This was the most upbeat of the Annals Alzheimer's Meta-Analyses Meta Analyses. And it wasn't incredibly upbeat. It basically shows that cognitive training DOES IMPROVE PERFORMANCE, but only in the areas trained. Little to no data to show that overall cognitive impairment is improved.
Again, the meta-analysis was great. Excellent study techniques. But you can't "make a silk purse of a sow's ear". In other words, the Alzheimer's prevention research, original studies, just have not been done. There is insufficient focus on insulin resistance, long-term follow-up, and multi-system approach.