Get Newsletter
AlzRisk Risk Factor OverView
Back

Risk Factor:
Risk Factor Type: Medications
Current Understanding:
The tables below summarize results from a series of observational studies of statin use in relation to AD dementia risk. Although there is significant heterogeneity of findings across studies, there is currently insufficient evidence to conclude that statin use affects risk of Alzheimer’s disease or all cause dementia. While research considering statin use at or within one year of the time of dementia consistently reports protective associations between statin use and Alzheimer’s disease, such studies are particularly susceptible to bias from and reverse causation (given that persons with cognitive deterioration may be less likely to be prescribed or continue taking a statin). Conversely, consistent with the results of randomized clinical trials, studies that tracked AD dementia risk over several years following assessment of statin use at a set “baseline” individually and collectively report no association. However, the relation between statin use and Alzheimer’s disease in the general population remains an open question. Most of these studies were conducted prior to the wide adoption of statin use for primary prevention. They are limited by small numbers of statin users and could not consider the impact of use of statins in midlife or sustained statin use from midlife to late life on dementia risk. Regardless, statins have demonstrated benefits on cardiovascular disease and mortality and are therefore are increasingly widely recommended for both primary and secondary prevention of cardiovascular events and associated mortality. For a more in-depth discussion of these studies considering the association between statin use and AD, please see the Discussion. An expanded examination of the relation between statins and dementia or cognitive decline can be found in the related published review, Power MC, Weuve J, Sharrett AR, Blacker D, Gottesman RF (Nature Reviews Neurology 2015, doi:10.1038/nrneurol.2015.35.
Literature Extraction: Search strategy  * New *
Last Search Completed: 20 March 2015 - Last content update released on 30 March 2015.


Table 1:   Baseline use of statins (yes vs. no)
Notes These reports evaluate the relation of statin use, measured at a set baseline, to risk of incident Alzheimer disease (AD) over subsequent years. All reported estimates compared AD risk among statin users with AD risk among non-users.  
  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
Ancelin, 2012 3C Incidence study reporting hazard ratios (HRs) 2784
(0%)		 -
(detail)
Men only
Did not use:         
Used:         
(detail)
-
-
Total: 126
1.00
1.09
Ref.
0.67-1.76
Ref.
0.73
*
-
-
Total: 193
1.00
0.81
Ref.
0.53-1.23
Ref.
0.33
*		 French
 74 (5)
(65 - )		 Screening: BVRT, IST, MMSE, Neuropsych Testing, Other, TMT

AD Diagnosis: DSM IV, NINCDS ADRDA
(detail)		 A, E, CC‡
(detail)		 Ancelin, 2012
Ancelin, 2012 3C Incidence study reporting hazard ratios (HRs) 4272
(100%)		 -
(detail)
Women only
Did not use:         
Used:         
(detail)
-
-
Total: 206
1.00
1.17
Ref.
0.80-1.70
Ref.
0.41
*
-
-
Total: 290
1.00
1.20
Ref.
0.88-1.64
Ref.
0.25
*		 French
 74 (5)
(65 - )		 Screening: BVRT, IST, MMSE, Neuropsych Testing, Other, TMT

AD Diagnosis: DSM IV, NINCDS ADRDA
(detail)		 A, E, CC‡
(detail)		 Ancelin, 2012
Arvanitakis, 2008 ROS Incidence study reporting hazard ratios (HRs) 877
(-)		 -
(detail)
Lipophilic statin use
Did not use: 92%
Used: 8%
(detail)
-
-
1.00
1.05
Ref.
0.57-1.95
Ref.
0.88
*
 
 
 
 
 		 Caucasian
(detail)		 75 (7)
( - )		 Screening: CERAD, MMSE, Neuropsych Testing

AD Diagnosis: Autopsy, CERAD, Neuropsychological examination
(detail)		 A, E, G‡ Arvanitakis, 2008
Arvanitakis, 2008 ROS Incidence study reporting hazard ratios (HRs) 862
(-)		 -
(detail)
Less lipophilic statin use
Did not use: 94%
Used: 6%
(detail)
-
-
1.00
0.71
Ref.
0.29-1.74
Ref.
0.45
*
 
