This is based on the newly published Scandinavian study:
Homocysteine and holotranscobalamin and the risk of Alzheimer disease
Results: The odds ratios (ORs) (95% confidence interval [CI]) for AD were 1.16 (1.04–1.31) per increase of 1 µmol/L of tHcy at baseline and 0.980 (0.965–0.995) for each increase of 1 pmol/L baseline holoTC.
Notice how did the authors skillfully manage to conceal the impact (really huge!) - by not putting the risk ratio over the full range in the abstract. Instead they mitigated their grant-loosing risk by showing only an incremental change per 1umol/L or per 1pmol/L of the intependent varaibles. (Note: tHcy=homocysteine, holoTC=B12, AD=Alzheimer's Disease).
The impact seems to be huge indeed, based on some quick-and-dirty calculations of the typical tHcy and B12 variability among some populations. Taking two other studies (see  and  below) giving the average difference between vegetarians and non-vegetarians for tHcy and B12 ( 2.6 to 12.3 umol/L and -94 to -195 pmol/L respectively), applying the above incremental ratios of (1.16 and 0.980 respectively) one obtains the estimated ralative risk ratio for AD for vegetarians versus non-vegetarians of: 1.5 to 6.2 based on the homocysteine, and 6.7 to 51 based on the B12 data.
(Disclaimer: risk ratio estimates that are based on statistical data do not prove nor disprove causuation.)
My conclusion: the study indicates that the B12 defficiency as well as (independently) high homocysteine seem to be very strong predictors of Alzheimer's disease risk. The overal risk estimate is also much stronger than the washed-out figures presented in the paper's abstract. This should be of utmost importance for vegans and vegetarians!
1. Effect of Vegetarian Diet on Homocysteine Levels
2. Plasma Homocysteine Levels in Taiwanese Vegetarians Are Higher than Those of Omnivores
3. BBC Health News: Vitamin B12 link to Alzheimer's backed by study
4. BBC Health News: Vitamin B 'puts off Alzheimer's'