- Email: [email protected]
Screening for Memory Loss, Alzheimer's Disease, and Dementia
B Screening for Memory Loss, Alzheimer’s Disease, and Dementia
Appendix B
Screening for Memory Loss, Alzheimer’s Disease, and Dementia Primary care physicians and other healthcare practi-
tioners routinely screen for many diseases, including hypercholesterolemia, hypertension, and prostate and breast cancer, to mention a few. Yet Alzheimer’s disease, one of the most common disorders in older adults, is rarely screened for (Solomon, 2008; Solomon & Murphy, 2008). Interestingly, not only do most patients readily agree to memory screening, but also many are enthusiastic about having their memory treated as seriously as other functions of the body. In a study that we conducted several years ago (Solomon et al., 2000), all patients over the age of 60 visiting a primary care practice over an eight-week period were asked if they were willing to undergo a brief (less than 10 minutes) screen of their memory. Although both our group and the primary care physician who collaborated on the study had some initial concern regarding the patients’ reaction to being approached about the memory screen, we were all relieved when the typical response of patients was being pleased that their physician was concerned about their memory. Of the 145 patients aged over 60 who visited the practice during this period, 139 agreed to be screened. The few who did not agree to being screened at that time typically did so because of time constraints, and most indicated that they would be willing to be screened at a later time. The results of the study indicated that, of the 139 screened, 13 screened positive (requiring follow-up regarding their memory function); this result was about what we would expect given an incidence of dementia of approximately 10% in individuals over the age of 60. Ten of these patients were later diagnosed with Alzheimer’s disease, two with a mixed dementia, and one had mild cognitive impairment. This study showed both that screening in primary care can identify patients with Alzheimer’s disease and that patients are overwhelmingly in favor of having their memory screened.
To Screen or Not to Screen? To an increasing number of clinicians, the argument that we screen our patients for memory disorders seems obvious. But screening for cognitive function is not without its critics. The criticisms come from both practical and theoretical concerns (Box B1). TO SCREEN There are a number of reasons to screen our patients in the context of day-to-day clinical practice. Below we discuss a number of the more important reasons to screen patients for memory loss. Underdiagnosis of Alzheimer’s Disease
One of our colleagues who has a very busy primary care practice tells a story about a 72-year-old woman who came in for a check-up for her type 2 diabetes. For reasons our colleague did not understand, her medications did not seem to be controlling her insulin levels. She reported taking the medications as prescribed, monitoring her diet, and even exercising daily. He was puzzled. At the end of the visit she talked to him about a trip she had taken with her sister last month. She conveyed the details of the trip without difficulty. They then walked out of the examination room to the waiting room together, where they met the patient’s sister. Our colleague mentioned the trip the two sisters had recently taken only to be confronted with a blank look. It turns out that the two sisters had taken the trip, but it was six years earlier. This interaction led to a memory screen for the patient and the eventual diagnosis of Alzheimer’s disease. It also explained the patient’s poor control of her diabetes: it seems she was not reliably taking her medication nor was she reliably controlling her diet or exercising. Perhaps because the symptoms of early-stage Alzheimer’s disease are subtle and not readily detectable and because patients with early Alzheimer’s e5
e6
Online Only Appendices
BOX B1 BARRIERS TO EARLY DETECTION OF ALZHEIMER’S DISEASE • Patient may be unaware; denies or minimizes symptoms • Evaluation may be time-consuming and not compensated financially • Belief that memory loss may be part of normal aging • Belief that Alzheimer’s disease is not treatable by some health care professionals and the general public Reproduced from Solomon PR, Murphy CA. Geriatrics. 2005 Nov;60(11):26–31.