 
 
 
 		 Caucasian
(detail)		 75 (7)
( - )		 Screening: CERAD, MMSE, Neuropsych Testing

AD Diagnosis: Autopsy, CERAD, Neuropsychological examination
(detail)		 A, E, G‡ Arvanitakis, 2008
Arvanitakis, 2008 ROS Incidence study reporting hazard ratios (HRs) 929
(69%)		 -
(detail)
Did not use: 87%
Used: 13%
(detail)		 175
16
Total: 191		 1.00
0.91		 Ref.
0.54-1.52		 Ref.
0.72
*		  
 		       Caucasian
(detail)		 75 (7)
( - )		 Screening: CERAD, MMSE, Neuropsych Testing

AD Diagnosis: Autopsy, CERAD, Neuropsychological examination
(detail)		 A, E, G‡ Arvanitakis, 2008
Zandi, 2005 Cache County Study Incidence study reporting hazard ratios (HRs) 3297
(-)		 3.3 y
*
Did not use: 93%
Used: 7%
(detail)		 98
4
Total: 102		 1.00
1.19†		 Ref.
0.35-2.96		 Ref.
0.75
*		 174
8
Total: 182		 1.00
1.19		 Ref.
0.53-2.34		 Ref.
0.65
*		 Caucasian
(detail)		 - (-)
(65 - )
(detail)		 Screening: "Modified" 3MSE, DQ, Informant interview, IQ-CODE, Neuropsych Testing

AD Diagnosis: Brain Imaging, DSM IIIR, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, Age*APOE4, APOE4, DM, HXHTN‡ Zandi, 2005
* Derived value.
 † Five or fewer cases exist.
 ‡ Covariates: "A" (age), "E" (education), "G" (gender), "Age*APOE4" (Age*APOE4 interaction), "APOE4" (APOE e4 genotype), "CC" (city center), "DM" (diabetes mellitus), "HXHTN" (history of hypertension)
 
Table 2:   Statin use at time of diagnostic assessment
Notes These reports examine statin use at the time of the diagnostic assessment (or 1 year prior) in relation to AD risk. These estimates differ from those of Table 1 in that updates to statin use after baseline (also called time-updated use) were incorporated into the analyses. Reported estimates compare AD risk among persons who used statins with risk among person who did not use statins, where definitions of use and non-use vary by study. Also included in this table are estimates that incorporated other post-baseline data on statin use in alternative ways.  
  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
Bettermann, 2012 GEMS Incidence study reporting hazard ratios (HRs) 2587
(-)		 -
(detail)
Never used: 75%
Ever used: 25%
(detail)		 -
-
Total: 212		 1.00
0.83		 Ref.
0.61-1.12		 Ref.
0.21		 -
-
Total: 324		 1.00
0.77		 Ref.
0.60-0.98		 Ref.
0.03		 Caucasian, African-American (Black)
(detail)		 - (-)
(75 - )		 Screening: "Modified" 3MSE, CDR, ADAS_cog, Informant interview, Neuropsych Testing, TICS

AD Diagnosis: Brain Imaging, DSM IV, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, APOE4, CHD, FC, RE, SH, TRTGRP‡ Bettermann, 2012
Bettermann, 2012 GEMS Incidence study reporting hazard ratios (HRs) -
(-)		 -
(detail)
Lipophilic statin use
Never used:         
Ever used:         
(detail)
-
-
1.00
0.79
Ref.
0.58-1.08
Ref.
0.14
-
-
1.00
0.76
Ref.
0.60-0.98
Ref.
0.03		 Caucasian, African-American (Black)
(detail)		 - (-)
(75 - )		 Screening: "Modified" 3MSE, CDR, ADAS_cog, Informant interview, Neuropsych Testing, TICS

