Get Newsletter
AlzRisk Risk Factor OverView
Back

Risk Factor:
  (postmenopausal hormone use)
Risk Factor Type: Medications
Current Understanding:
The tables on the Risk Factor Overview present results from a series of observational prospective studies and one randomized controlled trial of postmenopausal hormone therapy (including estrogen alone or in combination with progestin) in relation to AD risk. Overall, in contrast to earlier systematic reviews and meta-analyses, the evidence from these studies does not support benefits on AD risk. The relationship between postmenopausal hormone therapy and AD risk has been controversial; early observational studies, many of them retrospective, suggested a beneficial effect of hormone therapy, and randomized controlled trial results in postmenopausal women suggest that hormone therapy increases dementia risk, and also has adverse effects on other health outcomes. Confounding, as well as differences in the women’s ages, in the timing of use relative to menopause, and in hormone therapy formulation, may account for the disparate findings in earlier observational studies vs. subsequent clinical trials, but more recent observational studies suggest no benefit. Changes over time in the observational findings may be related to changes in hormone therapy use over time. In any case, estrogen remains one of the most effective treatments for relief from menopausal vasomotor symptoms, and, according to the North American Menopause Society, temporary use of hormone therapy may be appropriate to treat low-risk women with severe menopausal symptoms. For a discussion of the putative mechanisms by which hormone therapy may be related to AD risk and commentary on interpreting the findings below in a broader context, please see the Discussion. A longer review and discussion can be found in the closely related published review and meta-analysis, O'Brien J, Jackson JW, Grodstein F, Blacker D, Weuve J. Postmenopausal hormone therapy is not associated with risk of all-cause dementia and Alzheimer's Disease (Epidemiologic Reviews 2014;36:83-103).
Literature Extraction: Search strategy  * New *
Last Search Completed: 20 March 2014 - (Same date as last content update.)


Table 1:   Hormone replacement therapy (any v. minimal or no use)
Meta-Analysis
Notes These reports examine the relationship between postmenopausal use of hormone replacement therapy (HRT) and AD. Use of HRT can refer to use of estrogen alone or estrogen in combination with a progestogen (progestin or progesterone). Most commonly, HRT refers to use of estrogens combined with a progestogen and ERT refers to use of estrogens only. Not all papers provided information about the types of HRT used by their participants. Results here compare AD risk among women who ever used HRT with risk among those who never used HRT. Some studies in this table compare those with chronic HRT usage with those with "minimal" HRT use. (In other words, some women in the reference group used HRT for a brief interval that investigators considered insufficient to bear meaningfully on AD risk.)  
  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
Brenner, 1994 ADPR-GHC Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 227
(100%)		 -
(detail)
No HRT: 52%
HRT - all: 48%
(detail)		 55
52
Total: 107		 1.00
1.10		 Ref.
0.60-1.80		 Ref.
0.73
*		  
 		       Caucasian
 74 (7)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, HYS‡
(detail)		 Brenner, 1994
Kawas, 1997 BLSA Incidence study reporting hazard ratios (HRs) 472
(100%)		 -
(detail)
No HRT: 51%
HRT: 49%
(detail)		 25
9
Total: 34		 1.00
0.46		 Ref.
0.21-0.99		 Ref.
0.05
*		  
 		       Caucasian
 62 (-)
(28 - 94)		 AD Diagnosis: Brain Imaging, DSM IIIR, NINCDS ADRDA, Neurologic examination
(detail)		 A, E‡
(detail)		 Kawas, 1997
Lindsay, 2002 CSHA Cumulative incidence study reporting odds ratios (ORs) 2079
(100%)		 5.0 y
(detail)
No HRT: 94%
HRT: 6%
(detail)		 106
4
Total: 110		 1.00
1.37†		 Ref.
0.48-3.95		 Ref.
0.56
*		  
 		        (detail) 73 (-)
( - )
(detail)		 Screening: 3MSE

