Reducing Medicare Costs: The Risk, Prevention, and Treatment of Cognitive Impairment
Reducing Medicare costs will be an important enterprise in the future as the baby boom generation retires. In the Medicare Prescription Drug Improvement and Modernization Act of 2003, Congressional debate was as much about “limiting government’s future exposure to cost increases”1 as it was about creating a new prescription drug benefit for Medicare. Grassroots efforts to reduce costs through incentives to physicians have been attempted, but without much success. An actuarial team from the Centers for Medicare & Medicaid Services (CMS) reported in 1998 that when faced with reduced fees, physicians “simply increased the amount of services to each patient and submitted claims for slightly higher levels of service so that they could continue to earn as much money as they had in the past.”2
Lieberman et al3 proposed two approaches for reducing the growth of Medicare spending. First, spending in the different regions of the country can be made more equal by lowering the amount of Medicare spending in high-spending regions.3 Healthcare researchers have found that Medicare could save as much as $120 billion a year, or 30% of its budget, by reducing spending in all areas of the United States to that of the lowest-spending regions.2 There is a clear logic to this, because the quality of care and outcomes in higher-spending areas are no better than similar measures in lower-spending regions.3 But, there are practical problems in reducing costs in high-spending areas, not the least of which is physician resistance to reductions in income, and the resultant circuitous paths around the system.
A better approach is the second one, which targets those individuals who spend the most in terms of the Medicare dollar. It is estimated that the 5% of Medicare recipients spend 47% of the Medicare dollar, and the diseases that account for most of this spending are, in order of importance: congestive heart failure, diabetes, and cognitive impairment. Lieberman et al3 suggested that we need to identify individuals who are likely to become high-spenders and intervene to prevent them from falling into these categories. In other words, we need to identify those at risk and accurately diagnose, prevent, and treat their problems in the early stages.
This article focuses on cognitive impairment, not because it is more important than the other diseases, but because it is more likely to affect a greater number of the elderly. About 23% of persons using Medicare have some form of cognitive impairment.4 It is a leading cause of placement in nursing homes and, therefore, a major source of expense for both Medicare and Medicaid. It is more difficult to diagnose than heart disease or diabetes, so improvement in diagnosis at the primary care level may greatly help in preventing the progression of the disease and aid in its treatment. The diagnosis of cognitive impairment must begin with an accurate definition of its symptoms and those of related diseases (eg, dementia, Alzheimer’s disease [AD]).
Cognitive Impairment and Related Diseases
The literature on cognitive impairment identifies the early symptoms of the disease with the term “mild cognitive impairment,” or MCI. MCI is “a transitional state between normal aging and dementia” that involves both memory loss and some loss of cognitive functioning.5 The type of cognitive loss varies greatly with the individual, as some may lose language skills first, while others may decline in executive functioning or learning abilities. MCI is very difficult to diagnose in a primary care setting because of the subtlety of its symptoms and the hurried nature of most encounters with the physician. (See Hill et al6 for diagnosis procedures.)
The symptoms for MCI are more subtle than those for dementia, and so are the potential obstacles for prevention and treatment. It has been demonstrated that those with MCI are more likely to develop dementia, and perhaps AD, so patients with MCI tend to resist being classified as “cognitively impaired,” as opposed to experiencing “normal aging.” The fear of further deterioration is amplified by the stigma attached to having MCI, which naturally leads to a denial of the condition and minimal attempts to prevent its further development.
The diagnosis of dementia is somewhat easier. Dementia has a number of clear symptoms that can be recognized by family members, if not in a brief encounter with the physician. The symptoms of dementia are gradual increasing memory loss, confusion, unclear thinking, losing problem-solving skills, agitated behavior or delusions, becoming lost in formerly familiar circumstances, and loss of interest in daily or usual activities.7 Dementia may be caused by “very small stokes” that affect the flow of blood to the brain, a condition called vascular dementia (VaD).7
The prevention of VaD is affected by how we define it, and distinguish it from other nonvascular forms of dementia. It might be relatively mild and “not interfere substantially with everyday activities,”8 or it may “exhibit deficits across multiple cognitive domains,” and significantly affect memory function.9 (See Broich10 for comments on prevention.)
In defining cognitive impairment, one must include AD since it is “the most common cause of dementia.”7 With AD, patients “lose functioning neurons (nerve cells) in areas of the brain dealing with cognitive function and memory,” often because of abnormal protein buildup in the brain.7 Epidemiological studies report that 41% of patients with AD experience psychosis, 36% delusions, and 18% hallucinations, with most of the psychosis occurring in first 3 years of the disease.11 In the end stages, hallucinations become more common and are associated with increased risk of death.12 Limited research indicates that AD and VaD may occur together, with similar brain lesions occurring in patients, combining to form a condition known as “mixed dementia.”13 (See Kohler et al14 and Langa et al13 for more on symptomatology.)
