Estimating Risk Factors for Pathogenic Dose Accrual From Longitudinal Data

Area of Science:

• Microbiology
• Epidemiology
• Biostatistics

Background:

• Estimating disease risk factors is vital for understanding etiology.
• Current statistical models for enteric pathogen diseases lack biological mechanism consideration, limiting association statements.
• Quantitative microbiological risk assessment provides a foundation for mechanistic modeling.

Purpose of the Study:

• To propose a novel, mechanistic approach for determining the association between risk factors and dose accrual rates for enteric pathogens.
• To develop regression parameters interpretable as dose accrual rate ratios.
• To provide a method for leveraging information across multiple pathogens.

Main Methods:

• Developed a mechanistic approach building on quantitative microbiological risk assessment.
• Incorporated previously ignored sources of variability.
• Leveraged information across multiple pathogens.
• Created an R package (dare) for the proposed methods.

Main Results:

• The proposed approach demonstrates higher biological plausibility and interpretability compared to traditional methods.
• Simulation studies showed generalized linear models had unacceptable coverage rates, while the proposed approach maintained nominal rates even when misspecified.
• Application to infant data from the PATHOME study identified environmental factors impacting enteric infections.

Conclusions:

• The proposed mechanistic approach offers a more robust and biologically plausible method for analyzing disease risk factors and dose accrual rates.
• This method improves upon existing statistical models, particularly for enteric pathogen-related diseases.
• The approach is validated through simulations and real-world application in infant enteric infection studies.