Get Newsletter
AlzRisk Risk Factor OverView
Back

Risk Factor:
Risk Factor Type: Behavior, Metabolic
Current Understanding:
The tables below present a modest number of reports whose results, taken collectively, suggest an inverse association between physical activity and risk for both Alzheimer disease (AD) and dementia. Overall, these data suggest that physical activity is a modifiable protective factor. Results from other lines of research corroborate the value of physical activity in relation to cognitive decline in older adults, and support many collateral benefits of physical activity as well. However, specific aspects of the relationship between physical activity and cognitive outcomes remain to be clarified, including the optimal duration, intensity, and timing during the lifespan of physical activity necessary to reduce cognitive risks, and the durability of the benefits of physical activity. For a review of the putative mechanisms by which physical activity may influence AD risk and detailed commentary on interpreting the findings below in a broader context, please view the Discussion.
Literature Extraction: Search strategy  * New *
Last Search Completed: 06 July 2016 - Last content update released on 17 October 2016.


Table 1:   Physical activity - categorical
Notes This table represents a variety of exposure definitions for leisure-time physical activity in relation to AD risk. Generally, the reference category corresponds to the least amount of physical activity; however, in the Abbott 2004 paper the reference category is represented by those with the highest number of miles walked per day.  
  Alzheimer Disease Total Dementia  
Paper Cohort Study Type # Subjects
(% Female)
Average Follow-up Time Exposure Distribution
# of Cases Effect Size 95% CI P-value # of Cases Effect Size 95% CI P-value Ethnicity Age at Start of Follow-up:
Mean (SD)
(Range)
Diagnostic Assessment Covariates & Analysis Comment Paper
Abbott, 2004 HAAS Incidence study reporting hazard ratios (HRs) 2257
(0%)
7.0 y
(detail)
< 0.25 miles/d: 27%
0.25 to 1 miles/d: 34%
> 1 to 2 miles/d: 19%
> 2 miles/d: 20%
(detail)
-
-
-
-
Total: 101
2.21
1.86
1.88
1.00
1.06-4.57
0.91-3.79
0.87-4.04
Ref.
0.03
0.09
0.11
Ref.
-
-
-
-
Total: 158
1.93
1.75
1.33
1.00
1.11-3.32
1.03-2.99
0.73-2.45
Ref.
0.02
0.04
0.35
Ref.
Japanese-American
(detail)
77 (4)
(71 - 93)
Screening: CASI

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, APOE4, BMI, CHD, DM, HTN, O, PP, TC‡
(detail)
Abbott, 2004
Akbaraly, 2009 3C Incidence study reporting hazard ratios (HRs) 5698
(61%)
-
(detail)
Low PA: 45%
Middle PA: 24%
High PA: 30%
(detail)
-
-
-
Total: 97
1.00
0.87
1.29
Ref.
0.50-1.51
0.80-2.09
Ref.
0.43
0.3
*
-
-
-
Total: 147
1.00
0.91
1.09
Ref.
0.59-1.39
0.73-1.63
Ref.
0.67
0.67
*
Caucasian
(detail)
74 (6)
(65 - )
Screening: BVRT, IST, MMSE

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, ADLI, APOE4, MMSE, DEP, DM, HC, HTN, MS, OS, SP, VD‡
(detail)
Akbaraly, 2009
Andel, 2003 HARMONY Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 3125
(61%)
31 y
(detail)
Physical activity in midlife
Hardly any PA: 11%
Light PA: 71%
Regular PA: 9%
Hard PA: 8%
(detail)

35
138
6
18
Total: 197

1.00
0.64
0.34
0.65

Ref.
0.41-1.00
0.14-0.86
0.33-1.29

Ref.
0.05
0.02
0.22

49
188
8
27
Total: 272

1.00
0.63
0.34
0.70

Ref.
0.43-0.91
0.16-0.72
0.40-1.24

Ref.
0.01
0.005
0.22
 (detail) 48 (5)
( - )
(detail)
Screening: BDRS, TELE

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, ALC, ANG, BMI, FVG, SM‡
(detail)
Andel, 2003
Buchman, 2012 MAP Incidence study reporting hazard ratios (HRs) 716
(76%)
3.5 y
(detail)
10th percentile: 10%
90th percentile: 10%
(detail)
-
-
Total: 71
1.00
0.53
Ref.
0.29-0.95
Ref.
0.03
*
 
