Table 31Physical activity and risk of developing AD

StudySample (n)Followup/EventsExposureCase definitionConfounding adjustmentResults
Andel et al., 2008196Community cohort (twins) (3134)31 years (mean)

197 AD
Time of exposure: Mid-life

Self-report responses to the question: “How much exercise have you had from age 25 to 50?”

Responses were coded: 0 (hardly any exercise), 1 (light exercise such as walking or light gardening), 2 (regular exercise involving sports), or 3 (hard physical training)
NINCDS-ADRDA
DSM
Age
Sex
Educational level
Diet
BMI
Alcohol use
Smoking
Angina
Adjusted OR (95% CI) for risk of AD compared with hardly any exercise:
Light exercise: 0.64 (0.41 to 1.00)
Regular (moderate) exercise: 0.34 (0.14 to 0.86)
Hard training: 0.65 (0.33 to 1.29)
Abbott et al., 2004197Community cohort (2257 )7 years

101 AD
Time of exposure: Late-life

Self-report responses to questions about the average amount of distance walked per day.
NINCDS-ADRDA
DSM
Age
APOE
Baseline CASI
Declines in activity since mid adulthood
Physical performance score
Education
BMI
Childhood years spent living in Japan
Occupation
Health conditions
Adjusted HR (95% CI) – reference is individuals who walked > 2 miles per day:
Those who walked < 0.25 miles per day: 2.21 (1.06 to 4.57)
Those who walked < 0.25 to 1 mile per day: 1.86 (0.91–3.79)
Those who walked 1 to 2 miles per day: 1.88 (0.87–4.04)
Laurin et al., 2001
Laurin, 2001199
Community cohort (4615)5 years

194 AD
Time of exposure: Late life
Self-report responses to two questions about frequency and intensity of exercise for individuals who reported physical activity.
Composite physical activity score categorized:
  1. “Low” = less than weekly
  2. “Moderate” = weekly
  3. “High“= ≥ 3 times weekly
NINCDS-ADRDA
DSM
Sex
Educational level
Family history of dementia
Tobacco use
Alcohol use
NSAID use
Daily living activities
Clinical variables
Adjusted OR (95% CI) for AD for levels of physical activity compared to no physical activity (adjusted for age, sex, and education):
Low activity: 0.67 (0.39 to 1.14)
Moderate activity: 0.67 (0.46 to 0.98)
High activity: 0.50 (0.28 to 0.90)

ORs increased somewhat when males and females were examined separately and when additional covariates were added to the model, with the exception that the ORs decreased (even lower risk) for females with high levels of physical activity
Larson et al. 2006198Clinical cohort (Group Health Cooperative, HMO) (1740)Mean 6.2 years (SD, 2.0)

107 AD
Time of exposure: Late-life

Self-report responses for the number of days per week during the past year the individual did the following activities for at least 15 minutes at a time:
Walking, hiking, bicycling, aerobics or calisthenics
Swimming, water aerobics, weight training
Stretching, or other exercise.
Responses dichotomized as “exercised regularly,” defined as self-report of exercise ≥ 3 times/week, vs. “did not exercise regularly.”
NINCDS-ADRDA
DSM
Age
Ethnicity
Sex
Educational level
Baseline cognitive function
Physical function
Depression
Health conditions
Lifestyle characteristics
Supplements
APOE
Risk of AD for those who exercised regularly compared to those who did not exercise regularly:
Age- and sex-adjusted HR: 0.64 (95% CI 0.43 to 0.96, p = 0.031)
HR adjusted for all potential confounders: 0.69 (95% CI 0.45 to 1.05; p = 0.081)

Risk reduction associated with exercise was greater in those with lower performance levels. (p = 0.021 for interaction of exercise with performance- based physical function)
Rovio et al. 2005200Community cohort (1449)Mean 21 years (SD 4.9)

76 AD
Time of exposure: Mid-life

Self-report responses to: “How often do you participate in leisure- time physical activity that lasts at least 20–30 minutes and causes breathlessness and sweating?”
Responses dichotomized:
“Active” = active ≥ 2 times/week
“Sedentary” = < 2 times/week
NINCDS-ADRDA
DSM
Age
Sex
Educational level
Followup time
Locomotor disorders
APOE
Clinical variables
Smoking status
Alcohol use
Risk of AD for “active”’ versus “sedentary”:
Adjusted OR: 0.35 (95% CI 0.16 to 0.80)
Physical activity had same effect on both sexes

