literature. Not all experts scored each/every outcome/risk factor combination, reflecting the different types of expertise on the panel. The compilation of these WOE scores was used to create a publicly available database for birth outcomes risk factors (https://scipinion-rfbo.onrender.com) that is intended to be updated over time so as to serve as an up to date resource for the research and medical community. CONCLUSIONS: This expert approach serves as a unique and valuable approach to data reduction that can help to inform research priorities, while providing an open access resource to identify risk factors that may act as confounders and/or modifiers in future epidemiology studies and clinical trials.

  PublisherOA PDFDOI

• Hog CAFO construction and temporal trends in racial/ethnic demographics, economic growth, and land use in North Carolina

  Journal of Environmental Management · 2026-04-01

  articleOpen accessSenior author

  Since the mid-twentieth century, the massive demand for meat production in the US has led to the development of Concentrated Animal Feeding Operations (CAFOs). Research has shown that CAFOs are associated with local environmental degradation and community health disparities; however, there is a limited understanding of their impact on community development. The purpose of our study is to investigate how the population within hog CAFO communities evolved with respect to race, class, education, and income as compared to non-hog CAFO communities in North Carolina (NC). We investigated the changes in demographic, household, and local economic trends associated with varying levels of hog CAFO construction (i.e., area-weighted CAFO count within a 15 km buffer) in counties compared to counties without such facilities during the peak period of hog growth and policy cycles regulating hog farm construction in NC. A higher density of hog CAFOs was constructed in areas with a higher percentage of Black and Hispanic residents and residents of lower socio-economic status (SES) in NC. Findings suggest CAFOs are associated with the displacement of Black residents. In turn, the percent of residents who are White is increasing in hog farming communities, even with increased growth rates in the Hispanic population. Hog CAFO construction is associated with lower wage growth, a tempering in the growth of employment opportunities, and cropland expansion compared to non-hog farming communities. This research aims to provide insight into and better inform the planning and development of CAFOs, so that rural communities can grow healthily and sustainably. • Hog farm construction in North Carolina peaked in the mid-90s. • Hog farms were constructed mainly in areas with a higher percentage of residents of color and low-income communities. • Hog farm construction was associated with disparities in race/ethnicity and local economic growth trends. • Community dynamics may have been influenced by policies regulating hog farm construction.

  PublisherDOI

• Atmospheric stressors and kidney diseases

  Nature Reviews Nephrology · 2026-04-09

  article
  PublisherDOI

• Density of animal feeding operations, including concentrated animal feeding operations (CAFOs), and cancer incidence: A county-level ecological study across three U.S. states

  Environmental Research · 2026-03-17

  articleOpen accessSenior author

  BACKGROUND: Animal feeding operations (AFOs) including concentrated animal feeding operations (CAFOs) are significant sources of environmental pollution with potential public health implications. Despite growing concern of environmental health risk, few studies have assessed the associations between exposure to AFOs/CAFOs and cancer incidence across diverse geographic regions and populations. OBJECTIVE: This study investigates county-level cancer incidence by state in relation to AFO/CAFO exposure in three US states. METHODS: We analyzed county-level incidence data for all- and site-specific cancers from 2000 to 2021 and AFO/CAFO density for three states (i.e., California, Iowa, and Texas). To address confounding, we applied propensity score matching to compare counties with high AFO/CAFO exposure to control counties. Stratified analyses were conducted by state and cancer type. RESULTS: Higher exposure to AFO/CAFOs was associated with increased cancer incidence in all three states, although the magnitude and statistical significance of the associations varied by state. Compared to control counties, exposed counties had significantly higher all-cancer incidence rate ratios (IRRs): 1.044 (95% CI 1.040, 1.047) in California, 1.079 (1.066, 1.091) in Iowa, and 1.078 (1.073, 1.084) in Texas. Stratified analyses by cancer type showed higher associations for specific cancers in each state (e.g., bladder cancer in California, colorectal cancer for Iowa, and lung and bronchus cancer in Texas). CONCLUSION: Our findings suggest a link between higher AFO/CAFO exposure and increased cancer incidence across various US states. Future research using individual-level data, refined exposure assessment, and longitudinal approaches are needed to strengthen the evidence.

