Is the USPSTF Right About Early Screening For Cognitive Impairment?

Alzheimer’s and other dementias pose an enormous economic threat, but for now, the task force says, routine screening is unwarranted

Susan Worley

In November 2013, shortly after the U.S. Preventive Services Task Force (USPSTF) released its draft recommendation on screening for cognitive impairment in adults age 65 and older, it began to hear from prominent academic and clinical researchers and institutions, who urged the task force to assign a grade A or B, which would signify full endorsement.

Some argued that by failing to support early cognitive screening, the task force was reinforcing a discredited notion that physicians ought to diagnose only conditions for which they have a cure. Others were concerned about impeding efforts to grasp the full scope of the public health care burden of dementia, and about undermining efforts to recruit participants in clinical trials that might eventually lead to a treatment. Perhaps most important, critics listed the many ways in which early identification of cognitive impairment might benefit both the impaired and society. Many noted, for example, that early recognition of dementia could significantly reduce health care costs by preventing avoidable complications related to delirium, ranging from the mishandling of multiple medications to unnecessary hospitalizations.

Albert Siu, MD

“We found insufficient evidence to recommend routine screening of asymptomatic individuals,” but if a patient complains of memory loss, then test, says Albert Siu, MD, co-vice chairman of the USPSTF.

Then in March, the task force, which does not factor health care costs into its decisions, issued its final recommendation, declaring that there is currently insufficient evidence to recommend early cognitive screening. Unlike controversial USPSTF decisions regarding screening for prostate cancer and breast cancer, which inspired passionate opposition, this decision has been met with some confusion. Considered logical by some experts, the decision is regarded by many others as oddly out of tune with current events. Recent efforts to increase awareness that Alzheimer’s disease may begin decades before symptoms emerge, a new era of prevention trials in the realm of clinical research and Medicare’s decision to include the detection of cognitive impairment as part of its annual wellness visit suggest a collective interest in addressing dementia at the earliest possible stage. All experts agree on an urgent need for more research, but as critical research resumes, it may be necessary to identify practices that, at least on a general population level, offer greater health care gains and cost savings than screening of asymptomatic individuals.

Highlights of the decision

“Our recent recommendations really pertain to the individual age 65 or older who visits his or her health care provider without expressing any concerns or complaints about memory or cognition whatsoever,” says Albert Siu, MD, co-vice chairman of the USPSTF. “We set out to determine whether health care providers should be routinely screening such individuals. It’s important to note that our I grade [not enough evidence] does not recommend against screening; rather we have determined that the decision about whether to screen must be left to the judgment of the clinician. We found insufficient evidence to recommend routine screening of asymptomatic individuals, but clearly, if a patient complains of memory impairment, or if a family member comes in expressing concerns about memory, our recommendations don’t apply. If a patient is complaining about an issue, then, of course, testing might well be appropriate.”

Was cognitive impairment defined?

Some experts found that the current USPSTF recommendation lacks a clear definition of cognitive impairment. Of particular concern is the fact that the terms cognitive impairment and dementia are used interchangeably in some parts of the USPSTF statement. Howard Fillit, MD, executive director and chief science officer of the Alzheimer’s Drug Discovery Foundation, shares this concern with some other experts:

“When a clinician uses screening tests such as the MMSE [Mini-Mental State Examination], it is really to evaluate cognitive function across multiple domains. It would be a misnomer to talk about screening for dementia. The diagnosis of dementia is, of course, made not from screening tests but by a clinician, usually according to DSM IV criteria. The clinician determines whether the patient has cognitive impairment in more than one domain, whether it has been progressing, and whether the impairment is severe enough to lead to loss of functioning in the patient’s daily life. The clinician must ensure that the impairment is not caused by something potentially reversible, such as depression or the side effects of medications. So the clinician must evaluate not just cognitive functioning but the patient’s overall status, in proper context. The tests mentioned in the statement are used to screen for cognitive impairment, and if impairment is detected, the appropriate steps must be taken to diagnose dementia.”

Screening the wrong population?

Many experts agree that general screening of asymptomatic people beginning at age 65 is unwarranted, and for many this seems to raise the question of whether the task force examined the right population. The current recommendation results in a lack of guidance, and indirectly, in a lack of payment for screening elderly people who have a much greater likelihood of being impaired than do younger people.

