New Method for Analyzing Pesticide Data with Missing Values
Author Information
Author(s): Chen Haiying, Quandt Sara A., Grzywacz Joseph G., Arcury Thomas A.
Primary Institution: Wake Forest University School of Medicine
Hypothesis
Can distribution-based multiple imputation methods effectively handle bivariate pesticide data with values below the limit of detection?
Conclusion
The distribution-based multiple imputation method is a valid and feasible approach for analyzing bivariate data with values below the limit of detection.
Supporting Evidence
- The study demonstrated that imputed and observed values were consistent with the assumed distribution.
- Urinary APE concentration was significantly associated with potential pesticide poisoning symptoms.
- Results based on simple substitution methods differed significantly from those based on the MI method.
Takeaway
This study shows a new way to fill in missing pesticide data so scientists can better understand how pesticides affect health.
Methodology
The study used a simulation study and applied distribution-based multiple imputation methods to analyze longitudinal pesticide exposure data.
Potential Biases
Potential bias from using simple substitution methods instead of the proposed MI method.
Limitations
The method may not perform well with very high levels of left censoring or when the underlying distribution assumption is incorrect.
Participant Demographics
287 farmworkers from 11 counties in eastern North Carolina.
Statistical Information
P-Value
p<0.0001
Confidence Interval
1.00–1.14
Statistical Significance
p<0.0001
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website