 
      Caucasian
(detail)
82 (7)
( - )
AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, CA-Past, PA, SA‡
(detail)
Buchman, 2012
Gatz, 2006 HARMONY Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 196
(62%)
31 y
(detail)
Physical activity at midlife
Low PA:         
High PA:         
(detail)

-
-

1.00
0.67

Ref.
0.27-1.63

Ref.
0.3
*

-
-

1.00
1.54

Ref.
0.72-3.30

Ref.
0.27
*
 (detail) - (-)
( - )
(detail)
Screening: BDRS, TELE

AD Diagnosis: NINCDS ADRDA
(detail)
A, G‡
(detail)
Gatz, 2006
Gelber, 2012 HAAS Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 3468
(0%)
25 y
Low PA:         
High PA:         
(detail)
69
48
Total: 117
1.44
1.00
0.94-2.21
Ref.
0.09
Ref.
*
139
84
Total: 223
1.59
1.00
1.15-2.18
Ref.
0.004
Ref.
*
Japanese-American
52 (-)
( - )
Screening: CASI, HDS, MMSE, 3MSE

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, APOE4, CVD, CYJPN, DM, HC, HXHTN, OS‡ Gelber, 2012
Kishimoto, 2016 Hisayama Study Incidence study reporting hazard ratios (HRs) 803
(61%)
12 y
(detail)
Inactive: 67%
Active: 33%
(detail)
123
42
Total: 165
1.00
0.59
Ref.
0.41-0.84
Ref.
0.003
201
90
Total: 291
1.00
0.78
Ref.
0.60-1.01
Ref.
0.06
 (detail) 74 (-)
(65 - )
Screening: HDS, HDS-R, MMSE

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, ALC, AHD, BMI, DM, ECG, SMKH, SH, SBP, TC‡ Kishimoto, 2016
Larson, 2006 ACT-GHC Incidence study reporting hazard ratios (HRs) 1740
(60%)
6.2 y
(detail)
< 3 times/wk: 26%
> 3 times/wk: 74%
(detail)
-
-
Total: 107
1.00
0.69
Ref.
0.45-1.05
Ref.
0.08
52
106
Total: 158
1.00
0.68
Ref.
0.48-0.96
Ref.
0.03
Caucasian, Other, African-American (Black)
(detail)
74 (6)
(65 - )
(detail)
Screening: CASI

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, ALC, APOE4, CVD, CF, CHD, DEP, DM, DSU, HS, HTN, PP, SM‡§
(detail)
Larson, 2006
Laurin, 2001 CSHA Cumulative incidence study reporting odds ratios (ORs) 3848
(60%)
5.0 y
*

(detail)
None: 32%
Low PA: 13%
Moderate PA: 36%
High PA: 19%
(detail)
80
21
52
16
Total: 169
1.00
0.67
0.67
0.50
Ref.
0.39-1.14
0.46-0.98
0.28-0.90
Ref.
0.14
0.04
0.02
*
110
28
79
31
Total: 248
1.00
0.64
0.69
0.63
Ref.
0.41-1.02
0.50-0.95
0.40-0.98
Ref.
0.05
0.02
0.04
*
Caucasian
(detail)
- (-)
(65 - )
(detail)
Screening: 3MSE

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G‡
(detail)
Laurin, 2001
Lindsay, 2002 CSHA Cumulative incidence study reporting odds ratios (ORs) 4019
(61%)
5.0 y
*

(detail)
No regular PA: 29%
Regular PA: 71%
(detail)
80
108
Total: 188
1.00
0.69
Ref.
0.50-0.96
Ref.
0.03
*
 