“Cognitive screening in a general population is obviously wrong,” says Jason Karlawish, MD, professor of medicine, medical ethics, and health policy at the University of Pennsylvania. “Screening really must be focused on an older population. The lower we set the age, the more we increase the chance of false positive detection. I’m in favor of screening when the probability that you are going to pick up cognitive impairment is reasonably high, and that is in the setting of practices that care for older adults. Among geriatricians, I would say, it is a fairly standard practice to assess your patients’ cognition.”

Jason Karlawish, MD

The benefits of dementia treatment may include improvements in mood or cognition, but they are not easy to measure, says Jason Karlawish, MD, a medical ethicist.

Jason Brandt, PhD, professor of psychiatry and behavioral sciences at Johns Hopkins School of Medicine, concurs. “There are many good cognitive screening tests that hover at about 90% sensitivity and about 90% specificity, but to get the best value from these tests, it’s important to administer them to the right population. If you administer a particular cognitive test to all patients over 65 at a general practitioner’s office, you will have many more false positives than true positives, while if you test a geriatric population, you will have many more true positives. The prevalence of a condition in the population you are testing makes a huge difference.”

Letting clinicians decide

Most experts agree with the task force that decisions regarding the screening of some, but not all, people age 65 and older should be left to the judgment of each clinician. Currently, testing is covered by private insurers and Medicare only when patients have received a diagnosis based on signs and symptoms.

“To say that we will only do testing in patients who have signs or symptoms of cognitive impairment is highly problematic in my opinion,” says Pierre Tariot, MD, a geriatric psychiatrist and director of the Banner Alzheimer’s Institute, “and it raises this question: How do you know whether patients have signs or symptoms unless you look? You are left with a tautological absurdity. I’m embedded in a large health care system that is attempting to grapple with this very problem. Baseline testing on everybody is not practical, but we are designing a pathway for people for who have a reasonably high risk of having dementia. These patients will be evaluated in a primary care setting, and the results of that evaluation will trigger a care path for clinicians to follow.”

Reid Blackwelder, MD, president of the American Academy of Family Physicians offers his perspective. “Approximately 25% of our patient base is made up of individuals age 65 or older, so family physicians play a significant role in caring for an aging population. We encourage our members to regard USPSTF recommendations as important guidelines for practice, and we agree that routine screening of asymptomatic individuals over the age of 65 is not warranted. However, we also recognize that the I recommendation by the task force is not a recommendation against screening. Guidelines are for general populations, and clinicians don’t treat general populations, they treat individuals. Depending on each patient’s unique circumstances, we outline current evidence regarding any preventive test and what the benefits would be for a particular patient, and then allow that patient to decide.“

Value of early detection

In assessing the value of early detection, the task force focused largely on the limited effectiveness and adverse events associated with current medical interventions, as well as the limited evidence supporting harms or benefits of current non-pharmacologic interventions.

“Difficulties associated with measuring the benefits of diagnosing and treating dementia have haunted the field of Alzheimer’s disease for a long time,” says Karlawish. “The benefits of treating dementia are not easily measured, compared with the benefits of treating, for example, heart disease, which has easy-to-measure events such as death, stroke, or blood lipid levels. The benefits of dementia treatment may include improvements in functioning, mood, behavior, or cognition. These are not so easy to measure and in clinical practice clinicians don’t routinely use these scales. Instead, a family may tell you that a patient is less agitated, that the patient feels better, and also that the family feels better since they aren’t spending so much time and money traveling to and from the ER, and so forth. You can measure some of these things, but it’s difficult to do so, and therefore it’s difficult to prove that there is a benefit to screening or that a treatment is cost-effective. The task force has acknowledged that we have good instruments to detect cognitive impairment, but our inability to measure the benefits of identifying and addressing impairment is still a source of frustration.”

“It’s estimated,” says Tariot, “that more than half of all patients with dementia never receive a diagnosis, and patients with dementia consume approximately 300% more health care dollars per year than patients without dementia who are otherwise clinically identical. They cost more because they develop delirium and behavioral problems, the mismanagement of their other medical conditions is common, and they have inappropriate hospitalizations and emergency room visits – and prolonged hospitalizations. Patients with dementia and their families have very unsatisfactory interactions with the health care system, and providers also may have unsatisfactory interactions because they aren’t sure what to do. We have to fix this, and ignoring this problem is not going to make it go away.”

Are interventions available?

