2021-2023 data from 823 populous US counties estimates contributions of motor vehicle emissions and cigarette smoking to age-adjusted cancer and heart disease mortality. The analysis, by Leon Robertson, links GPS-derived vehicle kilometers with survey data on smoking and other risk factors. It finds over 600,000 annual deaths attributable to these two factors.
Use Cases
- Model age-adjusted cancer mortality rates from daily vehicle kilometers and smoking prevalence data.
- Estimate heart disease mortality reductions by simulating zero vehicle emissions using regression coefficients.
- Analyze gender-specific mortality patterns (men and women separately) across 823 counties.
- Control for other low-intercorrelated risk factors like excessive alcohol use in mortality models.
Strengths
- Covers 823 US counties with populations over 50,000.
- Provides a 3-year temporal coverage from 2021 to 2023.
- Analyzes two major mortality causes (cancer and heart disease) and two primary risk factors.
Limitations
- Limited to populous counties, excluding rural areas and smaller populations.
- Specific sample sizes, row counts, and column details are unknown.
- Relies on statistical modeling and control of confounders, which may not capture all causal relationships.
Provenance
- Source
- Leon Robertson via Harvard Dataverse.
- Collection Method
- GPS and cell phone movement data for vehicle kilometers matched to survey data on risk factors and official mortality statistics.
- Time Range
- 2021-2023
- Freshness
- Data current through 2023, with repository entry updated in April 2026.
- Geography
- 823 populous counties in the United States.