Profile: A triple threat, Kristine Yaffe is an associate professor in the UCSF departments of psychiatry, neurology, and epidemiology and biostatistics. She received her medical degree from the University of Pennsylvania and completed her residency and fellowship training at UCSF. Yaffe, who is chief of geriatric psychiatry at the UCSF-affiliated Veterans Affairs Medical Center, studies the causes of cognitive decline in older adults and possible ways to prevent it.

Q. Isn't cognitive decline just an inevitable part of aging, and our efforts to convince ourselves otherwise a delusional bow to baby boomer anxiety?

It's true that baby boomers are worried about maintaining their quality of life as they age. And, as part of that concern, they are fearful of dementia. But there is a difference between normal cognitive aging and real, measurable impairment. Think of it this way: Healthy 80-year-olds will not be able to match the speed of 20-year-olds in completing certain mental tasks such as putting a list of items in their correct order, particularly if there is a time limit. And they may be occasionally forgetful about where they placed things or have mild trouble recalling someone's name. Yet these same healthy 80-year-olds can still learn new things and remember them. Yes, there is a decline in mental agility, and yes, it may be inevitable, but that is a far cry from functional impairment seen with clinical cognitive impairment.

Q. So considering the increased incidence of Alzheimer's disease, other dementias and related diseases that do in fact impair us functionally, we really do have something to worry about. True?

I'm not sure we're all worrying about the same things. As we successfully manage chronic diseases that used to kill people at a younger age, the number of people over age 85 continues to grow. We see more old people, so we see more Alzheimer's and the memory loss, language impairment and visual spatial distortion that go with that disease and other dementias. Does that mean that we want to go back to a different era and lead shorter lives? I don't think so. But for others, the problem is that they don't worry enough. Too many people still accept and expect that when we get older we get "senile." There is no distinction made between the normal slowing down and dementia and no understanding that we can do things to help.

Q. When you say "we" do you mean patients helping themselves or doctors prescribing drugs?

Both. Depending upon what we learn in our clinical evaluation, which includes neurophysiological exams and family interviews, we can prescribe drugs that sometimes treat the symptoms or lessen their impact. And I know it's what you always hear, but healthy lifestyle, proper diet and exercise seem to make a difference in how the brain functions.

Q. Has exercise really been shown to strengthen the connections among neurons?

Studies in mice have shown that there are more synaptic connections among mice especially in brain regions involved in learning and memory - that exercised than those that were the rodent version of human couch potatoes.

Q. So does that mean that lack of exercise is a risk factor for memory loss?

I would put it another way: That exercise seems to offer some protection. In one of our studies - an 8-year observational study of women 65 and older - we found that women who were most physically active at the beginning of the study were the least likely to experience a decline in cognitive or mental functions. And the activity didn't have to be strenuous. For every mile walked per day, women had a 13 percent lower chance of developing cognitive decline. Now, whether or not exercise has any effect at all on the progression to serious dementia, or if there was something else that made the exercise beneficial, we don't know.

Q. You also have reported that estrogen acts as a memory-loss preventive in older women. Are you rethinking that now in light of the evidence that hormone replacement therapy is harmful?

I think everyone is rethinking what kind of bias might be creeping into observational studies in general and, in particular, those involving estrogen therapy. That doesn't mean our finding was wrong, just that it has to be refined. Perhaps the benefit is selective and genetically based. Perhaps it reflects something about those who choose to take estrogen. Or maybe it's the dose or the way it was administered. That's the problem with studying humans. It's not always easy to get precise biological measurements. You sometimes have to infer. Randomized clinical trials, where some people are being treated and others are not - and no one knows which - are the most insensitive to bias, and we should be doing more of them.

Q. Back to observational studies for a moment: Don't those inferences get

a little tricky when you observe, for example, that those with more education are less likely to get Alzheimer's?

Yes they are tricky, since we need to consider if there is something about those who are more educated that might be causing that observation. Is it a cognitive reserve? Do they do more crossword puzzles? Are they more fit? Is higher education a by-product of some greater neuronal capacity to begin with? We don't know. But we are very much aware of what we call healthy user bias and it's very important to sort that out in observational studies.

Q. Is that also true when you're looking at the relationship between inflammation and dementia?

Definitely. We recently reported that the higher the concentration of markers that indicate inflammation in a person's

blood, the more cognitive decline a person showed. But is the inflammation a by-product of dementia or is it causing the problem? We hope to get the answer from a large prevention trial that is testing different anti-inflammatory drugs. My gut instinct is that inflammation is not causing dementia, but making it worse.

Q. How about obesity?

As you know, obese people have a higher risk for developing diabetes. There is some evidence that the same enzyme that degrades insulin also degrades amyloid beta protein in the brain. The concentration of amyloid beta plaques in the brain is a sign of Alzheimer's. So there could be a connection.

Q. Speaking of connections and inferences, what explains this penchant for researchers to report that something helps with memory loss, then in the next sentence, advise people not to take advantage of that information? Testosterone supplements are an example. You have done studies that show that raising levels of testosterone in older men improves cognitive function.

Yes, we have done these observational studies with testosterone, but again they were not clinical trials. They were only a first step. It is very difficult to convey a sense of risk to the public or to advise on safe dosages when there is no data yet. Just because something seems to work in one way doesn't mean that there are no side effects. In this case, higher levels of testosterone are also associated with everything from baldness to increased risk for prostate cancer. But even if these are reported in the media, the public may choose to read selectively. Still, I agree we need to add stronger caveats to our studies so that people understand the risks. People also need to be skeptical when they see headlines that include the word "cause" without the word "may."

Q. Aging has an impact on a lot of different body systems, yet the entire biomedical research focus in this country is on treating and curing individual diseases, not preventing them in the first place. Don't we have our priorities confused?

America has a quick-fix culture; we would rather take a pill than do a push-up. Our fixation on single diseases reflects that bias. Having said that, we also have to acknowledge that it is tough to find good biological markers that prove that a preventive measure works, and how it works. Prevention trials also are very expensive and take a long time. It's much easier to focus on one outcome as the FDA and NIH tend to prefer and organize around this goal. But, as you stated, aging involves an entire organism and many different systems. I think we could construct interdisciplinary clinical trials that allowed us to look at things simultaneously - and that also would be cost-effective.

Q. In the meantime, what should we do to keep our brains young?

Don't smoke. Don't become obese. And exercise both your body and your mind. It's going to be hard to improve on this strategy for now.