Get Newsletter
AlzRisk Paper Detail

Reference: Wilson, 2002b
Cohort: Chicago Health and Aging Project
Risk Factor: Physical Activity

Average Follow-up Time Detail
Follow-up began from October 1993 to May 1997.

"After an average interval of 4.1 years, 1,249 of those judged free of disease at baseline were sampled for clinical evaluation of incident disease. Of these, 109 died and 9 moved before they could be scheduled, leaving 1,131 persons, of whom 842 (74%) participated. Participation in the clinical evaluation was not related to age, sex, or education, but was slightly lower in black persons compared to white persons. Because there was insufficient information to diagnose AD in 7 persons, analyses are based on the remaining 835."

Exposure Detail
We assessed current frequency of participation in physical activities with questions from the 1985 Health Interview Survey21 adapted for use with older persons.22 The activities were walking for exercise, jogging or running, gardening or yard work, dancing, calisthenics or general exercise, golf, bowling, bicycle riding, and swimming or water exercises. Persons were asked if they had participated in the activity in the past 2 weeks and if so the number of times and average duration in minutes. Minutes in each activity were summed and divided by 120 to form a composite index of physical activity expressed as hours per week. Because weighting each activity by the estimated energy expended23 did not change the results reported below, we used total weekly hours in all analyses."

Ethnicity Detail
"Participation in the clinical evaluation was not related to age, sex, or education, but was slightly lower in black persons compared to white persons."

Age Detail
Age at start of follow-up is the age at enrollment.

Screening and Diagnosis Detail
Screening Method:

AD Diagnosis:
NINCDS ADRDA National Institute of Neurological and Communicative Diseases and Stroke/Alzheimer's Disease and Related Disorders Association Criteria (McKhann 1984)

"Persons sampled for ascertainment of incident AD underwent a uniform structured clinical evaluation with examiners blinded to previously collected data. The evaluation included a medical history, neurologic examination, detailed cognitive function testing, standard laboratory studies, informant interview for those with evidence of cognitive impairment, and brain MRI in those with cognitive impairment and inconclusive clinical evidence of stroke. On the basis of this evaluation, a board-certified neurologist diagnosed AD and other common conditions. The diagnosis of AD was based on the criteria of the joint working group of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA).19 These criteria require a history of cognitive decline and evidence of impairment in memory and at least one other domain of cognition. APOE genotyping was done by an investigator blinded to all clinical data using methods adapted from Hixson and Vernier.20"

Covariates & Analysis Detail
Analysis Type:
Logistic regression

"All [logistic regression] models included terms for age, sex, race, possession of one or more APOE {epsilon}4 alleles, the interaction of race and {epsilon}4, and time from baseline to follow-up. The core model also had terms for years of education and cognitive activity score. In additional analyses, we constructed separate models for black and white subjects; included terms for the interaction of cognitive activity with age, sex, and education; excluded persons with low memory scores at baseline; and added terms for medical conditions and depressive symptoms at baseline.
"Because of the skewed distribution of physical activity scores, we used a categorical version in a secondary analysis.
"Analyses were carried out in SAS28 with the models weighted to account for the stratified random sampling using standard error estimates obtained by a jackknife repeated replication procedure.29 Models were validated with graphical and analytic techniques to check for possible nonlinearity and interactions."

AD Covariates:
APOE4APOE e4 genotype
SSstratified sampling

Additional covariates included time from baseline to follow-up and an interaction between race and (epsilon)4.