  PublisherDOI

• The burden of El Niño–Southern Oscillation-related dengue attributable to anthropogenic climate change: a multicountry modelling study

  The Lancet Planetary Health · 2026-04-01

  articleOpen access

  BACKGROUND: Dengue is known to be associated with El Niño-Southern Oscillation (ENSO) but the size of the effect is unclear, as is the influence of anthropogenic climate change (ACC). We aimed to quantify the associations between ENSO and dengue risk in 21 countries, and to estimate the contribution of ACC to the ENSO-related dengue burden. METHODS: We collected monthly dengue cases and observed and simulated climate data from 21 countries including 1237 locations from 2000 to 2019. We characterised Eastern Pacific (EP) and Central Pacific (CP) ENSO exposures for each location based on the E and C indices and their respective teleconnections. Location-specific association between ENSO exposure and dengue cases was estimated using negative binomial generalised linear model combined with best linear unbiased predictions. We also estimated the ENSO-related dengue burden under scenarios with and without ACC. FINDINGS: For each standard deviation increase in EP-El Niño strength and CP-La Niña strength, the overall risk of dengue cases across locations changed by 23·70% (95% CI 21·50 to 25·94) and -9·07% (-9·91 to -8·21), respectively. During 2000 to 2019, 4·45% (95% empirical CI [eCI] 3·75 to 5·32) and -3·34% (-4·01 to -2·64) of dengue cases were attributable to EP-El Niño strength and CP-La Niña strength, respectively. ACC accounted for 48·64% (95% eCI 38·01 to 60·19) of the EP-El Niño-attributable dengue increment and 33·05% (28·66 to 38·25) of the CP-La Niña-attributable reduction. These estimates corresponded to 403 197 (95% eCI 315 109 to 498 940) and -205 641 (-238 030 to -178 329) dengue cases across 1237 locations, respectively. The associations with ENSO varied strongly across the 21 countries. INTERPRETATION: This study presents new model-based evidence of the strong associations between ENSO and dengue risk at a multicountry level, and suggests that the contribution of ACC to the effects of ENSO might differ geographically. FUNDING: Prevention and Control of Emerging and Major Infectious Diseases National Science and Technology Major Project, the National Natural Science Foundation of China, and the Czech Ministry of Education Youth and Sport's programme ERC CZ.

  PublisherDOI

• Tropical Cyclone Exposure and Risk of Adverse Birth Outcomes in Urban and Rural Areas of Georgia

  Environmental Research Communications · 2026-05-21

  articleOpen access

  Abstract Tropical cyclones (TCs) are highly destructive weather disasters. Prior studies of TCs and birth outcomes often examined single storms or single TC characteristics (e.g., maximum wind speed) and rarely assessed urban/rural differences. To better understand TC impacts on perinatal health, we evaluated associations between multiple TC exposure metrics and several adverse birth outcomes and considered urban-rural variations. We conducted a population-based time series analysis of 2,436,478 singleton births in Georgia from 2000–2018, in which individual-level state birth records were aggregated to the county–week level and linked to weekly, county-level TC data from the National Oceanic and Atmospheric Administration and the National Aeronautics and Space Administration. TC exposure metrics included maximum sustained wind speed (>17, 22, 25 m/s), cumulative rainfall (>125, 150, 175 mm), storm proximity (<5, 10, 20 km), and flooding events (yes/no). Generalized linear models with a Poisson distribution were used to estimate relative risks (RRs) and 95% confidence intervals (CIs) for weekly rates of preterm birth (PTB, <37 weeks), low birthweight (LBW, <2500 g), small-for-gestational-age (SGA, <10th percentile), and proportion of male births. Winds >22 m/s were associated with higher risk of PTB (RR=1.58 [95% CI: 1.27, 1.96]), LBW (RR=1.77 [95%CI: 1.40, 2.23]), and SGA (RR=1.38 [95%CI: 1.12, 1.70]). Rainfall >175 mm was associated with PTB (RR=1.44 [95%CI: 1.08, 1.93]) and LBW (RR=2.06 [95%CI: 1.50, 2.83]). Proximity <5 km was associated with PTB (RR=1.41 [95%CI: 1.04, 1.90]) and LBW (RR=2.11 [95%CI: 1.38, 3.23]). Flooding was associated with LBW (RR=1.11 [95%CI: 1.03, 1.21]) and SGA (RR=1.08 [95%CI: 1.01, 1.16]). Risk estimates were generally higher in rural versus metropolitan counties. Across multiple metrics, TC exposures were linked to increased PTB and fetal growth restriction, with stronger effects in rural counties. These findings bolster information on perinatal risks of TCs and inform more targeted disaster preparedness.

  PublisherDOI

• Advanced Practice Provider Hybrid Role: A Novel Approach to Patient Care in Liver Transplant

  American Journal of Transplantation · 2025-08-01

  article
  PublisherDOI

• Rainfall variability and under-five child mortality in 59 low- and middle-income countries

  Nature Water · 2025-08-11

  article
  PublisherDOI

• Global excess deaths associated with heatwaves in 2023 and the contribution of human-induced climate change

