Obesity + life expectancy

Photograph of rising series of ridges taken by Mo Eid for Pexels
University of Michigan
Demographic projection, survival models, ordered logistic regression
STATA, R, Excel
project year



Lead researcher, with mentorship from senior scholar


Hispanics in the United States (US) hold a long-standing and paradoxical mortality advantage over non-Hispanic whites despite their relative socioeconomic disadvantage. Yet in the past few decades US population trends of increasing prevalence, intensity, and earlier onset of obesity have all disproportionately affected Hispanics. It remains unclear how and to what extent these trends could reshape future mortality patterns in the US. In this project, I draw upon multiple nationally representative population data sources to project the impact of observed, historical patterns in obesity prevalence among non-Hispanic whites and US-born and foreign-born Hispanics on mid-life life expectancy over the next three decades. These demographic projections provide a data-driven forecast of how obesity will likely impact life expectancy patterns among the adult population in the US in future years.


Project what obesity distributions will look like for male and female US-born Hispanics, foreign-born Hispanics, and non-Hispanic whites over the next three decades starting with a cohort of adults living in the US in 2015.

Estimate the impact of shifting obesity distributions on future life expectancy three decades into the future to evaluate whether the Hispanic paradox in mortality is expected to persist or erode.


In this project I used data from the National Health and Nutrition Examination Survey (NHANES), the National Health Interview Survey (NHIS), the National Center for Health Statistics (NCHS), and the US Census Bureau to project the impact of observed, historical patterns in obesity prevalence among non-Hispanic whites and US-born and foreign-born Hispanics on age-40 life expectancy (e40) over the next three decades. I followed a multi-stage analytic approach incorporating survival models, weight category transition probabilities from ordered logistic regression, and a series of stochastic projections based on weight category transitions and mortality risk in the baseline population. Somewhat counter to the expectations set by historical trends and early research published on this topic, I found that obesity is expected to have a large and comparable effect on US mortality for all racial-ethnic/nativity groups, and thus is unlikely to significantly reshape the existing Hispanic advantage in life expectancy within the foreseeable future.

End Result

Designed and executed a multi-stage, multi-method project incorporating four population data sources to present the first set of formal demographic projections detailing how widening racial-ethnic/nativity disparities in obesity prevalence and intensity will influence trends in life expectancy among the US population over the next three decades.


Not all research needs to start with new data collection efforts. Exploring and identifying new analytic purposes for existing data sources can also yield novel and impactful insights.

The complexity of the data strategy and novelty of the results resulting from this project has taught me how much value and insight lies in existing data sources that too often go under-utilized. Whether the goal is to forecast future obesity levels and the related life-expectancy impacts or to guide product development based on emergent trends in user behavior, there is a wealth of actionable insights to be made when existing data sources can be incorporated into new research strategies for continuous discovery.

A white arrow icon

Other Projects

Let's connect!

Interested in my work? Want to know more about my research skills? Have a job opportunity to share? Feel free to drop me a line.

Thank you for reaching out!
Oops! Something went wrong while submitting the form.