AD Diagnosis: DSM IV, NINCDS ADRDA
(detail)		 A, E‡
(detail)		 Lindsay, 2002
Roberts, 2006 Rochester Epidemiology Project Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 486
(100%)		 -
(detail)
< 6 mo or never: 89%
≥ 6 mo: 11%
(detail)		 216
28
Total: 244		 1.00
1.10		 Ref.
0.63-1.93		 Ref.
0.73		  
 		       Caucasian
(detail)		 84 (-)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IV
(detail)		 A, TM‡
(detail)		 Roberts, 2006
Shao, 2012 Cache County Study Incidence study reporting hazard ratios (HRs) 1768
(100%)		 7.0 y
(detail)
No HRT: 38%
HRT: 62%
(detail)		 89
87
Total: 176		 1.00
0.80		 Ref.
0.58-1.09		 Ref.
0.17
*		 -
-
Total: 248		 1.00
0.84		 Ref.
0.65-1.09		 Ref.
0.19
*		  (detail) 75 (6)
( - )
(detail)		 Screening: DQ, IQ-CODE, 3MSE

AD Diagnosis: Modified NINCDS ADRDA, Brain Imaging
(detail)		 A, E, APOE4, PS‡
(detail)		 Shao, 2012
Shumaker, 2004 WHIMS RCT, Incidence study reporting hazard ratios (HRs) 7479
(100%)		 4.5 y
(detail)
Placebo: 51%
E or E+P: 49%
(detail)		 21
33
Total: 54		 1.00
1.62
*		 Ref.
0.91-2.96
*		 Ref.
0.11
*		 40
68
Total: 108		 1.00
1.76		 Ref.
1.19-2.60		 Ref.
0.005
*		  (detail) 71 (-)
(65 - )
(detail)		 Screening: "Modified" 3MSE, CERAD, Informant interview, Neuropsych Testing

AD Diagnosis: DSM IV
(detail)		  
(detail)		 Shumaker, 2004
Tang, 1996 WHICAP Incidence study reporting hazard ratios (HRs) 1124
(100%)		 -
(detail)
No HRT: 86%
HRT: 14%
(detail)		 158
9
Total: 167		 1.00
0.50		 Ref.
0.25-0.90		 Ref.
0.02		  
 		       Caucasian, Hispanic, African-American (Black)
 74 (7)
( - )
(detail)		 AD Diagnosis: Brain Imaging, DSM IIIR, Medical History, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, APOE4, RE‡
(detail)		 Tang, 1996
Waring, 1999 Rochester Epidemiology Project Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 444
(100%)		 -
(detail)
< 6 mo or never: 93%
≥ 6 mo: 7%
(detail)		 211
11
Total: 222		 1.00
0.48		 Ref.
0.22-1.01		 Ref.
0.05		  
 		       Caucasian
(detail)		 - (-)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A‡
(detail)		 Waring, 1999
Zandi, 2002 Cache County Study Incidence study reporting hazard ratios (HRs) 1866
(100%)		 3.0 y
(detail)
No HRT: 43%
HRT: 57%
(detail)		 58
26
Total: 84		 1.00
0.59		 Ref.
0.36-0.96		 Ref.
0.03
*		  
 		        (detail) 74 (6)
( - )
(detail)		 Screening: DQ, Informant interview, 3MSE

AD Diagnosis: Medical History, NINCDS ADRDA, Neurologic examination, Neuropsychological examination
(detail)		 A, E, APOE4‡
(detail)		 Zandi, 2002
* Derived value.
 † Five or fewer cases exist.
 ‡ Covariates: "A" (age), "E" (education), "AAMr" (age at menarche), "AAMp" (age at menopause), "ALC" (alcohol intake), "APOE4" (APOE e4 genotype), "BMI" (body mass index), "BW50" (body weight at age 50), "HYS" (history of hysterectomy), "CH" (number of children), "PS" (propensity score), "RE" (race/ethnicity), "SM" (smoking status), "TM" (type of menopause)
 
Table 2:   Hormone replacement therapy, duration of use (categorical)
Notes These reports examine the relationship between the duration of use of hormone replacement therapy (HRT; either in years or number of prescriptions filled) and AD. The risk among women in each category of duration of use is compared with the risk among never users. As in Table 1, use of HRT can refer to use of estrogen alone or estrogen in combination with a progestogen (progestin or progesterone), although some studies are not explicit about which types of HRT are represented.  
  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
Barnes, 2003 ROS Incidence study reporting hazard ratios (HRs) 577
(100%)		 4.8 y
(detail)
No HRT: 64%
< 10 years:         
≥ 10 years:         
(detail)		 -
-
-
Total: 92		 1.00
0.57
1.38		 Ref.
0.21-1.60
0.81-2.36		 Ref.
0.28
0.24
*		  
 