 
       (detail) - (-)
(65 - )
Screening: "Modified" 3MSE

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G‡
(detail)
Lindsay, 2002
Podewils, 2005 CHCS Incidence study reporting hazard ratios (HRs) 3375
(59%)
5.4 y
(detail)
< 248 kcal/week: 25%
248-742 kcal/wk: 25%
743-1,657 kcal/wk: 25%
> 1,657 kcal/wk: 25%
(detail)
69
70
58
48
Total: 245
1.00
1.07
0.92
0.70
Ref.
0.73-1.57
0.62-1.39
0.44-1.33
Ref.
0.73
0.69
0.14
*
129
136
111
103
Total: 479
1.00
1.22
0.94
0.85
Ref.
0.93-1.60
0.69-1.28
0.61-1.19
Ref.
0.15
0.69
0.34
*
Caucasian, Other
75 (5)
(65 - )
Screening: IQ-CODE, 3MSE, Other, TICS

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, ADLI, APOE4, MMSE, IADLI, MRI, RE, SNS, SSS‡
(detail)
Podewils, 2005
Ravaglia, 2008 CSBA Incidence study reporting hazard ratios (HRs) 749
(54%)
3.9 y
(detail)
< 4,774 kcal/wk: 33%
4,774-8,090 kcal/wk: 33%
> 8,090 kcal/wk: 33%
(detail)
21
20
13
Total: 54
1.00
0.70
0.95
Ref.
0.33-1.49
0.50-1.80
Ref.
0.35
0.88
*
41
26
19
Total: 86
1.00
0.58
0.69
Ref.
0.32-1.06
0.41-1.15
Ref.
0.07
0.16
*
Caucasian
(detail)
73 (6)
(65 - )
Screening: MMSE

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, APOE4, C, CVD, CPD, CHD, DM, HTN, HCY‡
(detail)
Ravaglia, 2008
Rovio, 2005 NKP-FINMONICA-CAIDE Cumulative incidence study reporting odds ratios (ORs) 1449
(62%)
21 y
(detail)
Physical activity in midlife
Sedentary: 59%
Active: 41%
(detail)

31
10
Total: 41

1.00
0.35

Ref.
0.16-0.80

Ref.
0.01
*

38
15
Total: 53

1.00
0.47

Ref.
0.25-0.90

Ref.
0.02
*
 (detail) 51 (6)
(39 - 64)
Screening: MMSE

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, ALC, APOE234, BMI, LD, MI, SM, SH, SBP, TC‡§
(detail)
Rovio, 2005
Sattler, 2011 ILSE Cohort study reporting odds ratios (ORs) 300
(49%)
12 y
(detail)
Baseline physical activity
Inactive: 67%
Active: 33%
(detail)

16
8
Total: 24

1.00
0.93

Ref.
0.45-1.90

Ref.
0.84
*

 
 

 

 

 
 (detail) 62 (-)
( - )
(detail)
Screening: MMSE, Other, TMT

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, BMI, DEP, SES‡
(detail)
Sattler, 2011
Sattler, 2011 ILSE Cohort study reporting odds ratios (ORs) 300
(49%)
12 y
(detail)
Lifetime physical activity
Never active: 29%
Not continually active (vs. Never active): 61%
Always active (vs. Never active): 10%
(detail)

8
12
4
Total: 24

1.00
1.01
1.25†

Ref.
0.54-1.90
0.38-4.06

Ref.
0.98
0.71
*

 
 
 

 

 

 
 (detail) 62 (-)
( - )
(detail)
Screening: MMSE, Other, TMT

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, BMI, DEP, SES‡
(detail)
Sattler, 2011
Scarmeas, 2009 WHICAP Incidence study reporting hazard ratios (HRs) 1880
(69%)
5.4 y
(detail)
No PA: 25%
Some PA: 36%
Much PA: 39%
(detail)
71
84
69
Total: 224
1.00
0.75
0.67
Ref.
0.54-1.04
0.47-0.95
Ref.
0.08
0.02
 
 
 
      Caucasian, Other, Hispanic, African-American (Black)
77 (7)
( - )
Screening: Other

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G, APOE4, bsln BMI, Kcal, CF, CI, DEP, LA, O, RE, SM, SP‡
(detail)
Scarmeas, 2009
Tolppanen, 2015 NKP-FINMONICA-CAIDE Incidence study reporting hazard ratios (HRs) 1432
(62%)
28 y
Physical activity in midlife
Low PA: 24%
Medium PA: 35%
High PA: 41%
(detail)

56
70
68
Total: 194

1.40
1.43
1.00

0.97-2.02
1.02-2.01
Ref.

