Serum homocysteine and dementia: meta-analysis of eight cohort studies including 8669 participants

Alzheimers Dement. 2011 Jul;7(4):412-7. doi: 10.1016/j.jalz.2010.08.234.


Background: Prospective cohort studies have not been consistent in showing an association between serum homocysteine and dementia.

Objective: To conduct a meta-analysis of cohort studies that examined the relationship between serum homocysteine and dementia, and to estimate the change in risk of dementia for a unit change in serum homocysteine.

Methods: The data from eight cohort studies (involving 8,669 participants; range of mean ages, 47-81 years; median duration of study, 5 years) of serum homocysteine on the incidence of dementia were combined and the odds ratio of dementia per 5 μmol/L increase in serum homocysteine was determined.

Results: There was a statistically significant association between serum homocysteine and the incidence of dementia: the odds ratio for a 5 μmol/L increase in serum homocysteine was 1.35 (95% confidence interval, 1.02-1.79) or 1.50 (1.13-2.00) adjusted for regression dilution bias. The odds ratio for a 3 μmol/L decrease in serum homocysteine (the average reduction expected using folic acid and B12) was 0.78 (0.66-0.93).

Conclusion: The meta-analysis of epidemiological cohort studies shows a positive association between serum homocysteine and dementia. Although the results do not provide evidence of cause and effect, they do provide an estimate of the expected effect if the relationship were causal; an approximate 20% reduction in risk of dementia from treatment with folic acid and B12.

Publication types

  • Meta-Analysis

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Cohort Studies
  • Confidence Intervals
  • Dementia / blood*
  • Dementia / epidemiology*
  • Female
  • Homocysteine / blood*
  • Humans
  • Male
  • Middle Aged
  • Odds Ratio


  • Homocysteine