Emily N Peterson, PhD
Department of Biostatistics and Bioinformatics, Emory University
Project maintained by Enpeterson Hosted on GitHub Pages — Theme by mattgraham
Contact Information
Emily N Peterson
Department of Biostatistics and Bioinformatics
Rollins School of Public Health
Emory University
1518 Clifton Rd., Atlanta, GA 30329
emily.nancy.peterson@emory.edu
Education
- Emory University, Atlanta, GA. 2020-2021. Post-Doctoral Fellowship. Postdoctoral mentor: Dr. Lance Waller.
- University of Massachusetts Amherst, Amherst, MA. 2019. Doctor of Philosophy in Biostatistics. Adviser: Dr. Leontine Alkema.
- Vanderbilt University, Nashville, TN. 2015. Master of Science in Biostatistics. Adviser: Dr. Tatsuki Koyama.
- Penn State University, Hershey, PA. 2013. Master of Science in Public Health Sciences. Adviser: Dr. Vernon Chinchilli.
- Davidson College, Davidson, NC. May 2006. Bachelor of Science in Psychology. With a research focus in Cognitive and Neuroscience Psychology.
Research
Primary Interests
I develop Bayesian statistical methods to improve the accuracy of population health data for small and historically underrepresented populations. My work focuses on correcting measurement error in mortality and demographic data, with applications to Tribal health, maternal mortality, and opioid surveillance. I specialize in Bayesian hierarchical modeling, spatial epidemiology, and statistical demography to produce policy-relevant estimates under data sparsity.
Research Themes
- Bayesian hierarchical modeling for small-area population health
- Measurement error correction in mortality and demographic data
- Spatial and spatio-temporal epidemiology
- Statistical demography and population estimation
- Data science for public health surveillance systems
Projects
- Data integration methods to estimate small area American Indian and Alaska Native mortality outcomes adjusting for measurement error.
- A Bayesian Spatio-Temporal Top-Down Framework for Estimating Opioid Use Disorder Risk Under Data Sparsity.
- Correcting Undercounting of GDM among Immigrant Women in Georgia.
- A Bayesian approach to estimate maternal mortality using national civil registration vital statistics data accounting for reporting errors.
- A Bayesian hierarchical small area misclassification model to estimate extent of misclassification errors of maternal mortality by U.S. Pregnancy Mortality Surveillance Systems.
- A Bayesian hierarchical small-area population model accounting for data source specific methods from American Community Survey, Population Estimates Program, and Decennial Census data.
- A Bayesian hierarchical model to estimate the extent of reporting errors of maternal mortality by national vital registration systems.
- BEACON: Bayesian Estimation and Correction for AI/AN Outcomes using Native-data — integrating Cherokee Nation clinical data with OSDH surveillance to estimate adjusted AI/AN gestational diabetes prevalence in Oklahoma.
Research Impact
My work supports public health agencies, Tribal Nations, and international organizations in producing more accurate mortality and demographic estimates for policy and resource allocation. I collaborate with partners to translate statistical methods into practical surveillance tools that promote data equity and public health decision-making.
Collaborations
- Cherokee Nation Public Health
- World Health Organization Maternal Mortality Interagency Group
- Harvard T.H. Chan School of Public Health
- UK Small Area Health Statistics Unit
- University of Massachusetts Amherst
- GA Department of Health
- CDC Maternal Mortality Surveillance Group and Michael Kramer at Emory University Epidemiology Department
- Oregon State Marine Mammal Institute
Featured Research
A Bayesian Hierarchical Small Area Population Model Accounting for Data Source Specific Methodologies From American Community Survey, Population Estimates Program, and Decennial Census Data
What a DAG Can Tell Us
Publications
- EN Peterson, G Guranich, J Cresswell, L Alkema. A Bayesian approach to estimate maternal mortality using national civil registration vital statistics data accounting for misclassification errors. (In progress)
- EN Peterson, RC Nethery, T Padellini, JT Chen, BA Coull, FB Piel, J Wakefield, M Blangiardo, L Waller. A Bayesian hierarchical small-area population model accounting for data source specific methodologies from American Community Survey, Population Estimates Program, and Decennial Census data. Journal of Applied Statistics (submitted).
- EN Peterson, AB Moller, A Gemmill, D Chou, L Alkema. A Bayesian temporal hierarchical model to assess levels of misclassification error in national vital registration maternal mortality data. Statistics in Medicine (accepted).
- RC Nethery, JT Chen, N Krieger, PD Waterman, EN Peterson, LA Waller, BA Coull. Statistical implications of endogeneity induced by residential segregation in small-area modelling of health inequities. The American Statistician (accepted).
- NB Jain, JE Kuhn, GD Ayers, A Song, EN Peterson. Geographical Variation in Rates of Shoulder and Knee Arthroscopy in U.S. States and Relationship to Orthopedist Density in Surgeon Volume. JAMA. 2019. 11; 2(12). doi: 10.1001/jamanetworkopen.2019.17315
- NB Jain, GD Ayers, EN Peterson, MB Harris, L Morse, KC O’Connor, E Garshick. Traumatic spinal cord injury in the United States, 1993-2012. JAMA. 2015. 9;313(22). 2236-43.
- A Morandi, LM Solberg, R Habermann, P Cleeton, EN Peterson, EW Ely, J Schnelle. Documentation and Management of Words Associated with Delirium Among Elderly Patients in Postacute Care: A Pilot Investigation. JAMDA. 2009. 34-339.
- SF Simmons, EN Peterson, C You. The Accuracy of Monthly Weight Assessments in Nursing Homes: Implications for the Identification of Weight Loss. Journal of Nutrition, Health and Aging. 2009. 13, 3, 284-288.
- JF Schnelle, SF Simmons, L Beuscher, EN Peterson, R Habermann, F Leung. Prevalence of Constipation Symptoms in Nursing Home Residents. Journal of Gerontology. 2009.
Code & Software
- BEACON — Reproducible demo scripts for Bayesian methods developed in collaboration with Tribal Nations. Synthetic data replicates analyses where original data cannot be shared due to Tribal data governance agreements.
Teaching
Current Courses
- Emory BIOS 526: Modern Regression Methods
- Multilevel models
- Nonlinear models
- GEEs
- Splines
- GAMs
- Workflow Development
