Mad Cows & Computer Models: Study Shows USA Response to BSE Inadequate

New studies show USDA relies on flawed "computer models" to judge risk factors of BSE;

Monday, June 16, 2008
Mad Cows and Computer Models: The U.S. Response to BSE
NEW SOLUTIONS: A Journal of Environmental and Occupational Health Policy

Issue: Volume 18, Number 2 / 2008 Pages: 145 - 156

Mad Cows and Computer Models: The U.S. Response to BSE

Frank Ackerman and Wendy A. Johnecheck

Abstract:

The proportion of slaughtered cattle tested for BSE is much smaller in the U.S. than in Europe and Japan, leaving the U.S. heavily dependent on statistical models to estimate both the current prevalence and the spread of BSE. We examine the models relied on by USDA, finding that the prevalence model provides only a rough estimate, due to limited data availability. Reassuring forecasts from the model of the spread of BSE depend on the arbitrary constraint that worst-case values are assumed by only one of 17 key parameters at a time. In three of the six published scenarios with multiple worst-case parameter values, there is at least a 25% probability that BSE will spread rapidly. In public policy terms, reliance on potentially flawed models can be seen as a gamble that no serious BSE outbreak will occur. Statistical modeling at this level of abstraction, with its myriad, compound uncertainties, is no substitute for precautionary policies to protect public health against the threat of epidemics such as BSE.

http://baywood.metapress.com/app/home/contribution.asp?referrer=parent&backto=issue,5,18;journal,1,41;linkingpublicationresults,1:300327,1

Click on title above to read full article;

http://bse-atypical.blogspot.com/2008/06/mad-cows-and-computer-models-us.html