 		       Caucasian, Other, African-American (Black)
(detail)		 76 (7)
(65 - )
(detail)		 Screening: Neuropsych Testing

AD Diagnosis: NINCDS ADRDA
(detail)		 A, E‡
(detail)		 Barnes, 2003
Brenner, 1994 ADPR-GHC Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 227
(100%)		 5.0 y
*
(detail)
All estrogens
No HRT: 52%
1 - 2 prescriptions: 17%
3 - 10 prescriptions: 15%
11 - 50 prescriptions: 16%
(detail)
55
22
15
50
Total: 142
1.00
1.50
0.90
0.80
Ref.
0.70-3.00
0.40-2.10
0.30-1.80
Ref.
0.27
0.8
0.63
*
 
 
 
 
 
 
 		 Caucasian
 74 (7)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, HYS‡
(detail)		 Brenner, 1994
Brenner, 1994 ADPR-GHC Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 175
(100%)		 5.0 y
*
(detail)
Oral estrogens
No HRT: 67%
1 - 4 prescriptions: 10%
5 - 16 prescriptions: 13%
17 - 50 prescriptions: 10%
(detail)
55
7
12
6
Total: 80
1.00
0.70
1.40
0.40
Ref.
0.20-2.00
0.50-3.80
0.10-1.50
Ref.
0.54
0.52
0.18
*
 
 
 
 
 
 
 		 Caucasian
 74 (7)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, HYS‡
(detail)		 Brenner, 1994
Brenner, 1994 ADPR-GHC Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 194
(100%)		 5.0 y
*
(detail)
Vaginal estrogens
No HRT: 60%
1 prescription: 13%
2-4 prescriptions: 15%
5 - 27 presciptions: 12%
(detail)
55
15
15
8
Total: 93
1.00
1.70
1.40
0.70
Ref.
0.70-4.00
0.60-3.10
0.30-1.90
Ref.
0.23
0.42
0.45
*
 
 
 
 
 
 
 		 Caucasian
 74 (7)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, HYS‡
(detail)		 Brenner, 1994
Kawas, 1997 BLSA Incidence study reporting hazard ratios (HRs) 472
(100%)		 -
(detail)
No HRT:         
> 0 - 5 years:         
> 5 - 10 years:         
> 10 years:         
(detail)		 -
-
-
-
Total: 34		 1.00
0.44
0.34
0.50		 Ref.
0.13-1.51
0.05-2.52
0.17-1.47		 Ref.
0.19
0.29
0.21		  
 
 
 		       Caucasian
 62 (-)
(28 - 94)		 AD Diagnosis: Brain Imaging, DSM IIIR, NINCDS ADRDA, Neurologic examination
(detail)		 A, E‡
(detail)		 Kawas, 1997
Roberts, 2006 Rochester Epidemiology Project Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 486
(100%)		 -
(detail)
< 0.5 years or never: 89%
0.5 - 3 years: 5%
> 3 years: 6%
(detail)		 216
14
14
Total: 244		 1.00
1.22
1.01		 Ref.
0.55-2.69
0.47-2.20		 Ref.
0.63
0.97		  
 
 		       Caucasian
(detail)		 84 (-)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IV
(detail)		 A, TM‡
(detail)		 Roberts, 2006
Tang, 1996 WHICAP Incidence study reporting hazard ratios (HRs) 1124
(100%)		 -
(detail)
No HRT: 86%
Unknown: 3%
≤ 1 year: 6%
> 1 year: 5%
(detail)		 158
3
5
1
Total: 167		 1.00
1.30†
0.47†
0.13†		 Ref.
0.40-4.20
0.20-1.10
0.02-0.92		 Ref.
0.66
0.06
0.01
*		  
 
 
 		       Caucasian, Hispanic, African-American (Black)
 74 (7)
( - )		 AD Diagnosis: Brain Imaging, DSM IIIR, Medical History, NINCDS ADRDA, Neuropsychological examination
(detail)		 A, E, APOE4, RE‡
(detail)		 Tang, 1996
Waring, 1999 Rochester Epidemiology Project Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 444
(100%)		 -
(detail)
No HRT: 82%
< 6 months: 11%
≥ 6 months: 7%
(detail)		 189
22
11
Total: 222		 1.00
0.85
0.42		 Ref.
0.44-1.62
0.18-0.96		 Ref.
0.62
0.04		  
 
