14 December 2011. Howard Fillit of the New York City-based Alzheimer’s Drug Discovery Foundation (ADDF) once visited a medical practice about 50 miles from Tampa, Florida. He was intrigued that the physician running the practice saw some 17,000 patients a year, half of whom were on Medicare, the federal health insurance program for Americans 65 and older. Yet none of the patients had been prescribed cholinesterase inhibitors, the standard treatment for Alzheimer’s disease (AD). So Fillit said to the Florida physician, “You have 8,000 to 9,000 people on Medicare. If you look at the statistics, you should have 800 to 900 people with dementia.” The physician responded, “Well, doc, we just don’t see it down here.”

Fillit shared the anecdote at the Dementia Screening and Detection Meeting held on 22 November in New York City. Organized jointly by ADDF and the Alzheimer’s Foundation of America (AFA), also based there, the conference brought together AD experts, with representatives from the National Institute on Aging (NIA), to brainstorm on how to increase the detection of dementia in primary care practice. According to the Patient Protection and Affordable Care Act signed into law in 2010, starting in 2011 Medicare will pay for an annual wellness visit, which includes the detection of any cognitive impairment. This change in policy provides an opportunity as well as a challenge to those interested in dementia care. “How do we go from our knowledge base of what is memory screening to something that fits into a wellness exam?” asked Fillit, laying out the charge to meeting participants. “We will need to put out some kind of guideline as an expert panel to give to the world in a peer-reviewed publication. It should lay out the steps for how we move forward.” Although the group did not settle on a definitive set of recommendations just yet, the discussions that day highlighted the need for identifying dementia early on, as well as the challenges to adding dementia screening and follow-up in an already crowded primary care system.

The AFA has been a strong proponent for dementia screening since its inception in 2002. “We knew there was a blood pressure screening day, so we asked ‘Why isn’t there a memory screening day?’ But when we proposed the idea it came under attack,” AFA president Eric Hall told meeting participants. Despite the criticism, AFA instituted an annual National Memory Screening Day in 2003. It is held every November. Although the event remains somewhat controversial among healthcare professionals, this year it drew a record 60,000 people who were concerned about their memory and wanted to be tested. “This tells us that we are going in the right direction; there is a need for what we are doing,” Hall told ARF. On 15 November 2011, healthcare professionals offered free, confidential memory screenings to the public at over 2,500 sites nationwide, including pharmacies, malls, and senior centers.

In the doctor’s office, dementia continues to be woefully under-detected. Malaz Boustani, Indiana University School of Medicine, Indianapolis, conducted an extensive review of the literature in 2003 on behalf of the U.S. Preventive Services Task Force (USPSTF) and concluded that 66 to 80 percent of dementia cases are unidentified in primary care (Boustani et al., 2003). Several presentations at the AFA-ADDF meeting confirmed that the situation has not improved since then. (These presentations are described in Part 2 of this series.)

“Why is dementia not being screened for?” asked Richard Stefanacci, University of the Sciences in Philadelphia, Pennsylvania, in his presentation. “One in a hundred patients ask to be screened. Unfortunately, primary care physicians tend to be reactive; they have to have a patient or caregiver complain before they investigate,” he answered his own question. Moreover, Stefanacci said, “Physicians don’t feel it pays them to make the diagnosis, in part because they are not convinced that there is anything they can offer Alzheimer’s patients.”

Although there is no cure for AD, presenters argued that screening can allay fears (if the test result is negative) or reveal other underlying conditions, such as hypothyroidism, depression, malnutrition, diabetes, that might be affecting cognition. “There are some reversible causes of cognitive problems, but even if those problems are due to Alzheimer’s dementia, for some individuals it may be better to know than not to know,” Lori Frank from MedImmune, Gaithersburg, Maryland, told ARF. In her presentation, Margaret “Marge” Dean of Texas Tech University Health Sciences Center School of Medicine at Amarillo explained that having a diagnosis of dementia can help patients and their families plan for the future. “If everyone has a better idea of what is coming down the line, you have improved their lives,” she said.

Most importantly, meeting participants argued that identifying dementia can improve a patient’s overall medical care and reduce the number of hospitalizations. Soo Borson, University of Washington, Seattle, told ARF that there is evidence from several different kinds of studies that persons with dementia need specialized care. For example, it happens that a patient’s treatment for heart disease or diabetes is ineffective because the person also has undiagnosed dementia and is forgetting to take medications or to follow instructions. “We need to move away from the concept of treating dementia, but rather look at dementia as the organizing principle for treating all of a patient’s healthcare needs,” said Frederick Schmitt of the University of Kentucky Medical Center in Lexington. “If you are demented, your healthcare is going to be affected.”