“At an early stage of Alzheimer’s disease or a related dementia,” says Jeffrey Cummings, MD, ScD, of the Lou Ruvo Center for Brain Health at the Cleveland Clinic, “the patient embarks on a journey in which he or she must learn to cope with the disease. It seems to be the opinion of the task force that because there is no disease-modifying treatment, and because symptomatic treatments have limited effectiveness, it is therefore best for a patient not to know that he or she is on a journey. It’s as if they are saying, ‘You’ve left shore, but we don’t think it’s important for you to know that you’ve left shore.’ At the Cleveland Clinic, we have a different perspective. We say, ‘You’ve left shore, you’re on a new journey, you need guidance, and we want to help guide you.’ ”

Jeffrey Cummings, MD, ScD

“It’s as if they are saying, ‘You’ve left shore, but we don’t think it’s important for you to know that you’ve left shore,’” says Jeffrey Cummings, MD, ScD.

Expressing the sentiments of most experts interviewed for this article, Cummings adds, “There are many types of intervention currently available. It’s important to educate the family, to help them understand that although the patient has a disorder that interferes with memory, the patient may retain some skills, and adjustments may be made to compensate for others. There are safety issues that should be guarded against. Patients may consider clinical trials. And early recognition may enable families to defer nursing home placement, one of the most costly aspects of health care, avoid emergency room visits, and better manage co-morbidities and medications related to them.”

Value of delaying dementia

Until a disease-modifying treatment is available, the greatest impact on health care outcomes and costs may lie in a focus on the modifiable risk factors associated with dementia. Indeed, this is one point on which task force members, researchers and clinicians, and insurers seem to agree. Evidence suggests that the incidence of dementia is greater among individuals with diseases that damage the heart or blood vessels, and it is estimated that up to 50% of all cases of Alzheimer’s disease may be related to modifiable risk factors associated with chronic conditions such as diabetes, hypertension, and obesity.

“If we could delay onset by five years, we would be able to reduce cases by approximately 50%,” says Howard Fillit, MD, head of the Alzheimer’s Drug Discovery Foundation.

“In the past,” says Fillit, “AD was regarded solely as a neurodegenerative illness of the brain, but now epidemiological and other data are clearly indicating that systemic diseases such as diabetes, hypertension, and obesity are risk factors for AD, and by managing these illnesses we can at least delay the onset of AD. If we could delay onset by five years, we would be able to reduce cases by approximately 50%.”

While research investigating the value of early screening for cognitive impairment continues, it may be a source of consolation for patients and clinicians that many screening tests for modifiable risks factors associated with dementia are covered by major insurers.

“Aetna supports the current position of the USPTF and does not cover screening for cognitive impairment,” says James Cross, MD, vice president and head of Aetna’s national medical policy and operations. “However, Aetna does support and provide coverage for the USPSTF recommendations that are related to several of the risk factors for cognitive impairment, including counseling on tobacco cessation, alcohol use, a healthful diet, and physical activity. We also cover screening for high cholesterol, hypertension, and depression.”

IBC has no specific coverage for routine screening

Don Liss, MD, senior medical director of clinical programs and policy at Independence Blue Cross, which covers approximately 3 million lives in greater Philadelphia and New Jersey, talks about the United States Preventive Services Task Force Guidelines on cognitive screening.

The USPSTF is one of the most respected bodies in the field of prevention. They apply a very elegant and careful process to the analysis of very complicated issues. In this case they found that while currently available cognitive tests, such as the MMSE [Mini-Mental State Examination], perform pretty well, the results of such tests make relatively little difference in outcomes for patients. The drug therapies that are available to slow cognitive decline are not very effective. We certainly hope that better treatments will be developed, and if that happens, cognitive screening may be appropriate.

It’s important to note that the current recommendation should not discourage a clinician from performing an analysis if the clinician believes there is a reason to do so – if, for example, the patient’s family is concerned. If a change in behavior or personality suggests cognitive decline, then it’s very appropriate to engage in testing.

Currently we do not categorize cognitive screening as a separately compensable service. The tests that the task force examined were largely tests that can be given in 10 minutes or less, and such tests can be performed as part of a routine office visit. If this testing is performed as part of an annual preventive or wellness visit, there is no separate claim, or CPT code for the service, and no separate payment is made.

