Dr. Michael Rosenbloom says primary-care physicians have a lousy track record of diagnosing Alzheimer's disease and other forms of dementia.

Now he and his colleagues at the HealthPartners Center for Memory and Aging want to give them a tool that could change that: a quick, pen-and-paper test that, if used during routine annual physicals, could detect cognitive problems in thousands of older patients who may have undetected Alzheimer's disease or other neurological disorders.

"[If] you're diagnosing these diseases when these patients are already mistaking their medications, having motor-vehicle accidents, losing their way from home — that's a failure," said Rosenbloom, clinical director of the Memory and Aging Center. "We have got to get to these patients earlier."

More than 5 million Americans older than 65 have been diagnosed with Alzheimer's disease and other types of dementia. And a recent study suggests that deaths from Alzheimer's disease have been grossly undercounted — that it may be the third-most-deadly disease in America.

Yet Rosenbloom and other Twin Cities neurologists say that by the time many people have a diagnosis, the disease already has caused serious damage, making intervention and treatment more difficult.

They want primary-care physicians to screen for the disease regardless of whether the patient complains of memory problems.

HealthPartners Care Group, one of Minnesota's largest medical practices, has been screening Medicare patients ages 70 and older at four of its clinics for more than 18 months. The test, called the Mini-Cog, takes a minute or two and is far more sensitive to mild cognitive impairment than a more-widely-used test developed 45 years ago that takes twice as long.

Among patients who were previously undiagnosed for dementia, 26 percent failed the Mini-Cog and were referred for further evaluation, said Terry Barclay, a neuropsychologist and clinical director at the Center for Memory and Aging.

Failing the test "is associated with a significant increase in emergency-room visits, hospitalizations, phone calls to the clinic … no-show rate and canceled appointments compared to those who pass the screen," Barclay wrote in an e-mail.

Essentia Health in Duluth has a similar study underway, he said, and the Mini-Cog is also being promoted by Act on Alzheimer's, a large public-private consortium in Minnesota that advocates for better dementia care.

Feared diagnosis?

Many primary-care physicians resist the idea of cognitive screening — worried that failure will lead to the diagnosis of a fatal, largely untreatable disorder, wreck their relationship with the patient and take too much time, Rosenbloom said.

He said doctors often worry that a diagnosis of dementia will cause the patient to become depressed — or worse. But he cited a study that found anxiety and depression either remain stable or decrease after diagnosis. Rosenbloom said in his experience, people often are relieved to learn the diagnosis and the disease can be managed with lifestyle changes.

"You take a huge weight off their shoulders, and their quality of life improves significantly," he said.

In addition, Rosenbloom said, about 9 percent of dementia cases result from treatable causes such as vitamin B12 deficiency, thyroid disease, depression or heavy-metals toxicity.

Barclay called routine cognitive screens "the gold standard" of care and said if patients and their families knew that, they wouldn't settle for less. "And if providers understand what the gold standard is, they won't provide anything less," he said.

HealthPartners is evaluating the potential risks and benefits of conducting cognitive screening systemwide in what it calls its "Sixth Vital Sign" project.

Memorize three words

The five-point Mini-Cog test has two parts. Memorize three words, each worth one point. Then draw a clock face and place the hands at 10 past 11, worth two points. Finally, repeat the three words. Patients scoring three points or less merit further evaluation. But it's not a diagnosis of anything, according to the geriatric psychiatrist who developed it, Dr. Soo Borson.

The Mini-Cog was developed at the University of Washington and was tested several years ago in more than 8,000 patients over age 70 in the Minneapolis VA Health Care System. None had a prior dementia diagnosis, but the Mini-Cog identified cognitive problems in nearly 26 percent of them.

Dr. J. Riley McCarten, a ­neurologist at the Minneapolis VA and the University of Minnesota who oversaw the Mini-Cog study, says standardized cognitive testing is especially important as people age. The rate of Alzheimer's diagnosis more than quadruples from age 65 to 85.

"We miss as many as 75 percent of the people who have dementia," McCarten said. Dementia, McCarten adds, is "not a priority for the health care system. It's a money loser."

That could change. In 2011, the rules governing Medicare changed to include a cognitive evaluation at the patient's annual wellness visit. But the rules don't specify how that should take place, McCarten said.

Researchers have found that exercise, learning another language and certain cognitive games might delay the progression of the disease, and some medications have shown modest positive effects as well.

"If you consider that [Alzheimer's] disease doubles every five years after the age of 65, if you can delay the onset by five years, you'd cut the [disability] rate in half," McCarten said.

While no cures are in sight, research into related neurological disorders, such as frontotemporal dementia, amyotrophic lateral sclerosis (ALS) and Parkinson's disease are showing promise and may have benefits for Alzheimer's patients, said Dr. Bruce Miller, a neurologist at the University of California, San Francisco.

Insufficient evidence?

The U.S. Preventive Services Task Force studied routine cognitive screens and, while it didn't dispute that they help spot certain forms of dementia, the panel concluded last year that there was insufficient evidence that routine administration of the tests improves decisionmaking.

"One interpretation is there's not enough evidence, so we shouldn't do it," McCarten said. "The other interpretation is that the experts all recognize that early recognition is important and the evidence base lags behind expert opinion."

Borson, who developed the Mini-Cog, said testing certainly could help doctors identify patients who need help managing their medications, with transportation and with other disorders. The downside is that physicians can misinterpret the results, she said.

Allina Health in recent years has screened an average of 14,000 Medicare patients annually with the Mini-Cog. Unlike at HealthPartners, failure doesn't automatically mean more tests. That's left to primary-care physicians, though most do so, said Dr. Steve Berg­eson, Allina's medical director of care improvement.

"It really makes you think carefully about how you give [medical] instructions to the patient or the family," said Bergeson, a practicing family physician. "You start to standardize how you're doing it, and it becomes more consistent, reliable information."

Dan Browning • 612-673-4493