 		       Caucasian
(detail)		 - (-)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A‡
(detail)		 Waring, 1999
Zandi, 2002 Cache County Study Incidence study reporting hazard ratios (HRs) 1856
(100%)		 3.0 y
(detail)
No HRT: 43%
< 3y: 17%
3 - 10y: 17%
> 10y: 23%
(detail)		 58
10
8
7
Total: 83		 1.00
0.82
0.60
0.41		 Ref.
0.38-1.57
0.26-1.22
0.17-0.86		 Ref.
0.58
0.2
0.03
*		  
 
 
 		        (detail) 74 (6)
( - )
(detail)		 Screening: DQ, Informant interview, 3MSE

AD Diagnosis: Medical History, NINCDS ADRDA, Neurologic examination, Neuropsychological examination
(detail)		 A, E, APOE4‡
(detail)		 Zandi, 2002
* Derived value.
 † Five or fewer cases exist.
 ‡ Covariates: "A" (age), "E" (education), "APOE4" (APOE e4 genotype), "BMI" (body mass index), "HYS" (history of hysterectomy), "RE" (race/ethnicity), "SM" (smoking status), "TM" (type of menopause)
 
Table 3:   Hormone replacement therapy, cumulative dose (categorical)
Notes These reports examine the relationship between cumulative dose of hormone replacement therapy (HRT) and AD. The risk among women in each category of cumulative dose is compared with the risk among never users. As in Table 1, use of HRT can refer to use of estrogen alone or estrogen in combination with a progestogen (progestin or progesterone), although some studies are not explicit about which types of HRT are represented.  
  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
Roberts, 2006 Rochester Epidemiology Project Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 475
(100%)		 -
(detail)
No HRT: 91%
≤ 756 mg: 5%
> 756 mg: 4%
(detail)		 216
13
9
Total: 238		 1.00
1.52
0.77		 Ref.
0.64-3.61
0.32-1.84		 Ref.
0.34
0.56		  
 
 		       Caucasian
(detail)		 84 (-)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IV
(detail)		 TM‡
(detail)		 Roberts, 2006
Waring, 1999 Rochester Epidemiology Project Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 444
(100%)		 -
(detail)
0 mg: 89%
≤ 20 mg: 3%
21 - 500 mg: 4%
> 500 mg: 4%
(detail)		 189
7
7
7
Total: 210		 1.00
1.34
0.77
0.60		 Ref.
0.42-4.28
0.26-2.25
0.28-2.06		 Ref.
0.63
0.63
0.6		  
 
 
 		       Caucasian
(detail)		 - (-)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A‡
(detail)		 Waring, 1999
‡ Covariates: "A" (age), "TM" (type of menopause)
 
Table 4:   Hormone replacement therapy, current/past/never (categorical)
Notes These reports examine the relationship between the timing of HRT use and AD. As in Table 1, use of HRT can refer to use of estrogen alone or estrogen in combination with a progestogen (progestin or progesterone), although some studies are not explicit about which types of HRT are represented.  
  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
Brenner, 1994 ADPR-GHC Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 227
(100%)		 5.0 y
*
(detail)
All estrogens
No HRT: 52%
Former use: 24%
Current Use: 23%
(detail)
55
33
19
Total: 107
1.00
1.70
0.60
Ref.
0.90-3.20
0.30-1.20
Ref.
0.1
0.15
*
 
 
 
 
 
 		 Caucasian
 74 (7)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, HYS‡
(detail)		 Brenner, 1994
Brenner, 1994 ADPR-GHC Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 175
(100%)		 5.0 y
*
(detail)
Oral estrogens
No HRT: 67%
Former use: 14%
Current Use: 19%
(detail)
55
13
12
Total: 80
1.00
1.40
0.40
Ref.
0.60-3.50
0.20-1.10
Ref.
0.45
0.07
*
 
 
 
 
 
