The man complained of memory problems but seemed perfectly normal. No specialist he visited detected any decline.
"He insisted that things were changing, but he aced all of our tests," said Rebecca Amariglio, a neuropsychologist at Brigham and Women's Hospital in Boston.
About seven years later, though, he began showing symptoms of dementia. Amariglio now believes he had recognized cognitive changes so subtle that "he was the only one who could identify it."
Patients like this have long been "called the worried well," said Creighton Phelps, acting chief of the dementias of aging branch of the National Institute on Aging. "People would complain, and we didn't really think it was very valid to take that into account."
Now, however, scientists are finding that some people with such complaints may, in fact, be detecting early harbingers of Alzheimer's.
Studies presented Wednesday at an Alzheimer's Association conference in Boston show that people with some types of cognitive concerns were more likely to have Alzheimer's pathology in their brains and to develop dementia later. Research presented by Amariglio, for example, found that people with more concerns about memory and organizing ability were more likely to have increased levels of amyloid, a key Alzheimer's-related protein, in their brains.
And, in a significant shift highlighted at the conference, leading Alzheimer's researchers are identifying a new category: "subjective cognitive decline," people's perception of fledgling changes still invisible to others.
"The whole field now is moving to this area, and saying, 'Hey, maybe there is something to this, and maybe we should pay attention to these people,'" said Dr. Ronald Petersen, chairman of the advisory panel to the federal government's new National Alzheimer's Project.
Petersen, director of the Mayo Clinic's Alzheimer's center, said preliminary results of a Mayo study of healthy older adults in Minnesota suggested something similar.
"Lo and behold, those who had a concern about their memory in fact had more likelihood" of later developing mild cognitive impairment, an early phase of dementia, he said. He said study participants with memory concerns were 56 percent more likely to be given a diagnosis of such impairment, even when results were adjusted for factors such as education, genetic risk, and such psychiatric issues as anxiety and depression.
"These people are sensing something, and there's some biological signals that correlate," Petersen said. "I think it's real."
Experts emphasize that many people with such complaints will not develop dementia. Some memory decline reflects normal aging, they say, and some concerns reflect psychological angst.
And much remains unknown about subjective concerns. In some studies, like Amariglio's, highly educated people noticed changes more readily, but in other studies, less educated people did.
Some studies suggest that people who worry more about deficits have more dementia risk, but it is unclear whether the worry reflects the changes they sense or whether worrying itself increases risk.
People with family histories of dementia could be reporting problems simply because they know about the disease and its genetic component. And, while a study presented Wednesday found slight correlations between subjective concerns and the highest-risk genetic mutation, ApoE4, that relationship remains unclear.
Not every memory concern correlated with amyloid buildup, Amariglio said.
People who forget what they wanted in the kitchen or the names of relatively unfamiliar people are probably aging normally. People who forget important details of recent events, get lost in familiar places, or lose track of book or television plots may not be, especially if they have more problems than others their age.
Experts also are not yet suggesting that doctors regularly screen people for "subjective cognitive decline" because much more research is needed and no effective dementia treatment exists.
Dr. Richard Caselli, a neurology professor at the Mayo Clinic in Arizona, said that when patients cited cognitive problems, he ruled out "reversible things" but did not recommend testing for Alzheimer's because "if we do a scan and say, 'Hey, we found some amyloid in your brain,' there's really nothing you can do."
But subjective screening has value now for clinical trials, experts say, because it can help pinpoint people at higher risk for dementia to better determine whether treatments can delay or prevent Alzheimer's.
Major studies like the Alzheimer's Disease Neuroimaging Initiative are adding subjective memory complaint categories. So will an important upcoming trial to see whether an anti-amyloid drug can prevent dementia in cognitively normal people with amyloid in their brains.
"People have been interested in this subjective concern for a long time, but we didn't have a way to say, 'Is this normal?'" said Dr. Reisa Sperling, who runs Brigham and Women's Alzheimer's program and is helping lead the coming anti-amyloid study.
One of Caselli's patients, Roger Siegel, 84, has noticed problems for at least five years, and said he now remembers about 30 percent of what he would like to, and has trouble concentrating.
"I take a shower and wonder did I wash that leg," he said.
Recently, he bought six packages of pie filling instead of one, "because I asked somebody where would I find it and the answer was aisle six, so I wrote down six, but by the time I got to the aisle, I picked up six of the thing."
Neither his wife nor Caselli perceives these difficulties.
"I know I'm losing my mind," Siegel said, "but according to Caselli, I'm fine."
Caselli said Siegel had "been saying he is declining for years" and was given a diagnosis at another clinic of mild cognitive impairment three years ago "based on his subjective complaints, when he had no evidence of decline and I told him so."
But recently fine-grained cognitive measures showed he had "slight decline on his tests, and so may possibly be at a very early stage of a very slowly progressive degenerative syndrome," Caselli said.
The working group aims to develop standardized subjective cognitive tests so when treatments become available doctors can eventually use them as a "cheap and noninvasive" way to help identify people at greater risk, Petersen said.
"We can't do MRIs and scanning on everyone," he said.