Finally, dementia with Lewy bodies (DLB) also is a common form of dementia. It is characterized by progressive cognitive decline, combined with fluctuations in alertness, visual hallucinations, and parkinsonian motor symptoms. There is no cure for DLB. Clinical Geriatrics has published several articles in recent years dealing with DBL to which the reader might refer, namely Osborne,15 Mahgoub,16 and Borson and Fraser.17
Risk Factors and Prevention
There are numerous risk factors for cognitive impairment, ranging from head injuries to adult-onset diabetes. But the good news is that severe cognitive impairment in the form of dementia has declined in the elderly population in the United States in recent years. From 1982 to 1999, there were 310,000 fewer cases of severe cognitive impairment, and the decline is more marked among men than women. Possible reasons for this decline include improved “medical therapies and better education” among the elderly.18 VaD has declined in part because of the overall decline in strokes and chronic disability in the population. Other forms of dementia have declined as more educated elderly persons avoid common risks and follow medical advice more carefully.
Risks for development of cognitive impairment in later life can come from obvious sources like trauma to the head. In a study of professional football players, cerebral concussions were found to be related to the development of MCI in later life. Retired professional players with three or more concussions during their years in the NFL were five times more likely to develop MCI than those without concussions, and they also were much more likely to suffer significant memory loss.19
Another risk is one’s overall physical condition. Obesity, high blood pressure, and high cholesterol can all affect the vascular system and increase the likelihood of cognitive impairment. Kivipelto et al20 reported that obesity at midlife is associated with “risk of dementia and AD even after adjusting for sociodemographic variables.” When high blood pressure and cholesterol are added, the risk increases in an additive fashion. Since obesity is a risk factor, it is not surprising that diabetes also increases the risk of cognitive impairment. Diabetes often results from obesity in adults, so the linkage between diabetes, cerebrovascular disease, and cognitive impairment is both logically and empirically verified. Diabetes is a risk factor for VaD, but it does not predispose patients toward development of AD.21
The decline that accompanies AD is related to a number of factors, both genetic and environmental, but one significant environmental risk is the lack of mental activity. Cortical atrophy is more likely to occur if the brain is not used in a challenging way on a regular basis. In a Swedish study of AD in twins (thereby controlling for the genetic factor), it was found that “greater complexity of work, and particularly complex work with people,” is associated with lower risk of AD.22 That comprehensive study involved over 10,000 members of the Swedish twin registry, and controlled for age, gender, and level of education.
In summary, the risk factors for cognitive impairment, in the form of MCI, AD, VaD, or other dementia, can include: head injuries, obesity, high blood pressure, high cholesterol, diabetes, and lack of challenging mental activities. In addition, there are other genetic factors that scientists are only now beginning to understand, as research on the human genome proceeds. These risks are well known and documented in the medical literature, but there is less agreement on how to prevent the various forms of cognitive impairment.
We know that prevention of cognitive impairment involves diet and exercise, thereby reducing the common risk factors, but there is little agreement on the best types of diet and forms of exercise. Diets and books on diets are so numerous that most Americans have tried at least one diet. When followed, most diets will result in weight loss in the short term, but over the long term weight is regained. Often, the reason is that the diet is either boring or too demanding to maintain over the long term.
One study published in The Journal of the American Medical Association reported that the Mediterranean diet reduces mortality rates when combined with nonsmoking, moderate alcohol consumption, and at least 30 minutes of exercise daily.23 But this is probably true of many diets, whether they are low carbohydrate, low fat, low calorie, or some other form of low consumption. Ultimately, in the not-too-distant future, human genome research may discover genes that promote obesity, and will develop drugs to suppress their activity. It will be necessary to show some restraint in eating and to have some balance of the major food groups in one’s diet, but the problem of obesity may be minimized. With those developments, cognitive impairment from the major vascular causes should decline significantly.
Currently, the new field of nutrigenetics is studying “the interaction of dietary components with the genome” in an attempt to understand “the gene-based differences in response to dietary components.”24 Nutrigenetics will be combined with nutraceuticals to find pharmaceuticals that are “most compatible with health based on individual genetic makeup.”24 This personalized medicine of the future, which is the product of genomic research, likely will develop drugs to control weight based on individualized genomic differences.
However, genomic research will not find a fitness gene that replaces the beneficial effects of exercise. Nor will it stimulate the mental activity required to prevent AD in elderly populations. Mental and physical exercise and self-discipline will always be required for health as people age.