 		 Caucasian
 74 (7)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, HYS‡
(detail)		 Brenner, 1994
Brenner, 1994 ADPR-GHC Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 194
(100%)		 5.0 y
*
(detail)
Vaginal estrogens
No HRT: 60%
Former use: 29%
Current Use: 11%
(detail)
55
31
7
Total: 93
1.00
1.60
0.60
Ref.
0.80-3.00
0.20-1.70
Ref.
0.16
0.35
*
 
 
 
 
 
 		 Caucasian
 74 (7)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, HYS‡
(detail)		 Brenner, 1994
Ryan, 2009 3C Incidence study reporting hazard ratios (HRs) 3130
(100%)		 4.0 y
(detail)
No HRT: 69%
Former use: 16%
Current Use: 15%
(detail)		 -
-
-
Total: 53		 1.00
0.39
1.36		 Ref.
0.39-2.23
0.44-4.20		 Ref.
0.86
0.59		 -
-
-
Total: 79		 1.00
0.74
0.83		 Ref.
0.35-1.55
0.32-2.17		 Ref.
0.42
0.7		 Caucasian
(detail)		 74 (5)
(65 - )
(detail)		 Screening: MMSE, Other

AD Diagnosis: DSM IV
(detail)		 A, E, AAMp, ANTCH, BMI, CI, DEP, HICAF, MS, O, PP‡
(detail)		 Ryan, 2009
Zandi, 2002 Cache County Study Incidence study reporting hazard ratios (HRs) 1866
(100%)		 3.0 y
(detail)
No HRT: 43%
Former use: 26%
Current Use: 31%
(detail)		 58
9
17
Total: 84		 1.00
0.33
1.08		 Ref.
0.15-0.65
0.59-1.91		 Ref.
0.003
0.8
*		  
 
 		        (detail) 74 (6)
( - )
(detail)		 Screening: DQ, Informant interview, 3MSE

AD Diagnosis: Medical History, NINCDS ADRDA, Neurologic examination, Neuropsychological examination
(detail)		 A, E, APOE4‡
(detail)		 Zandi, 2002
* Derived value.
 ‡ Covariates: "A" (age), "E" (education), "AAMp" (age at menopause), "ANTCH" (anticholingergic drugs), "APOE4" (APOE e4 genotype), "BMI" (body mass index), "CI" (comorbidity index), "DEP" (depression), "HICAF" (high caffeine intake), "HYS" (history of hysterectomy), "MS" (marital status), "O" (other), "PP" (physical performance), "SM" (smoking status)
 
Table 5:   Hormone replacement therapy, age at initiation (categorical)
Notes These reports examine the relationship between the age of initiation of hormone replacement therapy (HRT) and AD. The risk among women in each category of age at initiation is compared with the risk among never users. As in Table 1, use of HRT can refer to use of estrogen alone or estrogen in combination with a progestogen (progestin or progesterone), although some studies are not explicit about which types of HRT are represented.  
  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
Roberts, 2006 Rochester Epidemiology Project Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 486
(100%)		 -
(detail)
No HRT: 89%
age ≤ 49.5 years: 6%
age > 49.5 years: 6%
(detail)		 216
17
11
Total: 244		 1.00
1.81
0.67		 Ref.
0.81-4.05
0.30-1.51		 Ref.
0.15
0.34		  
 
 		       Caucasian
(detail)		 84 (-)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IV
(detail)		 A, TM‡
(detail)		 Roberts, 2006
‡ Covariates: "A" (age), "TM" (type of menopause)
 
Table 6:   Hormone replacement therapy, time between menopause and HRT initiation (categorical)
Notes This report examines the relationship of the time between menopause onset and initiation of HRT to AD risk (i.e., lag time from menopause to the onset of HRT). The risk among women in each category of lag time is compared with the risk among never users (which includes whose who took therapy for < 6 months). As in Table 1, use of HRT can refer to use of estrogen alone or estrogen in combination with a progestogen (progestin or progesterone).  
  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
Roberts, 2006 Rochester Epidemiology Project Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 485
(100%)		 -
(detail)
No HRT: 89%
≤ 2 y: 7%
> 2 y: 5%
(detail)		 216
17
11
Total: 244		 1.00
1.24
1.02		 Ref.
0.61-2.53
0.41-2.54		 Ref.
0.55
0.97		  
 
