Youth-Led Public Health Research · Santa Clara County
A data-driven analysis of obesity disparities across 370 census tracts reveals a stark east-west divide — and a roadmap for closing it.
Source: CDC PLACES (2025); U.S. Census Bureau, ACS (2019–2023); Random Forest Model (Test R² = 0.96)
Using CDC PLACES 2025 data and machine learning across 370 census tracts, this research identifies the strongest predictors of obesity at the neighborhood level — and where the greatest need for intervention lies.
Depression, smoking, and housing insecurity predict obesity better than race, income, or education — even when both sets of factors are given to the model together. Mental health is an obesity intervention, not a separate concern.
SHAP Analysis · Random ForestAn east/west fault line runs through the county, with roughly a 5× gap in composite risk scores between the highest and lowest burden areas — where structural disadvantage compounds the cycle of chronic stress and poor health outcomes.
Composite Risk Score · 370 TractsTracts with the highest obesity have the lowest checkup and cholesterol screening rates. The communities that need preventive care most are getting it least — a critical equity finding with direct policy implications.
Care Gap · Equity FindingThe composite risk score — combining RF prediction, depression, smoking, housing insecurity, and care gaps — identifies where intervention resources are most urgently needed.
Source: CDC PLACES (2025); U.S. Census Bureau, ACS (2019–2023)
* Predominantly Hispanic and Vietnamese, lower-income communities where structural disadvantage and limited access to mental health and preventive care compound existing health inequalities.
Sample project · Alum Rock · Low Income Households · Minority Communities
3 census tracts
COMBINED FOCUS GUIDANCE
Prioritize culturally-competent Medi-Cal enrollment through community health workers and bilingual access to county social services — language access must accompany all financial assistance programs.
Launch peer navigation at trusted community sites. Depression at 19.3% is the leading driver of health risk here — addressing stigma and access barriers is the entry point.
Whether you lead a community organization, shape public policy, or cover public health — there is a role for you in this work.
Community Organizations & Nonprofits
Every community faces its own combination of risk factors. We work with local organizations to identify the 4–5 highest-impact risk markers in your specific neighborhoods, then collaborate to design targeted, place-based actions and solutions — because no single approach fits every community.
Partner with us →Government & Policymakers
The research supports targeted investment in mental health access, housing stability, and preventive care in the highest-burden tracts. Policy briefs and custom solutions are available for city and county stakeholders.
Partner with us →Media & Press
A high school researcher using machine learning to expose a 5× health gap between East and West San Jose — and connecting the dots to depression, housing, and preventive care access. A press kit is available on request.
Request the press kit →Henri Smit is a community service leader and AI & data science practitioner with a deep passion for healthcare, public health, and the power of data to drive meaningful change in underserved communities.
This research grew out of Henri's role on the Youth Action Committee at Sacred Heart Community Service and his belief that data literacy is one of the most powerful tools available to the next generation of community advocates.
Whether you are a community organization, a policymaker, a journalist, or simply someone who cares about health equity in Santa Clara County — we would love to hear from you.