Among older patients coming to primary care, can a brief screen flag most of those with cognitive impairment? Image credit: Soo Borson

Studies have shown that patients with dementia who receive dementia-specific care have a better quality of overall medical care than those not receiving such care (Callahan et al., 2006; Vickrey et al., 2006; Boustani et al., 2011). But missing from the literature is a prospective randomized trial that follows cognitively normal people and shows that, in the long run, those who are screened for cognitive impairment fare better than those who aren’t. “We know that screening improves detection, but not if it improves outcome,” Borson told ARF. The lack of this kind of evidence is why the USPSTF, which makes recommendations about screening and other preventive services, concluded both in 1996 and 2003 that there was insufficient evidence to support dementia screening in the primary care setting (see Screening for Dementia: Recommendations and Rationale, USPSTF).

At the meeting, participants also debated the cost effectiveness of dementia screening. In her presentation, Paige Lin of Tufts Medical Center in Boston showed that among Medicare recipients 65 to 85 years of age, those with dementia have higher healthcare expenses than those without. Detecting dementia early on and providing the necessary care to patients could reduce those costs. However, “the efficacy of screening depends on several factors, such as test accuracy, screening frequency, the level of dementia risk in the target population, and most importantly, the downstream costs of care after detecting dementia. Currently, there are very few published studies on the cost effectiveness of dementia screening,” wrote Lin in an e-mail to ARF. “We need better data that demonstrate the value of screening by considering both the benefits and the costs.”

Although evidence on outcomes is lacking—and there was some debate at the meeting about precisely what kind of data are needed—what is in place are several screening tests that are effective in detecting dementia. Boustani, who was one of the scientists on the dementia screening USPSTF, found evidence of high accuracy for tests like the Memory Impairment Screen (MIS), the Mini-Cog (Borson et al., 2005), and the AD8 (Galvin et al., 2007; see also ARF Webinar). Separately, several presenters who described results of different dementia screening projects also came to the conclusion that these tests all work. (Those presentations are described in Part 2 of the series.) “One of the highlights of the meeting for me was the recognition that we have good screening tools and some work well in the primary care setting,” said Borson.

In the future, the diagnosis of dementia and cognitive impairment will be enhanced by various biomarkers currently being used in research settings, according to John Morris at Washington University School of Medicine in St. Louis. Once FDA approved and validated for clinical use, the PET amyloid imaging agent AV-45/florbetapir (see ARF related news story) or others like it “could be used widely,” Morris told ARF, adding, “I believe that is coming.” One caveat is that it is unclear at this point whether Medicare will pay for the scans.

If good screening tests are at hand, then the problem may be one of too many choices. In his presentation, J. Wesson Ashford of Stanford University in Palo Alto, California, reviewed the hundreds of tests available to clinicians. Ashford, who chairs AFA’s memory screening advisory board, argued that in future, testing should be done primarily using computers. “Computer calculation of test performance can give a much more meaningful indication of an individual's performance, improving the sensitivity and specificity of any test and considerably improving the cost effectiveness of screening,” he wrote in an e-mail to ARF.

For now, the Centers for Medicare & Medicaid Services (CMS) has asked the NIA to provide guidance on which test should be used to screen for cognitive impairment during the annual wellness visit. “It should be evidence based, appropriate for the Medicare population and providers, brief and easy to use, and cost is a consideration as well,” said NIA’s Molly Wagster, speaking at the AFA-ADDF meeting with her colleague Nina Silverberg. NIA so far has identified 12 tests that are free of cost and take less than five minutes to administer. These include the AD8, Mini-Cog, and MIS, among others. However, “there is no consensus as to which tests are the best or what to recommend to a primary care physician,” said NIA’s Creighton (Tony) Phelps, who did not attend the conference in New York City. NIA has already held several meetings with representatives from other federal agencies and dementia experts, and is continuing to collect input.

Even once a test, or group of tests, is officially anointed for screening, “dementia screening is going to be challenging to do,” said Morris. Busy primary care physicians have to find time to screen for dementia among the many other health conditions that typically afflict older patients; moreover, they may not have time or resources to conduct the follow-up evaluation on a positive screen, or to find dementia-specific care for a patient. Low reimbursement rates only compound these difficulties.

AFA-ADDF meeting participants will continue discussions by phone over the next several weeks with the goal of hammering out a set of recommendations that address some of these challenges. If the task may seem daunting, there is no lack of enthusiasm. “The policy changes that have occurred in Medicare, the field’s interest in the earliest stages of disease as indicated by the new diagnostic guidelines, and the greater focus on prevention are all very exciting and energizing,” Frank told ARF.—Laura Bonetta.

This is Part 1 of a two-part series. See also Part 2.