  The Innovation · 2025-09-04 · 3 citations

  articleOpen access

  An unprecedented heatwave swept the globe in 2023, marking it one of the hottest years on record and raising concerns about its health impacts. However, a comprehensive assessment of the heatwave-related mortality and its attribution to human-induced climate change remains lacking. We aim to address this gap by analyzing high-resolution climate and mortality data from 2,013 locations across 67 countries/territories using a three-stage modeling approach. First, we estimated historical heatwave-mortality associations using a quasi-Poisson regression model with distributed lag structures, considering lag effects, seasonality, and within-week variations. Second, we pooled the estimates in meta-regression, accounting for spatial heterogeneity and potential changes in heatwave-mortality associations over time. Third, we predicted grid-specific (0.5 0.5) association in 2023 and calculated the heatwave-related excess deaths, death ratio, and death rate per million people. Attribution analysis was conducted by comparing heatwave-related mortality under factual and counterfactual climate scenarios. We estimated 178,486 excess deaths (95% empirical confidence interval [eCI], 159,892≥204,147) related to the 2023 heatwave, accounting for 0.73% of global deaths, corresponding to 23 deaths per million people. The highest mortality rates occurred in Southern (120, 95% eCI, 116≥126), Eastern (107, 95% eCI, 100≥114), and Western Europe (66, 95% eCI, 62≥70), where the excess death ratio was also higher. Notably, 54.29% (95% eCI, 45.71%≥61.36%) of the global heatwave-related deaths were attributable to human-induced climate change. These results underscore the urgent need for adaptive public health interventions and climate mitigation strategies to reduce future mortality burdens in the context of increasing global warming.

  PublisherDOI

• Statistical downscaling of coarse-resolution fine particulate matter predictions over the contiguous United States: model development, evaluation, and implication in health impact assessment

  The Science of The Total Environment · 2025-08-26 · 1 citations

  articleOpen accessSenior author

  Fine particulate matter (PM 2.5 ) predictions at a high spatial resolution (i.e., neighborhood scale) are critically needed to better understand the health impacts of air pollution, especially at neighborhood scales. This work develops a statistical downscaling approach to predict PM 2.5 at a 1-km grid resolution over the contiguous United States (CONUS) under baseline and future energy transition scenarios and estimate health benefits utilizing the Environmental Benefits Mapping and Analysis Program (BenMAP). To this end, we incorporate the satellite-based high-resolution aerosol optical depth (AOD), land use data, and PM 2.5 composition predicted by the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) at 36-km into daily multi-linear regressions over different climate regions of the CONUS. Compared to the WRF-Chem baseline predictions in 2008–2012, 1-km PM 2.5 estimates enhance the accuracy by increasing the yearly correlation coefficients from ~0.4 to ~0.8 and reducing normalized mean errors from ~47 % to ~27 %. Future 1-km PM 2.5 is projected by combining the baseline 5-yr (2008–2012) monthly-averaged training coefficients with high-resolution statistically improved projected AOD and PM 2.5 subsets from WRF-Chem. BenMAP with WRF-Chem predictions under future energy scenarios shows an average of 2478 fewer deaths per year in 2050 in New York City and Boston due to PM 2.5 , while the downscaled PM 2.5 shows less PM 2.5 reduction and about half the health benefit of the WRF-Chem projections. The downscaling approach is more computationally efficient than running the 3-D air quality model with a 1-km spatial grid resolution. This work uniquely combines WRF-Chem outputs and statistical downscaling to provide high-resolution and high-fidelity PM 2.5 predictions. • High-resolution and high-fidelity 1-km PM 2.5 concentrations over the CONUS. • Novel future 1-km PM 2.5 projection under energy transition scenarios. • Improved spatial resolution shows less projected reduction in PM 2.5 and mortality.

  PublisherDOI

Recent grants

• NIH Grant R01ES016317

  NIH · $2.9M · 2014

• Environmental Health Disparities in an Older Population

  NIH · $3.9M · 2017–2024

• NIH Grant R21ES021427

  NIH · $466k · 2015

• NIH Grant R01ES015028

  NIH · $477k · 2011

• NIH Grant R01ES019587

  NIH · $2.6M · 2016

Frequent coauthors

• Éric Lavigne

  Wilfrid Laurier University

  135 shared

• Shilu Tong

  Chinese Center For Disease Control and Prevention

  128 shared

• Yuming Guo

  105 shared

• Whanhee Lee

  97 shared

• Ji-Young Son

  Yale University

  93 shared

• Martina S. Ragettli

  Swiss Tropical and Public Health Institute

  92 shared

• Aurelio Tobı́as

  92 shared

• Ho Kim

  Seoul National University

  91 shared

Education

• Masters of Philosophy, Philsophy

  University of Edinburgh

  2020

• Ph.D. Environmental Engineering, Department of Geography and Environmental Engineering

  Johns Hopkins University

  2002

• M.Sc. Environmental Management and Economics

  Johns Hopkins University

  1999

• M.Sc. Environmental Engineering, Civil and Environmental Engineering

  Stanford University

  1994

• B.Sc. Environmental Engineering, Minor in Music, Department of Civil and Environmental Engineering

  Massachusetts Institute of Technology

  1992