AD Diagnosis: Brain Imaging, DSM IV, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, APOE4, CHD, FC, OLLM, RE, SH, TRTGRP‡ Bettermann, 2012
Bettermann, 2012 GEMS Incidence study reporting hazard ratios (HRs) -
(-)		 -
(detail)
Non-lipophilic statin use
Never used:         
Ever used:         
(detail)
-
-
1.00
1.01
Ref.
0.65-1.58
Ref.
0.96
-
-
1.00
0.89
Ref.
0.61-1.29
Ref.
0.54		 Caucasian, African-American (Black)
(detail)		 - (-)
(75 - )		 Screening: "Modified" 3MSE, CDR, ADAS_cog, Informant interview, Neuropsych Testing, TICS

AD Diagnosis: Brain Imaging, DSM IV, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, APOE4, CHD, FC, OLLM, RE, SH, TRTGRP‡ Bettermann, 2012
Bettermann, 2012 GEMS Incidence study reporting hazard ratios (HRs) 1872
(-)		 -
(detail)
Never used:         
Newly initiated:         
(detail)		 -
-		 1.00
0.46		 Ref.
0.29-0.74		 Ref.
0.001
*		 -
-		 1.00
0.53		 Ref.
0.37-0.75		 Ref.
0.0004
*		 Caucasian, African-American (Black)
(detail)		 - (-)
(75 - )		 Screening: "Modified" 3MSE, CDR, ADAS_cog, Informant interview, Neuropsych Testing, TICS

AD Diagnosis: Brain Imaging, DSM IV, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, APOE4, CHD, FC, RE, SH, TRTGRP‡ Bettermann, 2012
Bettermann, 2012 GEMS Incidence study reporting hazard ratios (HRs) -
(-)		 -
(detail)
Lipophilic statin use
Never used:         
Newly initiated:         
(detail)
-
-
1.00
0.47
Ref.
0.29-0.77
Ref.
0.003
-
-
1.00
0.53
Ref.
0.36-0.77
Ref.
0.001		 Caucasian, African-American (Black)
(detail)		 - (-)
(75 - )		 Screening: "Modified" 3MSE, CDR, ADAS_cog, Informant interview, Neuropsych Testing, TICS

AD Diagnosis: Brain Imaging, DSM IV, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, APOE4, CHD, FC, OLLM, RE, SH, TRTGRP‡ Bettermann, 2012
Bettermann, 2012 GEMS Incidence study reporting hazard ratios (HRs) -
(-)		 -
(detail)
Non-lipophilic statin use
Never used:         
Newly initiated:         
(detail)
-
-
1.00
0.78
Ref.
0.34-1.80
Ref.
0.56
-
-
1.00
0.89
Ref.
0.46-1.69
Ref.
0.71		 Caucasian, African-American (Black)
(detail)		 - (-)
(75 - )		 Screening: "Modified" 3MSE, CDR, ADAS_cog, Informant interview, Neuropsych Testing, TICS

AD Diagnosis: Brain Imaging, DSM IV, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, APOE4, CHD, FC, RE, SH, TRTGRP‡ Bettermann, 2012
Beydoun, 2011 BLSA Incidence study reporting hazard ratios (HRs) 1604
(39%)		 -
(detail)
Never used: 93%
Ever used: 7%
(detail)		 -
-
Total: 178		 1.00
0.30		 Ref.
0.10-0.95		 Ref.
0.04
*		 -
-
Total: 252		 1.00
0.41		 Ref.
0.18-0.92		 Ref.
0.03
*		 Caucasian, African-American (Black)
 58 (18)
(50 - )		 Screening: Blessed, CDR, DQ

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, E, G, BMI, CVD, DLM, HTN, RE, SM, SBPठBeydoun, 2011
Haag, 2009 Rotterdam Study Incidence study reporting hazard ratios (HRs) -
(-)		 -
(detail)
Lipophilic statin use
Never used:         
Ever used:         
(detail)
438
23
Total: 461
1.00
0.54
Ref.
0.32-0.89
Ref.
0.02
*
 
 
 