APOE appears to be an effect modifier: among APOE carriers there is an association between physical activity and AD, but not among non-carriers (additive interaction RERI = 0.73, p = 0.02)
Rovio et al. 2007201Community cohort (1449)Mean 20.9 years (SD 4.9)

48 AD
Time of exposure: Mid-life

Self-report responses to the questions:
“How physically heavy is your work?”
Responses dichotomized as sedentary vs. active groups.
“How many minutes do you walk, bicycle, or have some other physical activity when you are going to and from work?”
Categorized as:
  1. not at all
  2. ≤ 59 minutes
  3. ≥ 60 minutes
NINCDS-ADRDA
DSM
Race
Educational level
Followup time
Locomotor symptoms
Occupation
Income at midlife
Leisure physical activity
APOE
Vascular disorder
Smoking status
Adjusted OR (95% CI) for risk of AD associated with high active vs. sedentary physical work activity: 1.90 (0.73 to 4.95)

Adjusted OR (95% CI) for risk of AD associated with physical activity during commuting:
No physical activity vs. moderate physical activity: 0.36 (0.13 to 0.96)
High physical activity vs. moderate activity: 0.48 (0.09 to 2.58)

No interactions with APOE
Verghese et al., 2003194Community cohort (469)Median 5.1 years

124 dementia, of which 86 were AD or mixed dementia
Time of exposure: Late-life

Self-report of frequency of involvement in the following 11 physical activities: Playing tennis or golf, swimming, bicycling, dancing, participating in group exercises, playing team games such as bowling, walking for exercise, climbing more than two flights of stairs, doing housework, and babysitting
Frequency of participation reported as “daily,” “several days per week,” “once weekly,” “monthly,” “occasionally,” or “never.”
Responses used to create index: 7 points for daily participation; 4 points for participating several days per week; 1point for participating once weekly; and 0 points for participating monthly, occasionally, or never. Summed the activity-days for each activity to generate a physical-activity score, ranging from 0 to 77.
Responses dichotomized as “rare participation” (once a week or less) versus “frequent participation” (several days a week or more).
NINCDS-ADRDA
DSM
Age
Sex
Educational level
Medical illness
Baseline Blessed test score
Participation in other leisure activities
Adjusted HR for dementia associated with dancing (frequent versus rare): HR 0.24 (95% CI 0.06 to 0.99)
No other physical activities showed a significant association with dementia

Adjusted HR (95% CI) for dementia for 1- point increment in the physical activity scale: 1.00 (0.98 to 1.03)

Adjusted HR (95% CI) for dementia using < 9 points on physical activity scale as comparison:
9 to 16 points: 1.44 (0.91 to 2.28)
> 16 points: 1.27 (0.78 to 2.06)
Podewils et al., 2005202Community cohort (3041)5.4 years
245 AD cases
Time of exposure: Late-life

Self-reported information about frequency and duration during the previous 2 weeks of the following activities: Walking, household chores, mowing, raking, gardening, hiking, jogging, biking, exercise cycling, dancing, aerobics, bowling, golfing, general exercise, and swimming.

Activities converted to number of kilocalories (kcal) expended per week and number of activities per week
NINCDS-ADRDAAge
Race
Sex
Educational level
Baseline cognitive status
Adjusted HR (95% CI) for AD for number of activities in previous 2 weeks versus 0 to 1 activity:
2 activities: 0.73 (0.49 to 1.08)
3 activities: 0.85 (0.57 to 1.29)
≥ 4 activities: 0.55 (0.34 to 0.88; p = 0.03)

Adjusted HR (95% CI) for AD associated with number of kilocalories expended per week compared to < 248 kcal/week:
248 to 742 kcal/week: 1.07 (0.73 to 1.57)
743 to 1657 kcal/week: 0.92 (0.62 to 1.39)
> 1657 kcal/week: 0.70 (0.44 to 1.13)
P for trend = 0.08

Association only significant for non-e4 carriers. P trend for kcal/week = 0.01 and p trend for number of activities = 0.001.
Yoshitake et al., 1995118Community cohort (577 at followup)7 years

42 AD
Time of exposure: Late-life

Self reported information about physical activity (four categories each for leisure and for work).

