There were 5 observational studies (1,379 women), 7 RCTs and 1 non-RCT (294 women) which evaluated the effect of oestrogen on cognition in nondemented women. There were 8 case-control studies (2,381 women) and 2 prospective cohort studies (1,596 women) which evaluated the use of oestrogen therapy and risk of dementia. There were also 1 RCT, 1 non-RCT and 2 uncontrolled trials (58 women) of oestrogen therapy in women with AD.
Effect of oestrogen on cognition in nondemented women.
Of the 5 observational studies, 2 reported inconclusive results, 2 reported no association, and 1 found that oestrogen use improved cognition. Of the 8 controlled trials, 6 found that oestrogen improved cognitive function. All of the available evidence had significant methodological problems. There was no clear evidence of a beneficial effect from oestrogen in asymptomatic women.
Effect of oestrogen therapy on cognition in women with AD.
All 4 identified studies (2 controlled and 2 uncontrolled) found some beneficial effect from oestrogen on cognition on at least one measure of dementia severity, but generally not on all measures used. The studies were all extremely small with short duration of therapy, and included participants with a wide range of dementia severity.
Use of oestrogen therapy and risk of dementia.
The 10 identified studies (8 case-control and 2 prospective cohort studies) produced results ranging from a suggestion of a protective effect from oestrogen, to an increased risk of developing AD in oestrogen users. The meta-analysis of all 10 studies produced a summary OR of 0.71 (95% CI: 0.53, 0.96) for developing any dementia. The 8 case-control studies gave an OR of 0.79 (95% CI: 0.56, 1.12) for any dementia, and the 2 prospective cohort studies gave an OR of 0.48 (95% CI: 0.29, 0.81) for AD. The definition of dementia had very little impact on the results of the meta-analysis.