A log-linear model for predicting risk factors for rabies positivity in raccoons in Virginia, 1984-1987

In response to an epidemic of rabies in the mid-Atlantic region, the Virginia Department of Health and the Consolidated Laboratory Services in 1982, redesigned their submission forms for animals being tested for rabies in an effort to elicit detailed information about the epidemiology of rabies in Virginia. The information collected from those submission forms was used in a mathematical model analysis of the epidemiology of raccoon rabies in Virginia for the years 1984 through 1987. Eleven explanatory variables and one response variable (positivity for rabies) were examined. The objective of this study was to develop a model, through logistic regression, that would explain the epidemiology of rabies in raccoons in Virginia, and determine the risk factors for prediction for positivity for rabies in raccoons in Virginia. This information would aid further surveillance efforts, preventive education programs, and in formulating future oral vaccination programs in raccoons. Multiway contingency tables were constructed (involving 2,3,4,5,and 6 way interactions), and log-linear models were fitted using an iterative fitting process to generate maximum likelihood estimates. The goodness of fit of each model was judged using the likelihood-ratio-chi-square p value (0.01). The backward stepwise model selection process was performed on logit models to find the best fitting model (0.01). The final model consisted of a combination of 17 four variable term models. Eight of the eleven explanatory variables remained important risk factors in the prediction of positivity for rabies in raccoons. To validate the model, it was applied to data collected in the years 1988 through July 1989. The model fit at the 0.01 level. Parameter estimates were calculated for each term in the model. All eight variables had main order effects (direct) on the response variable (positivity for rabies). Three second order effects were evident: age and season, behavior and season, and year and season. Future studies will involve applications of this model to other species to further explore the epidemiology of rabies, and to refine the model for practical applications.

Name thesis:degree_name

Ph. D.

Level thesis:degree_level

doctoral

Discipline thesis:degree_discipline

Veterinary Medical Sciences

Department dc:contributor.department

Veterinary Medical Sciences

Grantor dc:publisher

Virginia Tech

Year dc:date.issued

1990

Author dc:creator

• Torrence, Mary Elizabeth

Chair dc:contributor.committeechair

• Moore, David

Committee members dc:contributor.committeemember

• Lichtman, Marilyn V.
• Lee, J.C.
• Southard, Douglas R.
• Jenkins, Suzanne R.

Statement dc:rights

• In Copyright

Licence dc:rights.uri

• http://rightsstatements.org/vocab/InC/1.0/

Language dc:language.iso

en

Dc Identifier Other

etd-07282008-135746

OAI identifier oai:identifier

oai:vtechworks.lib.vt.edu:10919/38989