 
 		 Dutch
(detail)		 69 (9)
(55 - )
(detail)		 Screening: CAMDEX, GMS, MMSE, Neuropsych Testing, Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, E, G, BMI, CVD, CS, DM, SM, SBP, TC, CHLD‡ Haag, 2009
Haag, 2009 Rotterdam Study Incidence study reporting hazard ratios (HRs) -
(-)		 -
(detail)
Non-lipophilic statin use
Never used:         
Ever used:         
(detail)
438
9
Total: 447
1.00
0.54
Ref.
0.26-1.11
Ref.
0.1
*
 
 
 
 
 		 Dutch
(detail)		 69 (9)
(55 - )
(detail)		 Screening: CAMDEX, GMS, MMSE, Neuropsych Testing, Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, E, G, BMI, CVD, CS, DM, SM, SBP, TC, CHLD‡ Haag, 2009
Haag, 2009 Rotterdam Study Incidence study reporting hazard ratios (HRs) 6992
(60%)		 9.2 y
Status 1 year prior to AD assessment
Never used:         
Ever used:         
(detail)
-
-
1.00
0.51
Ref.
0.31-0.83
Ref.
0.007
*
 
 
 
 
 		 Dutch
(detail)		 69 (9)
(55 - )
(detail)		 Screening: CAMDEX, GMS, MMSE, Neuropsych Testing, Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, E, G, BMI, CVD, CS, DM, SM, SBP, TC, CHLD‡ Haag, 2009
Haag, 2009 Rotterdam Study Incidence study reporting hazard ratios (HRs) 6992
(60%)		 9.2 y
Never used:         
Ever used:         
(detail)		 438
28
Total: 466		 1.00
0.57		 Ref.
0.37-0.90		 Ref.
0.01
*		  
 		       Dutch
(detail)		 69 (9)
(55 - )
(detail)		 Screening: CAMDEX, GMS, MMSE, Neuropsych Testing, Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, E, G, BMI, CVD, CS, DM, SM, SBP, TC, CHLD‡ Haag, 2009
Li, 2010 ACT-GHC Incidence study reporting hazard ratios (HRs) 3099
(59%)		 6.1 y
Never used: 77%
Used: 23%
(detail)		 -
-
Total: 263		 1.00
0.62		 Ref.
0.40-0.97		 Ref.
0.03
*		  
 		       Caucasian
 75 (6)
(65 - )		 Screening: CASI, Neuropsych Testing, Other

AD Diagnosis: DSM IV, NINCDS ADRDA
(detail)		 A, E, G, APOE4, CASI, BMI, CHRT, CMVD, LLA, RE, SM‡
(detail)		 Li, 2010
Rea, 2005 CHCS Incidence study reporting hazard ratios (HRs) 2798
(-)		 -
*
Status 1 year prior to AD assessment
Never used: 91%
Ever used: 9%
(detail)
216
21
Total: 237
1.00
1.21
Ref.
0.76-1.91
Ref.
0.42
*
428
38
Total: 466
1.00
1.08
Ref.
0.77-1.52
Ref.
0.66
*		 Caucasian
(detail)		 - (-)
(65 - )
(detail)		 Screening: Brain Imaging, MMSE, Other
(detail)		 A, E, G, MMSE, BSTR, BCHD‡ Rea, 2005
Rea, 2005 CHCS Incidence study reporting hazard ratios (HRs) -
(-)		 -
Lipophilic statin use 1-year prior to AD assessment
Never used:         
Ever used:         
(detail)
-
-
1.00
1.03
Ref.
0.57-1.86
Ref.
0.92
*
-
-
1.00
0.94
Ref.
0.61-1.44
Ref.
0.78
*		 Caucasian
(detail)		 - (-)
(65 - )
(detail)		 Screening: Brain Imaging, MMSE, Other
(detail)		 A, E, G, MMSE, BSTR, BCHD‡ Rea, 2005
Rea, 2005 CHCS Incidence study reporting hazard ratios (HRs) -
(-)		 -
*
Less lipophilic statin use 1-year prior to AD assessment
Never used:         
Ever used:         
(detail)
-
-
1.00
1.58
Ref.
0.80-3.11
Ref.
0.19
*
-
-
1.00
-
Ref.
-
Ref.
-		 Caucasian
(detail)		 - (-)
(65 - )
(detail)		 Screening: Brain Imaging, MMSE, Other
(detail)		 A, E, G, MMSE, BSTR, BCHD‡ Rea, 2005
Rea, 2005 CHCS Incidence study reporting hazard ratios (HRs) 2798
(-)		 5.4 y
*
Never used: 91%
Ever used: 9%
(detail)		 -
-
Total: 245		 1.00
0.59		 Ref.
0.32-1.10		 Ref.
0.09
*		 -
-
Total: 480		 1.00
0.69		 Ref.
0.46-1.02		 Ref.
0.07
*		 Caucasian
(detail)		 - (-)
(65 - )
(detail)		 Screening: Brain Imaging, MMSE, Other
(detail)		 A, E, G, MMSE, BSTR, BCHD‡ Rea, 2005
Sparks, 2008 ADAPT Incidence study reporting hazard ratios (HRs) 1970
(-)		 -
(detail)
Never used: 63%
Consistently used: 37%
(detail)		 20
4
Total: 24		 1.00
0.33†		 Ref.
0.11-0.98		 Ref.
0.05
*		  
 		       Caucasian
(detail)		 - (-)
(70 - )		 Screening: Brain Imaging, CAB, Informant interview, LAB, MMSE, Neuropsych Testing, Other

