OffLine Fall 2000/Winter 2001
Vol. 13 Number 1
Alzheimer's Disease (AD) is an increasingly common condition, especially for Americans over age 85. The trajectory of this neuro-degenerative disease follows a sobering path: memory loss, behavioral problems, loss of language, progressive disability and death. Rare in people under age 65, AD is a disease of society's "oldest old." Given that this is the fastest growing segment of the U.S. population, the public health burden of AD can only be expected to increase.
Unfortunately, discovery of a cure for AD in the near future appears unlikely. While the defects that destroy neurons and cause loss of brain function in AD are well known—myloid plaque and neurofibrilary tangles—the mechanisms that cause these defects are still poorly understood. "Traditionally, researchers seek magic bullets to cure diseases," said Dr. Eric Larson, professor of medicine and medical director of the University of Washington Medical Center. "I don't think we're going to find a magic bullet for Alzheimer's in the next five years. Our best hope may be to find a modifiable risk factor that delays the onset of disease for four to five years," Larson said.
If doctors can delay the onset of AD by five years, actuaries predict a 30 to 50 percent reduction in the number of people dying with the disease. That is because 30 to 50 percent of people who develop the disease are over 85 years old. Five years for people in this group means the opportunity to live the remainder of their lives without becoming debilitated by disease.
While aging is universally recognized as a risk factor for AD, the disease is not a given for all people living into their 80's and 90's. What determines who develops the disease? Genetics play a big role. In addition, researchers suspect a number of environmental risk factors are involved, similar to the way poor diet and exercise are risk factors associated with heart disease. Potential risk factors for AD currently under study include diet, health habits like smoking and alcohol consumption, low levels of education, history of depression, history of head trauma, exposure to known neurotoxins and deprivation in early childhood.
The UW's Eric Larson and colleagues, including Center for Health Studies investigators Ed Wagner and Andrea LaCroix, are currently examining some of these risk factors as part of an AD research program in existence since 1986. A primary focus is on prevention-identifying risk factors that hold the greatest prospect for delaying onset of the disease. In addition, the AD researchers are studying the disease's natural history, factors influencing survival, genetic components and incidence rates. This issue of Off Line focuses on this collaborative Group Health–UW Alzheimer's research program, with particular emphasis on identification of environmental risk factors.
Alzheimer's Disease Patient Registry (ADPR)
Between 1987 and 1996 researchers attempted to identify all Group Health enrollees 65 years and older who had been diagnosed with dementia within the past year. Researchers reviewed the medical records of 6,899 patients with possible dementia who were identified either through physician referral or surveillance of clinic logs and automated databases. Out of this pool of 6,899 patients, almost three-quarters were excluded either because the chart indicated no symptoms of dementia or the chart recorded a dementia diagnosis established more than one year before review. Nine hundred and seventy (970) subjects eventually received a physical examination, a neuropsychological evaluation and a consensus diagnosis from ADPR clinicians. Of these 970 patients, 471 were diagnosed with "probable AD." AD cases were followed (and continue to be followed) through in-person and telephone contacts in order to gauge the natural history of the disease, as well as the stability of the initial diagnosis.
This Alzheimer's Disease Patient Registry (ADPR) has been a fertile source of subjects for case-control studies comparing patients with AD (cases) to demographically similar subjects without AD (controls). The following are some of the findings from case-control studies using the ADPR: (1) cigarette smoking may be associated with a reduced risk of developing AD; (2) estrogen replacement therapy in women had no association with AD in this population; (3) a history of depression may increase the risk of AD; (4) exposure to occupational organic solvents may increase the risk of AD; (5) a history of head trauma is positively associated with AD; and (6) exposure to electro-magnetic fields is not associated with AD.
Victoria Moceri along with Larson and colleagues recently published findings in Neurology from an ADPR-based case-control study suggesting that AD may have an early-life link. The areas of the brain that show the earliest signs of AD are the same areas that take the longest to mature during childhood and early adolescence. The researchers hypothesized that deprivation in childhood could prevent the brain from achieving complete levels of maturation. The study found that the risk of AD increased by 8 percent with each additional sibling; controls were more likely than cases to have lived in the suburbs before the age of 18. Researchers point out that suburban dwelling (especially in the early 1900s) and number of siblings are related to socioeconomic level and therefore to quality of the living environment. These could be surrogate markers for higher risk of lack of proper nutrition, lack of mentally stimulating activities, reduced levels of parental attention, and exposure to environmental toxins which, in turn, could lead to less complete development of brain capacity in early life.