So how does a society encourage that type of activity? The simplest form of exercise, and the most likely to be used among the elderly, is walking. A study of elderly men in Hawaii found that walking is related to a reduced risk of dementia, and especially the ability to walk quickly.25 However, the authors cautioned that other studies have not found associations, and the reason may be that the climate in Hawaii more easily permits year-round walking, which is often not possible in many other climates.
Blair and Church26 noted that cardiovascular fitness can be achieved through a variety of exercises—brisk walking, bicycling, swimming, housework, gardening, etc. The conventional wisdom today is that one should choose an exercise type that one enjoys, and simply do it every day for least 30 minutes. So, the challenge for physicians, researchers, and policymakers today is getting sedentary individuals off the couch and into some form of physical activity. The sooner one begins an exercise program in life, the more likely one is to ward off cognitive impairment in old age.
With regard to mental activity, there are several possible approaches. At the societal level, more effort needs to be placed on making jobs less routine and more stimulating. New approaches to work design can help workers to use their minds more often and with greater flexibility. At the individual level, time away from work should be spent doing things that are more mentally challenging, especially during the retirement years.
Verghese et al27 reported that the risk of dementia is reduced among the elderly by engaging in certain leisure activities. These activities include reading, playing board games, playing musical instruments, and dancing. They noted that these activities also increased happiness and quality of life among elderly persons in a community setting.27
While the evidence is not conclusive, it appears that cognitive decline can be delayed or prevented among all age groups by physical and mental exercise. Studenski et al28 noted that a “wide range of animal and human studies provide evidence for the potential of physical and cognitive exercise in promoting cognitive health in later life.” Further studies are needed to clarify the genetic versus environmental components related to decline.
One of the dangers of cognitive impairment among the elderly is that it can lead to a series of other mental problems, or neuropsychiatric symptoms. Some of these symptoms include apathy, depression, agitation, and aggression.29 As these symptoms develop, it becomes more likely that caregiving is impossible at home and nursing home care is required. This not only diminishes quality of life, but also creates enormous expense for families and programs like Medicare and Medicaid.
A recent study by Rapp et al30 reported that a lifetime history of depression is associated with increased plaques and tangles in the brain, which are markers of AD and cognitive decline. A history of major depressive disorder, like AD, is likely to affect the brain’s memory-related temporal lobes. Previous studies also have linked depression and AD, as noted in the study by Rapp et al.30
Treatment and Cost
The treatment of cognitive impairment today is limited by our knowledge of the genetic aspects of the disease. In the future, it may be possible to tailor drugs toward individual genetic makeup and treat cognitive impairment more effectively. But for now, the various treatments for cognitive impairment are only marginally effective. Cholinesterase inhibitors have had some positive effect in slowing the decline of patients with AD, but the clinical importance of the demonstrated effect of the cholinesterase inhibitors has been arguable.31 Statins also are being developed for use in patients with AD, based on preclinical and observation studies, but their effectiveness is not established. Anticonvulsive compounds have had some positive effect on mood stabilization and have reduced agitation to some degree in patients with AD and other dementias. Apathy has been treated with some effectiveness by psychostimulants and antidepressants, such as desipramine and fluoxetine.32 However, there are no medications at present that prevent dementia in any of its forms, even the relatively mild form of cognitive impairment.33
The best-case scenario for treatment is usually in the home, which is easier in the early stages. More than 70% of all patients with dementia “are cared for at home by family caregivers, usually spouses.”32 But in time, behavioral disturbances may force caregivers to institutionalize their spouses or other loved ones. Dementias tend to agitate patients and lead to aggressive behavior toward the caregivers, who in turn may retaliate and abuse their spouses and family members.32 The “absence of the usual affection and emotional engagement from patients toward caregivers is also a source of distress.”32 It is not surprising that neuropsychiatric symptoms such as agitation, apathy, and aggression are “important drivers in the cost of care for dementia patients.”34
Once a patient is institutionalized, treatment becomes much more expensive. Surprisingly, many caregivers in institutional settings do not know how to cope with patient behaviors. Larson et al35 reported that skilled nurses in nursing homes feel least knowledgeable in dealing with dementias and other mental problems, as opposed to dealing with physical problems such as heart disease. Oh et al36 found that caregiver training in a nursing home can be effective in reducing the amount of aggressive behavior of patients with cognitive impairment. Such training also is effective in reducing aggressive behavior from the nursing staff itself.
Treatment in the nursing home can be aided by relatively inexpensive methods to reduce aggression, agitation, and apathy. The physical environment can include music and wandering areas for patients; patient activities can include walking, massage, pet therapy, and cognitive remediation activities. Patient treatment might use one-on-one interaction, which is more expensive, along with behavior therapy and structured activities. And professional caregivers could benefit from knowledge of drugs used by patients and their side effects, as well as neuropsychiatric symptom recognition.32 Most of these interventions are inexpensive.