AD Diagnosis: DSM IV, NINCDS ADRDA
(detail)		 A, E, G, APOE4‡ Sparks, 2008
Zandi, 2005 Cache County Study Incidence study reporting hazard ratios (HRs) 3274
(-)		 3.1 y
*
Does not use: 83%
Currently uses: 17%
(detail)		 78
6
Total: 84		 1.00
0.85		 Ref.
0.32-1.88		 Ref.
-		 136
15
Total: 151		 1.00
1.04		 Ref.
0.56-1.81		 Ref.
0.9
*		 Caucasian
(detail)		 - (-)
(65 - )
(detail)		 Screening: "Modified" 3MSE, DQ, Informant interview, IQ-CODE, Neuropsych Testing

AD Diagnosis: Brain Imaging, DSM IIIR, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, Age*APOE4, APOE4, DM, HXHTN‡ Zandi, 2005
† Five or fewer cases exist.
 ‡ Covariates: "A" (age), "E" (education), "G" (gender), "Age*APOE4" (Age*APOE4 interaction), "APOE4" (APOE e4 genotype), "CASI" (baseline CASI score), "MMSE" (baseline MMSE), "BSTR" (baseline stroke), "BCHD" (basline CHD), "BMI" (body mass index), "CVD" (cardiovascular disease), "CS" (cerebrovascular symptoms), "CHRT" (cohort), "CMVD" (comorbid vascular disease), "CHD" (coronary heart disease), "DM" (diabetes mellitus), "DLM" (dyslipidemia), "FC" (field center), "HXHTN" (history of hypertension), "HTN" (hypertension), "LLA" (other lipid lowering agent use), "OLLM" (other lipid lowering drug use), "RE" (race/ethnicity), "SM" (smoking status), "SH" (stroke history), "SBP" (systolic blood pressure), "TC" (total cholesterol), "TRTGRP" (treatment group (not the exposure of interest)), "CHLD" (use of a cholesterol-lowering drug other than a statin)
 § Covariates for total dementia are different.
 
Table 3:   Statin use at time of diagnosis (current vs. former vs. never)
Notes These reports examine statin use at the time of the diagnostic assessment (or 1 year prior) in relation to AD risk. These estimates differ from those of Table 1 in that updates to statin use after baseline (also called time-updated use) were incorporated into the analyses. Reported estimates compared AD risk among among current, former or never users of statins. Also included in this table are estimates that incorporated other post-baseline data on statin use in alternative ways.  
  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
Bettermann, 2012 GEMS Incidence study reporting hazard ratios (HRs) -
(-)		 -
(detail)
Never used:         
Formerly used:         
Currently used:         
(detail)		 -
-
-		 1.00
1.39
0.69		 Ref.
0.89-2.15
0.49-0.97		 Ref.
0.14
0.03		 -
-
-		 1.00
0.98
0.71		 Ref.
0.66-1.46
0.55-0.93		 Ref.
0.92
0.01		 Caucasian, African-American (Black)
(detail)		 - (-)
(75 - )		 Screening: "Modified" 3MSE, CDR, ADAS_cog, Informant interview, Neuropsych Testing, TICS

