"Most surprising, says Dr. Isaacson, is that the MCI patients who followed at least 60% of their recommendations showed cognitive improvement. However, MCI pati

"Most surprising, says Dr. Isaacson, is that the MCI patients who followed at least 60% of their recommendations showed cognitive improvement. However, MCI patients who followed less than 60% of the recommendations experienced cognitive declines similar to the control groups, he notes." -- Another explanation: those able to follow these directions might have been in the best cognitive shape. What I'd wonder is if the less-than-60% cohort could have done worse than the control? Was this sorting masquerading as treatment? When they were in their early 80s, my mother and my Stepfather each started doing Sudoku puzzles to fight dementia. My mother lost interest and developed non-Alzheimer dementia six years later. My stepfather kept at it and did not develop dementia though he had cognitive issues. Where is the cause and where is the effect here, however? Edit: another thing that makes the study seem rather problematic is the "customized advice" part. Maybe some of the advice is better than others. How do you generalize? etc. For a more object study, you'd need to have people just meriting one kind of treatment in both control and treated groups. I understand how hard that could be - this stuff is just filled with quandaries. Customized treatment would logical at final patient level but very hard to objectively analyze.