This article is concerned primarily with reducing the costs of treatment for Medicare patients, so we need to ask, “How might overall costs in the system be reduced?” First and foremost is prevention. Cognitive impairment may be prevented through healthy habits of diet and exercise throughout a normal lifespan. Daviglus et al37 observed that body mass index (BMI) in young adulthood and middle age is related to Medicare expenditures in older age. A high BMI is associated with obesity, cardiovascular disease, diabetes, and probably with cognitive impairment. Arkansas is the only state in the United States that monitors the BMI of children in schools and suggests changes to parents of children who are overweight. If other states followed this example, it could set the stage for improved health throughout a lifetime, thereby reducing the likelihood of cognitive impairment in old age.
Medicare costs for cognitive impairment begin to rise significantly after the patient reaches the nursing home. Hill et al6 reported that patients with VaD have the highest annual healthcare costs of any nursing home patients, substantially higher than those with cardiovascular disease. Among the elderly who receive home healthcare from either Medicare or Medicaid, those with severe cognitive impairment are most likely to receive services, controlling for medical conditions and sociodemographic variables.38 Among African Americans, cognitive impairment is a significant predictor of those likely to be placed in nursing homes, based on Medicare expenditures and a diagnosis of AD.39
One way of limiting costs to the Medicare program is by limiting access to care. Murray et al40 reported that the prospective payment system (PPS) established by the federal government in the 1980s has been successful in limiting the costs of treatment for cognitive impairment. They find that post-PPS patients have “less cognitive impairment, more depression, and most family support.” PPS forces those with cognitive impairment to remain at home until it is absolutely necessary that they be institutionalized. However, one should question the ethics of this approach.
Another approach to reducing costs has involved the use of feeding tubes. More than one-third of patients with severe cognitive impairment in nursing homes have feeding tubes.41 While feeding tubes may prolong life for those unable to feed themselves, there is controversy over their use, especially in end-stage dementia. Ordinarily, patients must consent to allow caregivers to use feeding tubes to prolong their lives.
In the United States, state laws vary concerning the use of feeding tubes as a normal treatment or as a ‘heroic measure’ to save lives. In New York State, for example, they are not considered a heroic measure, so withholding a feeding tube requires a patient’s written request or “clear and convincing evidence” that they would refuse a feeding tube if given the choice.42 But even within states, the practice varies according to institutional and residential characteristics. For example, for-profit nursing homes use feeding tubes more frequently than their nonprofit counterparts, or state-affiliated institutions. It is used more often in for-profit institutions, possibly because feeding tubes are less expensive than staff-feeding by hand. In addition, Medicaid pays at a higher per-diem rate for patients using feeding tubes.41 The use of feeding tubes raises the price of care for the taxpayer while increasing reimbursement for nursing homes. The effect on patients also is negative, making them an object of care, isolating them from others, and demoralizing both patients and family members.
Current methods of reducing Medicare costs of caring for elderly patients with cognitive impairment are questionable. The use of the prospective payment system is rational when it keeps families from institutionalizing the elderly unnecessarily, but it is unethical to use it to deny nursing home care for those who need this level of care. Similarly, the use of feeding tubes for nursing home residents as a substitute for feeding by hand may be unethical.
In the future, the best hope for lowering the costs of treatment lies in genomic research and more systematic approaches to personal exercise. It is realistic to assume that within 20-30 years, genomic research will provide drugs that prevent obesity. Diet will still be important in maintaining the overall health of the body, through the proper balance of nutrients, but it should not contribute to excessive weight gain. There should be a great deal of freedom in choosing one’s diet. Genetic research also will provide drugs for directly treating cognitive impairment, so that elderly patients may be able to lead more normal lives, potentially with less institutionalization.
However, there is no magic bullet that will take the place of exercise. Exercise will always be necessary for the health of the body. The United States needs a more systematic approach to exercise, beginning in the schools. Once out of school, adults should be able to exercise in their neighborhoods without great trouble or expense. Neighborhoods should be walking and jogging “friendly,” with sidewalks, trees, jogging trails, and nearby parks. Other forms of exercise, like swimming and biking, should be available year-round in nearby public and private facilities for a modest cost.
Cognitive impairment is a condition in the elderly that may be preventable. It is a great tragedy to see loved ones ending their lives without dignity and without hope. The best way to reduce Medicare spending and maintain the health of our elderly population is to fund genomic research on obesity and cognitive impairment, while at the same time promoting a more systematic approach to encouraging exercise throughout a lifetime. Both of these ends will require a significant governmental and private sector investment, but it promises to be cost-effective in the long run.
The authors report no relevant financial relationships.