Researchers speaking here on September 26^th said this finding should not be viewed as the final word on a long controversy with years of contradictory results. Instead, looking at smoking as prognostic rather than predictive might help resolve the dilemma, said Sara M. Debanne, PhD, Professor of Epidemiology & Biostatistics, Case Western Reserve University, Cleveland, Ohio, United States.

"If you select smokers and nonsmokers and you follow them for the possible development of Alzheimer's, most of those studies appear to show that cigarette smoking increases the risk of Alzheimer's," Dr. Debanne said in an interview. "If instead, you choose people with and without Alzheimer's and you look back into their histories to see if they are smokers or not, you get the opposite result and it appears that fewer smokers have Alzheimer's disease."

The view of smoking as a prognostic factor is a new approach, she said. If future research confirms this role for smoking, it would explain the years of contradictory findings. The findings may help clinicians and families of a patient with Alzheimer's who smokes as they plan for a time that the patient will need supervised care.

Dr. Debanne and colleagues conducted a case-control study and quantified smoking history by pack-years. They then developed a multivariate logistic regression model in which they controlled for age, education, sex and the apolipoprotein E (APOE) e4 genotype, the gene implicated in hereditary forms of Alzheimer's disease.

The investigators then compared 315 patients with Alzheimer's disease against two control groups -- 394 subjects who were acquaintances of the cases and 176 who were selected randomly from a Medicare database. The investigators then coded pack-years recoded into 6 ordinal classes based on the distribution of pack-years in the first control group.

An initial unadjusted parallel analysis showed a trend towards a reduction in risk of Alzheimer's disease with increasing pack-years (P = .045). When the researchers analyzed the data by pack-years and compared subjects with less than 20 pack-years against those with 20 or more pack-years, they found an odds ratio of 0.373 for those with 20 or more pack-years, compared to those with less than 20 pack-years.

In further research, the investigative team will test whether smoking is a prognostic factor rather than a risk factor, she said.


[Presentation title: Alzheimers Disease and Smoking: A Dose-Response Association? Abstract 45]

Copyright © 2009 P\S\L Consulting Group Inc. All rights reserved. Republication or redistribution of P\S\L content is expressly prohibited without the prior written consent of P\S\L. P\S\L shall not be liable for any errors, omissions or delays in this content or any other content on its sites, newsletters or other publications, nor for any decisions or actions taken in reliance on such content.