 		       Caucasian
(detail)		 84 (-)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IV
(detail)		 A, TM‡
(detail)		 Roberts, 2006
Shao, 2012 Cache County Study Incidence study reporting hazard ratios (HRs) 1768
(100%)		 7.0 y
(detail)
No HRT: 38%
≤ 5 y: 41%
> 5 y: 21%
(detail)		 89
52
35
Total: 176		 1.00
0.70
1.03		 Ref.
0.49-0.99
0.68-1.55		 Ref.
0.05
0.89
*		 -
-
-
Total: 248		 1.00
0.74
1.06		 Ref.
0.55-1.00
0.76-1.49		 Ref.
0.05
0.73
*		  (detail) 75 (6)
( - )
(detail)		 Screening: DQ, IQ-CODE, 3MSE

AD Diagnosis: Modified NINCDS ADRDA, Brain Imaging
(detail)		 A, E, APOE4, PS‡
(detail)		 Shao, 2012
Shao, 2012 Cache County Study Incidence study reporting hazard ratios (HRs) 1705
(100%)		 7.0 y
(detail)
By duration of use
No HRT: 39%
≤ 5 y, duration < 3 y: 10%
≤ 5 y, duration 3-10 y: 5%
≤ 5 y, duration ≥ 10 y: 27%
> 5 y, duration < 3 y: 7%
> 5 y, duration 3-10 y: 6%
> 5 y, duration ≥ 10 y: 7%
(detail)
89
13
6
29
11
8
11
Total: 167
1.00
0.71
0.71
0.63
1.04
0.92
1.00
Ref.
0.39-1.28
0.31-1.65
0.41-0.98
0.54-1.97
0.44-1.92
0.53-1.91
Ref.
0.26
0.42
0.04
0.91
0.82
1.0
*
-
-
-
-
-
-
-
Total: 248
1.00
0.79
0.61
0.70
1.12
0.64
1.21
Ref.
0.49-1.28
0.28-1.32
0.49-1.00
0.68-1.87
0.31-1.32
0.74-2.00
Ref.
0.34
0.21
0.05
0.66
0.23
0.45
*		  (detail) 75 (6)
( - )
(detail)		 Screening: DQ, IQ-CODE, 3MSE

AD Diagnosis: Modified NINCDS ADRDA, Brain Imaging
(detail)		 A, E, APOE4, PS‡
(detail)		 Shao, 2012
* Derived value.
 ‡ Covariates: "A" (age), "E" (education), "APOE4" (APOE e4 genotype), "PS" (propensity score), "TM" (type of menopause)
 
Table 7:   Hormone replacement therapy, formulation (categorical)
Notes These reports examine the relationship between formulation of HRT and AD. "E" refers to estrogen, and "P" refers to the presence of a progestin.  
  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
Petitti, 2008 KPSC Incidence study reporting hazard ratios (HRs) 2906
(100%)		 5.0 y
(detail)
No HRT: 48%
E w/ P: 15%
E w/o P: 37%
(detail)		 -
-
-
Total: 191		 1.00
1.41
1.22		 Ref.
0.94-2.12
0.89-1.68		 Ref.
0.1
0.22
*		 50
50
112
Total: 212		 1.00
1.34
1.23		 Ref.
0.95-1.89
0.94-1.59		 Ref.
0.1
0.12
*		 Caucasian, Other, Hispanic, African-American (Black)
 79 (3)
(75 - )
(detail)		 Screening: Informant interview, Other, TICS-m

AD Diagnosis: Other
(detail)		 A, E, DM, HTN, MI, PD, SH‡
(detail)		 Petitti, 2008
Shao, 2012 Cache County Study Incidence study reporting hazard ratios (HRs) 1732
(100%)		 7.0 y
(detail)
No HRT: 38%
E w/ P: 24%
E w/o P: 38%
(detail)		 89
32
49
Total: 170		 1.00
0.93
0.70		 Ref.
0.60-1.43
0.49-1.01		 Ref.
0.74
0.05
*		 -
-
-
Total: 248		 1.00
0.93
0.79		 Ref.
0.64-1.35
0.59-1.07		 Ref.
0.7
0.12
*		  (detail) 75 (6)
( - )
(detail)		 Screening: DQ, IQ-CODE, 3MSE