According to Larson, these findings add support to the concept of "brain reserve", a potential unifying hypothesis regarding AD risk. According to this concept, genetic and environmental factors that maximize brain reserve and prevent neuronal loss or loss of neuronal connectivity will decrease the risk of AD and may delay onset of signs of disease. Conversely, factors that limit development of brain reserve (deprivation in early life and lack of education) or promote brain cell loss (head trauma and neurotoxins) increase the likelihood of AD. (This idea of brain reserve is similar to that of functional aerobic capacity and cardiac reserve).
In addition to relying on ADPR data to explore causes of the disease, investigators also use the registry to help improve the quality of life of people suffering from AD. For example, a recent ADPR study found that 70 percent of AD patients had anxiety disorders which were associated with more incidents of wandering, sexual misconduct, hallucinations, verbal threats and physical abuse. It is hoped that these findings will lead to improved recognition and treatment of anxiety among AD patients.
Adult Changes in Thought (ACT) Study
While case-control studies contribute important information to the body of knowledge about AD, some researchers are uneasy with results from these types of studies. One problem is that scientists must rely on information obtained from a spouse or family member of the AD patient since the patient herself cannot be interviewed directly. Family members' efforts to help investigators identify risk factors often lead to "colored" or inaccurate recall of the AD patient's pre-disease history. To remedy this potential for recall bias, researchers have called for more prospective cohort studies.
Prospective cohort studies enroll disease-free subjects and follow them over time. The Adult Changes in Thought (ACT) study, an example of this type of study, has enrolled 2,581 cognitively intact Group Health enrollees aged 65 and older. These patients are examined every 2 years to determine whether they have developed dementia and are also asked for updated risk factor information. Subjects who develop AD are then followed up annually. (It is estimated that among persons aged 65 years and older, about 1 person in every 100 develops AD each year). This study design avoids the problem of recall bias since data are collected directly from subjects before they are diagnosed with AD (that is, if they do go on to develop the disease). Goals of the ACT study include: (1) establishment of an incidence rate; (2) cross-cultural comparisons; (3) genetic studies; and (4) identification of genetic and environmental risk (protective) factors.
The ACT study will enable investigators to re-evaluate some of their findings from the case-control ADPR studies. For example, a case-control study using the ADPR found a protective effect for cigarette smoking. However, a recently completed AD prospective cohort study in Europe found the opposite—current smoking, (as well as female gender and low levels of education) was positively associated with AD. Researchers suspect that the protective effect of cigarette smoking found in several case-control studies may be explained by the presence of survival traits—that is, the same traits that enable people to survive the adverse consequences of smoking may also protect them from developing AD. The European cohort study also reported no association between a history of head trauma and AD, another finding contradictory to that of an ADPR case-control study.
Another primary focus of the ACT study is to examine the hypothesized protective effects of different kinds of medications, including nonsteroidal anti-inflammatories (aspirin, ibuprofen) and estrogen replacement therapy in women. The study is well positioned to measure exposure to these medications through Group Health's automated pharmacy database, in existence since 1977. The ACT study will also take a close look at occupational exposures to organic solvents such as paints, gasoline and acetone/ketone. Earlier case-control studies suggested a positive association between occupational exposures to such solvents and the development of AD, but problems with response and recall bias make the results less than definitive.
The fact that the ADPR and ACT studies are based at Group Health is no accident. Several features of the Cooperative strengthen the study designs, including its stable and well-defined population, relatively low attrition rate in this age group (less than 1 percent per year excluding deaths among enrollees aged 65+), and the presence of automated data systems that permit ascertainment of exposures, such as drug use, without relying on self-report. Further, the UW researchers have benefited from the expertise of CHS investigators Wagner and LaCroix who are very experienced in recruiting and retaining older enrollees into studies.
This is the second consecutive issue of Off Line to feature a highly successful collaboration among Group Health Center for Health Studies, Group Health Cooperative and the University of Washington. It is noteworthy that, of nine programs receiving federal funding in the 1980s for AD research, only three remain funded, one of which is the UW–CHS program. In today's highly competitive research environment, this team of CHS–UW scientists has emerged in the forefront of Alzheimer's Disease research.