AD Diagnosis: Brain Imaging, DSM IV, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, APOE4, CHD, FC, OLLM, RE, SH, TRTGRP‡ Bettermann, 2012
Bettermann, 2012 GEMS Incidence study reporting hazard ratios (HRs) -
(-)		 -
(detail)
New initiators and never users
Never used:         
Formerly used:         
Currently used:         
(detail)
-
-
-
1.00
0.48
0.45
Ref.
0.20-1.13
0.27-0.76
Ref.
0.09
0.003
-
-
-
1.00
0.47
0.53
Ref.
0.23-0.95
0.36-0.79
Ref.
0.04
0.002		 Caucasian, African-American (Black)
(detail)		 - (-)
(75 - )		 Screening: "Modified" 3MSE, CDR, ADAS_cog, Informant interview, Neuropsych Testing, TICS

AD Diagnosis: Brain Imaging, DSM IV, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, APOE4, CHD, FC, OLLM, RE, SH, TRTGRP‡ Bettermann, 2012
Rea, 2005 CHCS Incidence study reporting hazard ratios (HRs) 2798
(-)		 -
*
Status 1-year prior to AD assessment
Never used: 91%
Formerly used: 2%
Currently used: 7%
(detail)
216
8
13
Total: 237
1.00
2.54
0.92
Ref.
1.24-5.20
0.52-1.62
Ref.
0.01
0.42
*
428
12
26
Total: 466
1.00
1.88
0.90
Ref.
1.05-3.36
0.61-1.35
Ref.
0.03
0.6
*		 Caucasian
(detail)		 - (-)
(65 - )
(detail)		 Screening: Brain Imaging, MMSE, Other
(detail)		 A, E, G, MMSE, BSTR, BCHD‡ Rea, 2005
* Derived value.
 ‡ Covariates: "A" (age), "E" (education), "G" (gender), "APOE4" (APOE e4 genotype), "MMSE" (baseline MMSE), "BSTR" (baseline stroke), "BCHD" (basline CHD), "CHD" (coronary heart disease), "FC" (field center), "OLLM" (other lipid lowering drug use), "RE" (race/ethnicity), "SH" (stroke history), "TRTGRP" (treatment group (not the exposure of interest))
 
Table 4:   Statin use duration - categorical
Notes These reports evaluate the relation of statin use duration, modeled as a categorical variable, to risk of incident AD.  
  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
Bettermann, 2012 GEMS Incidence study reporting hazard ratios (HRs) 1872
(-)		 -
(detail)
Never used:         
< 1 year:         
1-3 years:         
> 3 years:         
(detail)		 -
-
-
-		 1.00
0.31
0.71
0.22		 Ref.
0.12-0.76
0.41-1.22
0.07-0.70		 Ref.
0.01
0.22
0.01
*		 -
-
-
-		 1.00
0.69
0.56
0.26		 Ref.
0.41-1.14
0.34-0.91
0.11-0.60		 Ref.
0.15
0.02
0.002
*		 Caucasian, African-American (Black)
(detail)		 - (-)
(75 - )		 Screening: "Modified" 3MSE, CDR, ADAS_cog, Informant interview, Neuropsych Testing, TICS

AD Diagnosis: Brain Imaging, DSM IV, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, APOE4, CHD, FC, OLLM, RE, SH, TRTGRP‡ Bettermann, 2012
Haag, 2009 Rotterdam Study Incidence study reporting hazard ratios (HRs) 6992
(60%)		 9.2 y
Never used:         
< 2.9 years:         
> 2.9 years:         
(detail)		 -
-
-
Total: 466		 1.00
0.44
0.78		 Ref.
0.25-0.80
0.44-1.32		 Ref.
0.006
0.38
*		  
 