AD Diagnosis: Modified NINCDS ADRDA, Brain Imaging
(detail)		 A, E, APOE4, PS‡
(detail)		 Shao, 2012
Shao, 2012 Cache County Study Incidence study reporting hazard ratios (HRs) 1732
(100%)		 7.0 y
(detail)
By time since menopause onset
No HRT: 38%
E w/ P, initiated ≤ 5 y after menopause: 13%
E w/ P, initiated > 5 y after menopause: 10%
E w/o P, initiated ≤ 5 y after menopause: 27%
E w/o P, initiated > 5 y after menopause: 11%
(detail)
89
13
19
34
15
Total: 170
1.00
0.65
1.32
0.65
0.86
Ref.
0.36-1.18
0.78-2.24
0.43-0.98
0.49-1.51
Ref.
0.15
0.3
0.04
0.6
*
-
-
-
-
-
Total: 248
1.00
0.65
1.32
0.65
0.86
Ref.
0.36-1.18
0.78-2.24
0.43-0.98
0.49-1.51
Ref.
0.15
0.3
0.04
0.6
*		  (detail) 75 (6)
( - )
(detail)		 Screening: DQ, IQ-CODE, 3MSE

AD Diagnosis: Modified NINCDS ADRDA, Brain Imaging
(detail)		 A, E, APOE4, PS‡
(detail)		 Shao, 2012
Shumaker, 2004 WHIMS RCT, Incidence study reporting hazard ratios (HRs) 4531
(100%)		 4.0 y
*
(detail)
Placebo: 51%
E w/ P: 49%
(detail)		 12
20
Total: 32		 1.00
1.74
*		 Ref.
0.71-3.91
*		 Ref.
0.2
*		 21
40
Total: 61		 1.00
2.05		 Ref.
1.21-3.48		 Ref.
0.01
*		  (detail) 71 (-)
(65 - )
(detail)		 Screening: "Modified" 3MSE, CERAD, Informant interview, Neuropsych Testing

AD Diagnosis: DSM IV
(detail)		  
(detail)		 Shumaker, 2004
Shumaker, 2004 WHIMS RCT, Incidence study reporting hazard ratios (HRs) 2947
(100%)		 5.2 y
*
(detail)
Placebo: 50%
E w/o P: 50%
(detail)		 9
13
Total: 22		 1.00
1.47
*		 Ref.
0.58-3.91
*		 Ref.
0.44
*		 19
28
Total: 47		 1.00
1.49		 Ref.
0.83-2.66		 Ref.
0.18
*		  (detail) 71 (-)
(65 - )
(detail)		 Screening: "Modified" 3MSE, CERAD, Informant interview, Neuropsych Testing

AD Diagnosis: DSM IV
(detail)		  
(detail)		 Shumaker, 2004
* Derived value.
 † Five or fewer cases exist.
 ‡ Covariates: "A" (age), "E" (education), "APOE4" (APOE e4 genotype), "BMI" (body mass index), "DM" (diabetes mellitus), "HTN" (hypertension), "MI" (mycardial infarction history), "PD" (Parkinson's disease), "PS" (propensity score), "SM" (smoking status), "SH" (stroke history)
 
Table 8:   Hormone replacement therapy, route of administration (categorical)
Notes These reports examine the relationship between the route of administration of HRT and AD. As in Table 1, use of HRT can refer to use of estrogen alone or estrogen in combination with a progestogen (progestin or progesterone), although some studies are not explicit about which types of HRT are represented. "E" refers to estrogen.  
  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
Brenner, 1994 ADPR-GHC Nested case control study with cumulative incidence sampling reporting odds ratios (ORs) 252
(100%)		 5.0 y
*
(detail)
No HRT: 46%
HRT use - oral E: 23%
HRT use - vaginal E: 31%
(detail)		 55
25
38
Total: 118		 1.00
0.70
1.30		 Ref.
0.40-1.50
0.70-2.30		 Ref.
0.29
0.39
*		  
 
 		       Caucasian
 74 (7)
( - )
(detail)		 Screening: Other

AD Diagnosis: DSM IIIR, NINCDS ADRDA
(detail)		 A, HYS‡
(detail)		 Brenner, 1994
* Derived value.
 ‡ Covariates: "A" (age), "HYS" (history of hysterectomy)