Dose-Response Model for Lassa Virus

ABSTRACT

This article develops dose-response models for Lassa fever virus using data sets found in the open literature. Dose-response data were drawn from two studies in which guinea pigs were given subcutaneous and aerosol exposure to Lassa virus. In one study, six groups of inbred guinea pigs were inoculated subcutaneously with doses of Lassa virus and five groups of out-bred guinea pigs were similarly treated. We found that the out-bred subcutaneously exposed guinea pig did not exhibit a dose-dependent trend in response. The inbred guinea pigs data were best fit by an exponential dose-response model. In a second study, four groups of out-bred guinea pigs were exposed to doses of Lassa virus via the aerosol route. In that study, aerosol diameter was less than 4.5 μ m and both mortality and morbidity were used as endpoints. The log-probit dose-response model provided a somewhat better fit than the Beta-Poisson model for data with mortality as the endpoint, but the Beta-Poisson is considered the best fit model because it can be derived using biological considerations. Morbidity data were best fit with an exponential dose-response model.

Key Words:

• Lassa fever virus
• dose-response
• microbial risk assessment
• exponential model
• beta-Poisson model
• log-probit model

ACKNOWLEDGMENTS

The advice of Timothy Bartrand (post-doctoral associate) and Mark Weir (doctoral student) is gratefully acknowledged. This work was performed as part of the Center for Advancing Microbial Risk Assessment (CAMRA). CAMRA is a USEPA/ Department of Homeland Security Cooperative Center of Excellence funded under the U.S. Environmental Protection Agency's (USEPA's) STAR grant R83236201.

This work does not express official policy of either the USEPA or the Department of Homeland Security.

Notes

**Failed test of trend.

*(Jahrling et al. 1982);

**Accepted as the best fit model.

**(Stephenson et al. 1984);

*Accepted best fit model.

**(Stephenson et al. 1984);

*Accepted best fit model.