0.07
0.03
Ref.
*

66
85
81
Total: 232

1.39
1.45
1.00

0.99-1.95
1.06-1.97
Ref.

0.06
0.02
Ref.
*
 (detail) 51 (6)
( - )
Screening: CERAD, Informant interview, MMSE

AD Diagnosis: DSM IV, NINCDS ADRDA
(detail)
A, E, G, ASM, AF, BMI, CVD, CVS, JNTPN, CPD, CHF, CAD, DM, MS, MI, OPA, O, PA, RA, SM, SH‡
(detail)
Tolppanen, 2015
Tyas, 2001 MSHA Cumulative incidence study reporting odds ratios (ORs) 635
(62%)
5.0 y
(detail)
No PA: 42%
Yes PA: 58%
(detail)
17
14
Total: 31
1.00
0.70
Ref.
0.32-1.51
Ref.
0.37
*
 
 
       (detail) 74 (6)
(65 - 93)
(detail)
Screening: "Modified" 3MSE, 3MSE

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G‡ Tyas, 2001
Wilson, 2002a ROS Incidence study reporting hazard ratios (HRs) 733
(67%)
4.5 y
(detail)
Lowest PA: 25%
Low-mid PA: 25%
Mid-high PA: 25%
Highest PA: 25%
(detail)
-
-
-
-
Total: 111
1.00
0.71
0.73
0.61
Ref.
0.42-1.19
0.44-1.20
0.35-1.05
Ref.
0.2
0.22
0.08
*
 
 
 
 
      Caucasian, Other
75 (6)
(65 - )
Screening: CERAD

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G‡
(detail)
Wilson, 2002a
Yoshitake, 1995 Hisayama Study Incidence study reporting hazard ratios (HRs) 826
(60%)
6.1 y
*

(detail)
No PA: 63%
Yes PA: 37%
(detail)
-
-
Total: 44
1.00
0.20
Ref.
0.06-0.68
Ref.
0.01
*
 
 
Total: 103
      Japanese
(detail)
74 (6)
(65 - )
(detail)
Screening: HDS

AD Diagnosis: NINCDS ADRDA
(detail)
A, G‡
(detail)
Yoshitake, 1995
* Derived value.
† Five or fewer cases exist.
‡ Covariates: "A" (age), "E" (education), "G" (gender), "ADLI" (activities of daily living impairment), "ALC" (alcohol intake), "ANG" (angina), "AHD" (antihypertensive drug use), "APOE234" (APOE e2 e3 e4 genotype), "APOE4" (APOE e4 genotype), "ASM" (asthma), "AF" (atrial fibrillation), "bsln BMI" (baseline BMI), "MMSE" (baseline MMSE), "BMI" (body mass index), "Kcal" (caloric intake), "C" (cancer), "CVD" (cardiovascular disease), "CVS" (cardiovascular surgery), "CYJPN" (childhood years spent in Japan), "JNTPN" (chronic joint pain), "CPD" (chronic pulmonary disease), "CF" (cognitive function), "CI" (comorbidity index), "CHF" (congestive heart failure history), "CAD" (coronary artery disease), "CHD" (coronary heart disease), "DEP" (depression), "DM" (diabetes mellitus), "DSU" (dietary supplement use), "ECG" (electrocardiogram abnormalities), "FVG" (fruit and vegetable consumption), "HS" (health status), "HC" (high cholesterol), "HXHTN" (history of hypertension), "HTN" (hypertension), "IADLI" (instrumental activities of daily living impairment), "LA" (leisure activities), "LD" (locomotor disorders), "MRI" (magnetic resonance imaging white-matter-grade score), "MS" (marital status), "MI" (mycardial infarction history), "OPA" (occupational physical activity), "OS" (Occupational status), "O" (other), "CA-Past" (Past Cognitive Activity ), "PA" (physical activity), "PP" (physical performance), "HCY" (plasma homocysteine), "RE" (race/ethnicity), "RA" (rheumatoid arthritis), "SMKH" (smoking habits), "SM" (smoking status), "SA" (social Activity), "SNS" (social network score), "SSS" (social support score), "SES" (socioeconomic status), "SH" (stroke history), "SP" (study population), "SBP" (systolic blood pressure), "TC" (total cholesterol), "VD" (vascular disease)
§ Covariates for total dementia are different.
 
