Investigating Air Pollution and Breast Cancer Mortality in California
Emanuel Alcala, PhD; Dilpreet Sekhon, M.P.H.; Tania Pacheco-Werner, PhD
Fresno State Central Valley Health Policy Institute
Background
In the United States, the most common cancer among women is breast cancer, 1 in 8 women at risk (ACS, 2023). At the state level, California’s age-adjusted breast cancer mortality (BCM) rate is 18.8 per 100,000 over the 5-year average of 2016-2020 (CDC,2023). Despite established risk factors for BC, there is emerging research on the impact of environmental factors on breast cancer mortality, mainly ambient air pollution (DuPre et al.,2019; Cheng et al., 2022). One of these air pollutants is particulate matter 2.5 (PM2.5), which is a group 1 carcinogen as defined by the World Health Organization (EPA, 2022; Zhou et al.,2022).
Methods
Secondary data was collected from the California Department of Public Health CDPH) Data and Vital Statistics, CalEnviroScreen 4.0, and the American Community Survey (ACS) from the Census. Multiple linear regression was used to test the significance between air pollution and age-standardized breast cancer mortality rates.
Figure 1. Particulate Matter 2.5 Concentrations at the Zip Code Level, California
Map description: PM2.5 concentrations are shown by zip code across California. The highest concentration category is concentrated in the San Joaquin Valley and parts of Southern California. Much of Northern California and the coast is in the lowest concentration category.
Concentrations are measured in micrograms per cubic meter (µg/m3).
| Map color | PM2.5 concentration (micrograms per cubic meter) |
|---|---|
| Yellow | 0.00–8.90 |
| Orange | 8.90–12.00 |
| Red | 12.00–15.93 |
Map scale: 0, 27, 54, 108, 162, and 216 miles.
Map credits: California State Parks, Esri, HERE, Garmin, FAO, NOAA, USGS, EPA.
Results
| Race/Ethnicity | Percentage of breast cancer mortality |
|---|---|
| Non-Hispanic White | 60% |
| Hispanic/Latinx | 18% |
| NH Asian | 11% |
| NH Black/African American | 9% |
| NH Pacific Islander | 1% |
| NH American Indian | 1% |
| Other | 0% |
NH = non-Hispanic.
- PM2.5 was significantly associated with BCM in both middle-aged group (b=1.53, 95% CI=[0.56,
2.51]) and the older age group (b=12.67, 95% CI=[10.91, 14.47]).
- A one SD increase in PM2.5 was associated with an increase in BCM rates by 12.7 per 100,000 women
- PM2.5 accounted for approximately 4% of the variance in BCM
- Diesel PM was significantly associated with BCM in middle-aged group (b=2.06, 95%
CI=[1.05, 3.07]) and the older age group (b=2.27, 95% CI=[0.35, 4.19]).
- Among the older age group, a one SD increase in Diesel PM was associated with an increase in BCM rates of 2.3 per 100,000 women.
Figure 3. Effect Size (beta) of Air Pollution Variables Estimated by Ordinary Least Squares Regression on Age-Adjusted Breast Cancer Mortality by Age Group
The table preserves the coefficients printed in Figure 3. Numerical confidence intervals are taken from the poster’s text. The older group’s PM2.5 coefficient is printed as 12.69 in Figure 3 and 12.67 in the text.
| Age group | Air pollutant | Effect size (beta) in Figure 3 | 95% confidence interval |
|---|---|---|---|
| Women Aged 45 to 64 | PM2.5 | 1.53 | 0.56–2.51 |
| Diesel PM2.5 | 2.06 | 1.05–3.07 | |
| Ozone | Positive; not numerically labeled | Not reported numerically; plotted interval crosses zero | |
| Women Aged 65+ | PM2.5 | 12.69 | 10.91–14.47 |
| Diesel PM2.5 | 2.27 | 0.35–4.19 | |
| Ozone | Negative; not numerically labeled | Not reported numerically; plotted interval crosses zero |
The effect-size axis runs from −1 to 4 in increments of 0.5 for women aged 45 to 64, and from −4 to 14 in increments of 2 for women aged 65+. The plotted PM2.5 and Diesel PM2.5 intervals are above zero in both age groups. Ozone intervals cross zero in both age groups.
Discussion
The model adjusted for race, poverty and insurance coverage. Along with air pollution, the percentage of the population that is privately insured was significantly associated with BCM in the older age group (b=26.85, 95% CI=[25.24,28.46]). The percentage of African Americans was also significantly associated with BCM in both the middle-aged group (b=2.79, 95% CI=[1.89,3.69]) and the older age group (b=2.75, 95% CI=[1.11,4.39]). While the findings may indicate statistical associations, it is essential to acknowledge the multifaceted nature of BCM, with various factors contributing to the observed patterns. Further research and a deeper exploration of potential mechanisms are warranted.
Conclusion
This is the first study to investigate the impact of air pollution and social determinants of breast cancer mortality in California at the zip code level. This study provides compelling evidence of a significant spatial association between PM2.5 and BCM rates with policy implications for stricter air quality regulations and urban planning policies. Further research is needed to establish causality and the mechanism of action at the population level.
For additional information please contact:
Emanuel Alcala, Ph.D.Assistant Director
ealcala@mail.fresnostate.edu