 		       Dutch
(detail)		 69 (9)
(55 - )
(detail)		 Screening: CAMDEX, GMS, MMSE, Neuropsych Testing, Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, E, G, BMI, CVD, CS, DM, SM, SBP, TC, CHLD‡ Haag, 2009
Rea, 2005 CHCS Incidence study reporting hazard ratios (HRs) 2798
(-)		 5.4 y
*
Never used: 91%
< 1 year:         
1-3 years:         
> 3 years:         
(detail)		 216
-
-
-
Total: 237		 1.00
1.52
1.05
1.04		 Ref.
0.78-2.98
0.49-2.24
0.42-2.56		 Ref.
0.22
0.9
0.93
*		 428
-
-
-
Total: 466		 1.00
0.98
1.41
0.74		 Ref.
0.55-1.74
0.89-2.25
0.35-1.57		 Ref.
0.95
0.15
0.43
*		 Caucasian
(detail)		 - (-)
(65 - )
(detail)		 Screening: Brain Imaging, MMSE, Other
(detail)		 A, E, G, MMSE, BSTR, BCHD‡ Rea, 2005
Zandi, 2005 Cache County Study Incidence study reporting hazard ratios (HRs) 3295
(-)		 3.3 y
*
Never used: 94%
< 3 years: 4%
> 3 years: 2%
(detail)		 98
2
1
Total: 101		 1.00
1.41†
0.62†		 Ref.
0.23-4.70
0.03-2.92		 Ref.
0.66
0.68
*		 174
5
2
Total: 181		 1.00
1.40†
0.71†		 Ref.
0.49-3.21
0.12-2.32		 Ref.
0.38
0.65
*		 Caucasian
(detail)		 - (-)
(65 - )
(detail)		 Screening: "Modified" 3MSE, DQ, Informant interview, IQ-CODE, Neuropsych Testing

AD Diagnosis: Brain Imaging, DSM IIIR, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, G, Age*APOE4, APOE4, DM, HXHTN‡ Zandi, 2005
* Derived value.
 † Five or fewer cases exist.
 ‡ Covariates: "A" (age), "E" (education), "G" (gender), "Age*APOE4" (Age*APOE4 interaction), "APOE4" (APOE e4 genotype), "MMSE" (baseline MMSE), "BSTR" (baseline stroke), "BCHD" (basline CHD), "BMI" (body mass index), "CVD" (cardiovascular disease), "CS" (cerebrovascular symptoms), "CHD" (coronary heart disease), "DM" (diabetes mellitus), "FC" (field center), "HXHTN" (history of hypertension), "OLLM" (other lipid lowering drug use), "RE" (race/ethnicity), "SM" (smoking status), "SH" (stroke history), "SBP" (systolic blood pressure), "TC" (total cholesterol), "TRTGRP" (treatment group (not the exposure of interest)), "CHLD" (use of a cholesterol-lowering drug other than a statin)
 
Table 5:   Statin dosage - categorical
Notes These reports evaluate the relation of statin dosage, modeled as a categorical variable, to risk of incident AD.  
  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
Haag, 2009 Rotterdam Study Incidence study reporting hazard ratios (HRs) 6992
(60%)		 9.2 y
Never used:         
< Median dosage:         
> Median dosage:         
(detail)		 546
-
-
Total: 466		 1.00
0.57
0.58		 Ref.
0.32-1.00
0.32-1.04		 Ref.
0.05
0.07
*		  
 
 		       Dutch
(detail)		 69 (9)
(55 - )
(detail)		 Screening: CAMDEX, GMS, MMSE, Neuropsych Testing, Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, E, G, BMI, CVD, CS, DM, SM, SBP, TC, CHLD‡ Haag, 2009
* Derived value.
 ‡ Covariates: "A" (age), "E" (education), "G" (gender), "BMI" (body mass index), "CVD" (cardiovascular disease), "CS" (cerebrovascular symptoms), "DM" (diabetes mellitus), "SM" (smoking status), "SBP" (systolic blood pressure), "TC" (total cholesterol), "CHLD" (use of a cholesterol-lowering drug other than a statin)