Two recent headlines caught my attention. “Eating berries linked to delay in cognitive decline” (Annals of Neurology, 2012, April 25) and “Coffee may ward off progression to dementia” (Journal of Alzheimer’s Disease, 2012, 30, 559-572). I drink blueberry flavored coffee each morning. Does this give me double the protection?

The first study demonstrated that higher consumption of either blueberries or strawberries “slowed memory decline by up to 2.5 years. The study, Nurses’ Health study, had a very large sample size, 16,010, with an average age of 74. Cognitive function was assessed every two years between 1995 and 2001. “Protective” associations were found for women who ate 1 or more servings of blueberries per week or ate 2 or more servings of strawberries per week.

On the surface the findings look encouraging. But the rub comes with some of the details. Most importantly, the study controls for age and education – a good thing. However, the study did not take into account life style choices. When the results of the study were conducted balancing those who exercised more and had higher income, the findings were attenuated. The bottom line is that it may be general life style choices and more money to buy healthy foods that accounts for the effects of not eating berries. So go ahead and eat strawberries and blueberries. But don’t forget to also exercise and engage in a generally healthy lifestyle.

The second study focused on coffee consumption (measured plasma levels of caffeine) in participants who were diagnosed with Mild Cognitive Impairment (MCI). The findings suggest that drinking 3-5 cups (500 mg of caffeine) was associated with failure to progress from MCI to dementia ‘over the course ‘of 2-4 years. More specifically, those diagnosed with MCI demonstrated significantly lower plasma levels of caffeine compared to those with normal memory. The study also reports lower plasma levels of caffeine in those diagnosed with dementia than those with normal memory. But this finding was not statistically significant which means it should not have been reported. As in the case of eating berries, the effects may have been a result of uncontrolled for lifestyle or health reasons not caffeine.

Statistical side bar. I have often seen studies report findings as being “nearly statistically significant” or as “trends toward statistical significance” There is no such thing. Statistical analysis is a tool for decision making. Either a finding is significant or not based on a priori decision rules. It’s sort of like one cannot be nearly pregnant. We have enough confusion from conflicting studies. We don’t need to clutter our understanding of these important issues with “trends.”

The conclusion I draw from these two studies is that there are still no magic bullets. We need randomized clinical trials to sort out factors that slow or prevent progression of memory loss or neurological disease. I will continue to eat berries as I have been and drink my two to three cups of coffee per day. I am also hedging my bets by being on the treadmill and staying engaged with the world.