The Case for Screening for Cognitive Impairment in Older Adults

John Riley McCarten, MD


J Am Geriatr Soc. 2013;61(7):1203–1205 

When considering whether to pursue an evaluation for any medical problem, the basic question is: Does the benefit of knowing the diagnosis outweigh the risk of pursuing it? The latter includes the cost not only to the healthcare system, but also to the individual, family, and provider. This is particularly true when the disorder in question involves cognitive impairment. Given the unparalled importance and complexity of cognitive function, screening for impairment is a simple first step to assist in making this critical risk–benefit assessment. The alternative—to relegate the evaluation of cognition to what is obvious—is indefensible. The default assessment of mental status, "A, Ox3″ (alert and oriented to person, place, and time), is not only inadequate, but also misleading, implying that an individual is cognitively intact. To be alert requires only the ability to keep one's eyes open (see Terri Schiavo). Criteria for orientation vary widely, are often insensitive, and are of no localizing value neurologically. Worse, orientation often is assumed to be intact and reported as such in an individual who appears to be normal.

Dementia in older adults is almost invariably the result of progressive, terminal brain disease (e.g., Alzheimer's disease (AD)).[1] Affected individuals are often unaware of the severity of their symptoms and appear normal in appearance, speech, and affect. Therefore, dementia, especially in its mild and moderate stages, is frequently an occult disease. Even when recognized, individuals and families may choose to believe that symptoms are part of "normal" aging and fail to report them. A provider may have little reason to suspect problems when none are reported and the individual appears to be a healthy older adult. A careful review of medications, appointments, and other individual contacts, such as telephone calls and unscheduled visits, may identify potential concerns, but the process is time consuming, and any problems identified are only indirect evidence of possible cognitive impairment. Furthermore, individuals and families, fearful of a diagnosis, may choose to wait until the provider raises concerns. It is not unreasonable that families may expect a provider to recognize symptoms and signs of a common, serious disease. A clinic visit at which the question of cognitive integrity is never addressed may be interpreted as an indication that symptoms or signs are not troubling.

The effect of unrecognized cognitive impairment for someone who is mildly impaired may seem trivial. Safety concerns may be minimal, and there may be no substantial effect on overall health. Still, symptoms can be ignored only so long before cognitive impairment will become evident, and the sentinel event may be upsetting if not devastating. The likelihood of a misadventure on the job or while driving, managing medications or finances, or using tools increases as someone becomes more cognitively impaired, and early recognition and intervention has the potential to avert catastrophe. The more pernicious risks of unrecognized cognitive impairment may be the frequently attendant social isolation and poor health habits. Although such problems may not be apparent to the healthcare provider, the ever-shrinking world of the individual with dementia—particularly when unrecognized—may be a breeding ground for boredom, lack of exercise, poor diet, sleep disruption, agitation, irritability, depression, paranoia, and other undesirable outcomes. Recognizing and diagnosing cognitive disorders, particularly early in the course of terminal dementia, empowers individuals and families to address predictable needs and potentially live their lives differently, should they so choose. Failure to recognize and diagnose not only robs them of this opportunity, but may also tie them to a more-limited, unhappy existence.

Brain function, and specifically cognitive function, is the most-complex and -important biological function. Unfortunately, there is a common misconception among providers that, if cognition were significantly impaired, they would recognize it. In fact, primary care providers have considerable difficulty identifying mild dementia and are generally poor at documenting dementia in the medical record..[2] Even individuals with moderate dementia, several years into progressive dementia, may appear cognitively normal during a typical office visit. This cannot be overstated.

Still, if an individual looks normal, and no one complains, is it reasonable to probe cognitive function? It is undoubtedly difficult to gauge accurately. The entire educational system is continually challenged by how best to assess cognitive abilities. A brief cognitive screen will have a high rate of false positives and negatives. Can such a screen really be of any value?

The hallmark of dementias, and specifically AD, is impaired recent memory. Individuals with AD, and even the precursory symptoms of AD (mild cognitive impairment, MCI), have difficulty learning and remembering new information. Common symptoms include a tendency to repeat and misplace, to depend more on notes and calendars, and to get lost driving. Most important for providers is that individuals with impaired recent memories are, de facto, unreliable historians. Although they may provide detailed histories, all such data are suspect. Furthermore, there can be no reasonable expectation that any explanation or advice from the provider—including any prescriptions for medications—will be remembered. Quite simply, if an individual has acquired difficulty in learning and remembering new information, the provider needs to speak to someone else. It is irresponsible for a provider to rely on a history from or provide advice to an individual with evidence of memory impairment. An assessment of cognition—a "screen"—should be integral to any examination in which the individual's cognitive integrity is not already established. Because deficits in recent or short-term memory are such a prominent feature of most dementias, screening for dementia is practical. Clinic intake staff can administer a brief 2- to 3-min cognitive screen that tests recent memory—also called "delayed recall"— essentially as a vital sign. Typically, the individual is given three words to remember and then is engaged in another "interference" task, such as drawing a clock, counting backward from 100 by 7s, or generating words (e.g., animal names in 1 min), before being asked to recall the three words. Such tests are well accepted.[3, 4] As with other vital signs, if the result is abnormal, the provider may elect to retest it or test it in another way, just as when a pulse or blood pressure is abnormal on intake. Recent memory in particular is too important to be tested only once if the screen indicates possible impairment. The provider's aim is to convince himself or herself that there is or is not a problem that needs further evaluation. Screening does not provide a diagnosis. Rather, it alerts the provider to the potential need for a further evaluation. Moreover, regardless of the result, screening may serve as an invitation for individuals to consider concerns and discuss them with the provider..[5]

Viewed in this way, screening for dementia meets criteria for a valid screening program:[6]

The condition being screened for is an important health problem.

The natural history of the condition is well understood.

There is a detectable early stage (MCI).

Treatment at an early stage—in the broad sense of actions that can be taken—offers potential benefits not available at a later stage.

Reasonable screening tests—within the context of the most challenging of biological functions to assess—are available.

Individuals and providers accept screening tests.

Optimal intervals for repeating the screen—given the typically insidious nature of dementing illnesses—could be established. A large-scale annual screening trial would provide the necessary data for analysis.

Adequate health service provisions for screening are available since the introduction of Medicare's Annual Wellness Visit.

Risks of screening for AD using biomarkers.[7] are largely unknown and may be substantial, but the vast majority of older adults do not object to a brief cognitive screen and do not react adversely even to failed screens. The cost of such screening is readily balanced against the benefit to the provider, who can better manage care knowing that his or her patient may be impaired, and to the individual and family, who have the opportunity to find an explanation for their problems and can proactively address the predictable consequences of dementia if diagnosed.

Too little is known about interventions to make a compelling economic case for screening for cognitive impairment. Current medications specifically for dementia are inadequate to recommend for or against screening.[8] Although many psychosocial interventions are promising, their effect is often difficult to assess rigorously. What is known is that there is a looming pandemic of dementia,[9] with associated staggering cost projections not only to the healthcare system, but also to individuals, families, providers, and all of society.[10,11] Too little is known about how to address it, and given the gravity of diseases that cause dementia, it is shocking that as many as half of those affected are undiagnosed.[12] Step one is to make finding the disease a priority. Until then, we are operating in the dark.