If, in the future, cognitive testing is given a grade of A or B [testing is recommended], we will need to consider whether such testing should be included as a part of a wellness visit – just as the task force recommends that blood pressure should be checked, and we consider that part of the visit – or whether it should be a separately compensable service.

Insurers faced a similar issue when the task force recommended screening for domestic violence as a woman’s preventive service. Questions such as “Have you been threatened or subjected to violence?” are part of a typical office visit, with no separate payment provided for this screening.

Should we all know our cognitive status?

Some experts anticipate the availability of a simple predictive test for Alzheimer’s disease (AD) sometime this decade. But if such a test should become available before the discovery of a disease-modifying treatment, who exactly should be screened?

“Screening refers to an attempt to detect, among the populace, disorders or risk factors of which people are unaware,” says Jason Brandt, PhD, professor of psychiatry and behavioral sciences at Johns Hopkins School of Medicine.

Jason Brandt, PhD

Most people at risk for Alzheimer’s disease don’t want to know, says Jason Brandt, PhD, professor of psychiatry at Johns Hopkins School of Medicine.

“Based on that definition, it certainly would be problematic for any organization to mandate screening, particularly for a disease for which there is no cure or highly effective treatment, and to which a lot of stigma may be attached.

“However, it is reasonable to make available, to those people who want it, information about their risk of getting AD or a related dementia – information that is individually tailored and helpful, and which may help an individual better plan for the future.”

Brandt draws on his experience conducting pre-symptomatic testing for Huntington’s disease (HD) at Johns Hopkins in one of the first programs in the country to offer such testing. He draws parallels between the two types of testing, while pointing out obvious differences.

“The people who were tested for HD in our program were at 50% risk before testing,” Brandt says, “because HD is caused by a single dominant gene and each person had a parent with the neurological disorder. Our program was one of the first to discover the consequences of informing people that they will develop a devastating disease. We learned a great deal about how people handle that news, and about whether most people really want to know their future risk in the first place. We learned how to provide predictive testing, only to those who really wanted to know, in a way that maximizes usefulness and minimizes psychiatric or social morbidity.

“We ended up testing many fewer people than we thought we would, because it turned out that most people who were at risk for HD didn’t want to know. However, for those who did want to know, having the knowledge was very beneficial. The at-risk people we tested all said that they would take the test again, even those who received bad news. We did not see significant psychological fallout among those who tested positive for the HD gene mutation.”

Brandt points out that testing for Alzheimer’s disease is more complicated. “In HD, we have a single-gene test with essentially 100% accuracy, and we don’t have anything nearly that predictive for AD yet. It’s unlikely that we will have a test for AD that has high predictive validity any time soon.

“But even if we did, how many people would want a test for such a late-life disorder? In HD, onset occurs around age 40; knowing your gene status can affect how you lead your life, including decisions regarding having children or career options.

“I suppose you might do things differently if you knew early in life that you were sure to get AD around age 75. You might not wait until retirement to take that round-the-world trip, or make different financial decisions, or plan for eventual nursing home care. But the implications for one’s life would clearly be different.”

For further reading

Moyer VA. Screening for cognitive impairment in older adults: U.S. Preventive Services Task Force Recommendation Statement. 2104; Ann intern Med

Lin P-J, Fillit HM, Cohen JT, Neumann PJ. Unintended benefits: The potential impact of addressing risk factors to prevent Alzheimer’s disease. Health Affairs. 2014; 33:547–554.

Karlawish, J. How are we going to live with Alzheimer’s disease? Health Affairs. 2014; 33:541–546

Cordell CB, Boorson S, Boustani M, et al. Alzheimer’s Association’s recommendations for operationalizing the detection of cognitive impairment during the Medicare Annual Wellness Visit in a primary care setting. Alzheimer’s and Dementia 2013; 9:141–150

Aretouli E, Konstantino KT, Brandt J. 4 year outcome of mild cognitive impairment – the contributions of executive dysfunction. Neuropsychology 2103; 27: 95–106


Legislation and regulation
By Richard Mark Kirkner
Legislation and regulation
By Richard Mark Kirkner
Legislation and regulation
By Richard Mark Kirkner

A blueprint for high-volume, high-quality lung cancer screening that is detecting cancer earlier—and helping to save lives

Clinical Brief

Multiple Sclerosis: New Perspectives on the Patient Journey–2019 Update
Summary of an Actuarial Analysis and Report