A comprehensive study found significant associations between agricultural pesticide use patterns and increased cancer incidence across the United States.

In the study, published in Frontiers in Cancer Control and Society, investigators utilized county-level data to analyze the relationships between pesticide use, cancer rates, and various sociodemographic factors. 

The investigators employed latent class analysis to define pesticide use patterns based on data from 3,143 U.S. counties. They developed models using a 2-class through 8-class approach, ultimately selecting an 8-class model for their analysis. Model fit metrics for the 8-class model included a log-likelihood estimate of –110,193.8, Akaike's information criterion of 163.22, and Bayesian information criterion of 220,949.

The study analyzed 69 pesticides monitored by the U.S. Geological Survey. Generalized linear models were used to evaluate associations between pesticide use patterns and cancer incidence rates, adjusting for covariates such as smoking prevalence, socioeconomic vulnerability, agricultural land use, and total population.

Data sources included:

• Agricultural pesticide use data from the U.S. Geological Survey
• Cancer incidence rates from the National Institutes of Health and U.S. Centers for Disease Control and Prevention (CDC) State Cancer Profiles database (2015 to 2019)
• Social Vulnerability Index data
• Agricultural land use data from the U.S. Department of Agriculture
• Population data from the 2019 American Community Survey.

Among the key findings were:

• Certain pesticide use patterns were associated with an increased incidence of leukemia, non-Hodgkin lymphoma, bladder cancer, colorectal cancer, lung cancer, pancreatic cancer, and all cancers combined.
• For some cancer types, the increased risk associated with pesticide exposure was comparable to that of smoking.
• The Midwest region, characterized by heavy corn production, consistently showed the highest added cancer risk across multiple cancer types.
• An estimated 154,541 (95% confidence interval [CI] = 154,529–154,552) additional cancer cases per year were attributed to differences in pesticide use patterns between the highest and lowest risk regions.

The difference between the highest and lowest risk regions was 154,541 (95% CI = 154,529–154,552) additional cancer cases per year. The Midwest showed consistently elevated risk. The study's estimate for all cancer incidence was 449.0 (95% CI = 435.7–462.3) per 100,000 individuals compared with the CDC's reported median rate of 442.4 per 100,000 individuals for 2013 to 2017.

Findings related to specific cancer types:

• Colorectal cancer: 20,927 (95% CI = 20,925–20,929) additional cases per year in highest risk regions, primarily in the Great Plains, Midwest, and Mississippi River Basin.
• Pancreatic cancer: 3,835 (95% CI = 3,834–3,836) additional cases per year, with similar geographical patterns to colorectal cancer.
• Leukemia and non-Hodgkin lymphoma: 4,595 (95% CI = 4,594–4,596) and 7,608 (95% CI = 7,607–7,609) additional cases per year, respectively, associated with use patterns in the Midwest and West Coast.
• Lung Cancer: 24,362 (95% CI = 24,359–24,365) additional cases per year, concentrated in the Midwest.

Pesticide-cancer associations compared with smoking:

• Non-Hodgkin lymphoma: 154.1% more cases attributed to pesticides vs smoking
• All cancers, bladder cancer, leukemia: 18.7%, 19.3%, and 21.0% more cases, respectively, attributed to pesticides
• Pancreatic cancer: 3.4% more cases attributed to pesticides
• Colorectal cancer: 73% more cases attributed to smoking
• Lung cancer: 324.8% more cases attributed to smoking.

Specific pesticide associations:

• Atrazine: Consistently associated with regions of high added risk for all cancers and colorectal cancer
• Boscalid: Associated with high-risk regions for leukemia, non-Hodgkin lymphoma, and pancreatic cancer
• Dicamba: Associated with high-risk regions for colorectal and pancreatic cancers
• Glyphosate: Associated with high-risk regions for all cancers, colorectal cancer, and pancreatic cancer
• Dimethomorph: Associated with high-risk regions for leukemia and non-Hodgkin lymphoma, but low-risk regions for colorectal cancer
• Dimethenamid: Associated with low-risk regions for bladder cancer, but high-risk regions for pancreatic cancer when combined with dimethenamid-P
• Dinotefuran: Associated with high-risk regions for leukemia and non-Hodgkin lymphoma, but low-risk regions for colorectal cancer
• Imazethapyr: Associated with high-risk regions for all cancers, colorectal cancer, and lung cancer.

Regional details:

States consistently appearing in regions with the highest added risk included Iowa, Illinois, Nebraska, Missouri, Indiana, Ohio, and Florida. California, as the top vegetable-producing state with more than 1.2 million acres of vegetables harvested in 2017, showed increased risk for certain cancer types. Florida, ranking second highest in orange production nationally, also showed elevated risks for bladder cancer, leukemia, and non-Hodgkin lymphoma.

The investigators noted that without pesticide use, fruit, vegetable, and cereal production was estimated to decrease by 32% to 78%. Organic farms that did not use pesticides often had 15% to 50% lower yields compared with conventional farms. In 2021, herbicides were applied to 96% of the 93.4 million acres of corn planted in the United States

The Social Vulnerability Index was significantly associated with bladder cancer, leukemia, and non-Hodgkin lymphoma. Agricultural land use was significantly associated with all cancers and lung cancer. Total population was not significantly associated with any cancer type.

Limitations:

• Ecological study design prevented causal inference or individual-level risk assessment
• Data censoring in some counties because of small populations
• Heterogeneity in county size and population
• Transient populations like seasonal farmworkers not accounted for
• Melanoma not included as a result of unique risk factors
• Lack of independent validation data.

The study provided a comprehensive analysis of pesticide use patterns and their potential associations with cancer incidence across the United States, offering insights for further research and public health considerations.

The authors declared that no financial support was received for the research, authorship, and/or publication of this article.