Background

In 2005, nearly 19,000 people died after infection with antibiotic resistant bacteria in the United States, more than the number of people who died from AIDS. By decreasing the effectiveness of current drug therapies, antibiotic resistance increases morbidity and mortality in humans and animals. Antibiotics and antibiotic resistant bacteria are found in soil, surface water, and groundwater, yet their levels are currently unregulated. Quantifying and assessing the risk of antibiotic resistant pathogens, and understanding the factors that cause the emergence and spread of antibiotic resistance is imperative for the development of policy that effectively protects human health.

A mathematical model, parameterized with experimental data, that explains the mechanisms and speed with which antibiotic resistance emerges and spreads through the environment, is the first step in developing a scientific based risk assessment for resistant bacteria in the environment.

Methods

Due to the infectious nature of resistant bacteria, a population-level framework is needed. Therefore, we are expanding a population prevalence model that accounts for host-to-host transmission with a multimedia environmental model of antibiotic resistant genes. The combined model, shown below, will be used to examine the flows of antibiotic resistance between different environmental compartments and the effect of these flows on the risk of exposure from the environment.

The environmental model will describe the evolution, dissemination, and persistence of antibiotic resistant genes in soil, surface water, and air. Spatial heterogeneity is included through the addition of physical inputs that affect bacterial population growth or the emergence of resistance: the presence of metals and antibiotics, biofilms, temperature, nutrient levels, and saturation. These inputs and the flow of resistance genes from humans and animals will be parameterized based on experimental data.

We will use published data on enterobacteria to parameterize and validate the environmental multimedia model. Enterobacteria are important for their role as human pathogens, and significant data exist about their presence in the environment. To understand the interaction between different compartments in the environmental model, we will adapt diagnostic tools that are in use for toxicant multimedia models. Current microbial exposure assessment techniques will be used to estimate human exposure to environmental resistance. To conduct a sensitivity analysis assessing the most important factors driving risk and remaining unknowns we will use a Regional Sensitivity approach.

Conclusions/Implications

The central result of this modeling approach will be the ability to identify gaps in current data, plan experiments to fill those gaps, test theories, and estimate the potential impacts of intervention strategies and policy decisions. We expect that the human-environment-human pathway will significantly affect antibiotic resistance prevalence in a population. This model will increase the level of understanding of the pertinent factors in this pathway, and can be used to evaluate the consequences of different prevention strategies. This project will aid in the understanding of how the environment affects the emergence and transmission of antibiotic resistance in a population, and it will provide direction for future research that can be used to support a scientifically based risk assessment for this important public health issue.