Table 2:   Physical activity - continuous (hours per week)
Notes These reports examine leisure-time physical activity modeled as a continuous variable (hours/week) in relation to AD risk.  
  Alzheimer Disease Total Dementia  
Paper Cohort Study Type # Subjects
(% Female)
Average Follow-up Time hrs/wk
Mean (SD)
(Range)
# of Cases Effect Size 95% CI P-value # of Cases Effect Size 95% CI P-value Ethnicity Age at Start of Follow-up:
Mean (SD)
(Range)
Diagnostic Assessment Covariates & Analysis Comment Paper
Wilson, 2002a ROS Incidence study reporting hazard ratios (HRs) 733
(67%)
4.5 y
(detail)
5.7 (8)
( - )
(detail)
111 1.00 0.97-1.02 1.0
*
        Caucasian, Other
75 (6)
(65 - )
Screening: CERAD

AD Diagnosis: NINCDS ADRDA
(detail)
A, E, G‡ Wilson, 2002a
Wilson, 2002b CHAP Cumulative incidence study reporting odds ratios (ORs) 835
(59%)
4.1 y
(detail)
3.5 (5)
( - )
(detail)
139 1.04 0.98-1.10 0.18
*
        Caucasian, African-American (Black)
(detail)
76 (4)
(65 - )
(detail)
Screening: Other

AD Diagnosis: NINCDS ADRDA
(detail)
A, G, APOE4, RE, SS‡
(detail)
Wilson, 2002b
* Derived value.
‡ Covariates: "A" (age), "E" (education), "G" (gender), "APOE4" (APOE e4 genotype), "RE" (race/ethnicity), "SS" (stratified sampling)
 
Table 3:   Physical activity - continuous (MET-hours per week)
Notes These reports examine leisure-time physical activity modeled as a continuous variable (per SD increase of MET-hours per week) in relation to AD risk.  
  Alzheimer Disease Total Dementia  
Paper Cohort Study Type # Subjects
(% Female)
Average Follow-up Time MET-hours/week
Mean (SD)
(Range)
# of Cases Effect Size 95% CI P-value # of Cases Effect Size 95% CI P-value Ethnicity Age at Start of Follow-up:
Mean (SD)
(Range)
Diagnostic Assessment Covariates & Analysis Comment Paper
de Bruijn, 2013 Rotterdam Study Incidence study reporting hazard ratios (HRs) 4406
(59%)
8.8 y
(detail)
75.5 (-)
( - )
(detail)
490 0.98 0.89-1.08 0.68
*
583 0.93 0.85-1.02 0.12
*
 (detail) 73 (7)
( - )
Screening: CAMDEX, GMS, MMSE

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)
A, E, G, APOE4, MMSE, BMI, DM, HDL, HTN, SM, TC‡
(detail)
de Bruijn, 2013
* Derived value.
‡ Covariates: "A" (age), "E" (education), "G" (gender), "APOE4" (APOE e4 genotype), "MMSE" (baseline MMSE), "BMI" (body mass index), "DM" (diabetes mellitus), "HDL" (HDL cholesterol), "HTN" (hypertension), "SM" (smoking status), "TC" (total cholesterol)