Logistic Regression

Dear Students,

In reading the newpaper this morning, I came across an article that seems to present the results of a logistic regression study. Since it wasn't a long article, I'd like to share it with you in its entirety. See if you can detect the 2 things that caused me to think that a logistic regression was used. Don't look for the two words "logistic regression," however. Neither of those words appears in the article.

Study Links Parkinson's, Exposure to Pesticide

New research suggests that exposure to pesticides at home increases the risk of Parkinson's disease.

Patients newly diagnosed with Parkinson's were twice as likely to have been exposed to pesticides as people in a control group free of the disease, researchers have found.

"This study is the largest yet of newly diagnosed individuals with Parkinson's and the first to show a significant association between home pesticide use and the risk of developing the disease," said Lorene Nelson, lead researcher and a neuroepidemiologist at Stanford University's School of Medicine.

The study was presented Friday at the American Academy of Neurology's annual meeting in San Diego.

Damage to nerve cells in a region of the brian called the substantia nigra causes the movement difficulties that characterize Parkinson's disease. It's thought that some chemicals, including several commonly used in pesticides, have a particular affinity for this part of the brain. Occupational exposure to pesticides already has been associated with increased risk of the disease, which affects more than 500,000 people in the United States.

"If we could understand why these neurons are being killed in certain circumstances, we can then try to prevent it," Nelson said.

The Stanford study, which looked at the pesticide exposure of 496 people with Parkinson's and 541 disease-free people, found that those with Parkinson's were more than twice as likely to have past exposure to insecticides in the home. A somewhat weaker link was noted fer herbicides, but exposure to insecticides and fungicides in the garden was not found to be a risk factor.

Nelson cautioned that more studies are needed before any conclusive statements can be made about the causes of Parkinson's, including setting any specific guidelines for pesticide exposure.

Since some families have a history of the disease, many researchers suspect that one or more genes play a role in making some people susceptible to environmental triggers of the disease.

OK, that's the full article. Did you see the "signals" that the study's data were analyzed by means of a logistic regression? The first (and main) signal is the phrase "twice as likely" that appeared in the 2nd and 7th paragraphs. Second, this study had a dichotomous dependent variable: whether or not a person had Parkinson's disease.

Sky Huck

Copyright © 2012

Schuyler W. Huck
All rights reserved.

| Book Info | Author Info |

Site URL: www.readingstats.com

Top | Site Map
Site Design: John W. Taylor V