Defined the physically active group as those including daily exercise during the leisure period or moderate to severe physical activity at work.
NINCDS-ADRDA
DSM
Age
Sex
Baseline cognitive status
Adjusted HR (95% CI) for AD for physically active group versus non-active group: 0.20 (0.06 to 0.68)
Akbaraly et al., 2009195Community cohort (5692)4 years

105 AD cases
Time of exposure: Late-life

Self-report of daily frequency physical activities that included doing odd jobs, gardening, and going for a walk
NINCDS-ADRDA
DSM
Age
Sex
Study center (Dijon or Montpelier)
Marital status
Educational level
Occupational grade
Vascular risk factors:
  • - Diabetes
  • - HTN
  • - High cholesterol
  • - History of vascular disease
Depressive symptoms (CES-D > 16)
Physical function (instrumental ADL score > 0)
Cognitive impairment (MMSE score < 24)
APOE genotype
HR for AD compared to the lowest tertile of physical leisure activities:
High tertile: 1.29 (95% CI 0.80 to 2.09)
Mild tertile: 0.87 (95% CI 0.50 to 1.51)
Scarmeas et al., 200985Community cohort (1880)5.4 years

282 AD cases
Time of exposure: Late-life

Self-reported responses about number of times participating and number of minutes per time participating in 3 different categories of activities: vigorous (aerobic dancing, jogging, playing handball), moderate (bicycling, swimming, hiking, playing tennis), and light (walking, dancing, calisthenics, golfing, bowling, gardening, horseback riding).

Estimation of adherence to a Mediterranean diet based on self-reported responses on a semi- quantitative food frequency questionnaire
NINCDS-ADRDA
DSM
Age
Race
Sex
Educational level
BMI
Smoking
Depression
Leisure activities
Comorbid medical conditions
Baseline CDR score
APOE
Interval between 1st dietary and 1st physical activity measure
Caloric intake
Considered simultaneously, both high adherence to a Mediterranean-type diet and high physical activity level were associated with lower risk of AD
Compared to low diet score:
Middle diet score: HR 0.98 (95% CI 0.72 to 1.33)
High diet score: HR 0.60 (95% CI 0.42 to 0.87); P = 0.008 for trend

Compared to no physical activity:
Some physical activity: HR 0.75 (95% CI 0.54 to 1.04)
Much physical activity: HR 0.67 (0.47 to 0.95); P = 0.03 for trend
Ravaglia, et al., 2008203Community cohort (also included institutionali zed individuals)

(749)
3.9 years

54 AD cases
Time of exposure: Late-life

Self reported physical activity: (a) number of city blocks walked daily; (b) number of flights of stairs climbed daily; (c) frequency and duration of weekly participation during past year in occupational, recreational and sport activity
NINCDS-ADRDA
DSM
Age
Sex
Educational level
APOE
Cardiovascular disease
Hypertension
Hyperhomo- cysteinemia
Cerebrovascular disease
Diabetes
COPD
Cancer
ADL motor impairment
Physical activities were categorized into a dichotomous variable for the type of activity or the number of kilocalories expended in the activity. None of the categorizations of physical activity was significantly associated with incident AD. Some HRs were above 1.0 and some were less than 1.0.

Abbreviations: AD = Alzheimer’s disease; ADL = activities of daily living; APOE = apolipoprotein E gene; BMI = body mass index; CASI = Cognitive Abilities Screening Instrument; CDR = Clinical Dementia Rating scale; CES-D = Center for Epidemiologic Studies Depression scale; CI = confidence interval; COPD = chronic obstructive pulmonary disease; CVD = cardiovascular disease; DSM = Diagnostic and Statistical Manual of Mental Disorders; HMO = health maintenance organization; HR = hazard ratio; HTN = hypertension; MCI = mild cognitive impairment; MMSE = Mini-Mental State Examination; NINCDS-ADRDA = National Institute of Neurological and Communicative Diseases and Stroke-Alzheimer’s Disease and Related Disorders Association; NSAID = non-steroidal anti-inflammatory drug; OR = odds ratio; RERI = relative excess risk from interaction; SD = standard deviation

From: 3, Results

Cover of Preventing Alzheimer's Disease and Cognitive Decline
Preventing Alzheimer's Disease and Cognitive Decline.
Evidence Reports/Technology Assessments, No. 193.
Williams JW, Plassman BL, Burke J, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.