disease may appear entirely appropriate in the context of a brief office visit (Resnick, 1994; Stoppe et al., 1994; Solomon et al., 2000), cases of Alzheimer’s disease in the early stages often go undetected during a primary care visit. By some estimates, up to 95% of cases of mild dementia are not detected by clinicians (Gifford & Cummings, 1999). In one small study, 78.6% of patients with mild dementia, 71.4% with moderate dementia, and 20% with severe dementia had no indication of cognitive deficits in their medical records (Valcour et al., 2000). Other studies in primary care practices have found rates of dementia between 50% and 66% (Olafsdottir et al., 2000; Valcour et al., 2000). In brief, many cases of early-stage Alzheimer’s disease are missed, and, as the example of our diabetic patient reveals, these patients’ memory problems can have potentially dangerous consequences. Benefits of Medication
As discussed in detail in Chapters 16 and 17, there are currently four approved medications for the treatment of Alzheimer’s disease: donepezil (generic and Aricept), rivastigmine (Exelon), galantamine (now generic, formerly under the names Razadyne and Reminyl), and memantine (generic and Namenda). Each of these medications provides “symptomatic benefit” (i.e., facilitates function of remaining neurons/ synapses), but does not protect against subsequent neuronal loss (i.e., no “disease-modifying” properties). The cholinesterase inhibitors are approved for all stages of Alzheimer’s disease (mild, moderate, and severe) and data suggest that they are most beneficial when administered early in the course of disease (Duncan & Siegal, 1998; Cummings & Jeste, 1999). Memantine is approved for moderate to severe Alzheimer’s disease. The emergence of disease-modifying
treatments further emphasizes the need for early detection and diagnosis. There are currently drugs in clinical trials that are being evaluated for their ability to alter amyloid production or aggregation, and others that may clear brain amyloid (see Chapter 19). Early detection and treatment will be the key to the eventual success of disease-modifying drugs. Implementation of Social Support Systems
We saw a patient in clinic who came accompanied by his wife and his son, who recently returned from Colorado. Our patient, who lived in central Vermont, had for many years run a small chain of sporting goods stores. The business had been successful for many years, but not over the past year. The son, who had worked with his father in the business for several years before going to Colorado, had returned to see if he could help. When he examined the records he was struck to find out that his father had cornered the market on scuba gear. And, while he got a very good price, unfortunately there was not much of a market for scuba gear in central Vermont in January. As his son learned more about the recent activities of the business, he discovered that his father had uncharacteristically made other poor financial judgments that collectively now threatened his business. The cause of the poor judgment turned out to be the onset of Alzheimer’s disease. Implementation of social support systems to address issues of care, nutrition, and safety, as well as financial and legal planning, are most beneficial when initiated early in the course of a progressive disease. For example, patients with Alzheimer’s disease have increased risk for automobile accidents (Cooper et al., 1993) and do not accurately report symptoms of comorbid illnesses (Larson et al., 1984). Early screening and diagnosis of Alzheimer’s disease would allow patients and families to make decisions regarding transportation, living arrangements, and other aspects of care when the patient is functioning at the highest possible level. As in the case of our patient, they allow families to make thoughtful plans for transition of businesses and other financial assets. NOT TO SCREEN … The US Preventive Services Task Force Position
The US Preventive Services Task Force routinely evaluates the risks and benefits of screening for many
B Screening for Memory Loss, Alzheimer’s Disease, and Dementia
diseases and conditions. This group evaluated screening for Alzheimer’s disease in 1996 and 2003. As in the 1996 report, the most recent report (Boustani et al., 2003) did not endorse routine screening for Alzheimer’s disease. Although it raised a number of concerns regarding screening for Alzheimer’s disease, including a low prevalence of the disease, insufficient evidence for the accuracy of screening tests, low accuracy of screening tests for mild dementia, and biases for age, education, and ethnicity, the primary rationale of the Preventive Services Task Force for not endorsing screening is that, until there are studies that demonstrate that screening (and, by implication, earlier diagnosis) provides better outcomes for patients with Alzheimer’s disease, endorsement is premature. Importantly, the Task Force did, however, recognize in 2003 the potential importance of screening to detect dementia at an early stage because only early intervention has the potential to modify an otherwise certain decline. But, rather than endorsing screening, the Task Force endorsed further research in screening for dementia. The Alzheimer’s Association Position
In 2000, the Alzheimer’s Association formed the Work Group on Screening for Cognitive Impairment and Alzheimer’s Disease in an effort to develop guidance for its chapters across the country. The 2000 Working Group listed 21 questions that should be considered before embarking on a public screening program, but it did not endorse community-based screening for Alzheimer’s disease. In the 2005 guidelines to the local chapters produced by the Alzheimer’s Association, the potential benefits of early detection of Alzheimer’s disease were recognized, but again concerns were also raised, including the possible need for informed consent (Irwig & Glasziou, 2000), the challenges of providing adequate follow-up in the case of a positive screen (Lawrence et al., 2003), and the variability in the professional training and ability of those carrying out screening; the concerns of the US Preventive Health Services were also reiterated. The Alzheimer’s Association’s position further distinguishes three levels of screening that are currently operating in the community: level I, memory awareness or pre-screening for memory problems; level II, community assessment of memory problems (using the Mini-Mental State Examination and other such
e7
tests administered in a variety of settings); and level III, comprehensive diagnostic work-up (by a physician). The association states that “there seem to be no major problems with Levels I and III.” It does not, however, endorse level II. In conclusion, the Alzheimer’s Association recommended against community screening by its chapters, but it is fine with physicians screening and diagnosing in their practices. In regard to community screening, the Alzheimer’s Association recommends instead public awareness programs based on “The 10 Warning Signs of Alzheimer’s Disease” as an alternative to most memory screening activities (www.alz.org). It is noteworthy that these symptoms have not been validated and are not being promoted as a screening instrument. Nevertheless, they bear a resemblance to recently developed and validated informant questionnaires (see below) that are now being used to screen. Of some concern is that individuals experiencing cognitive deficits and their families may treat the 10 Warning Signs as a screening instrument, raising exactly the same concerns of psychological distress that the Alzheimer’s Association raises regarding community screening. We have certainly experienced this issue with patients coming to our clinic.
Screening in Primary Care Practice WHO SHOULD BE SCREENED The goal of any screening program is to identify the largest number of individuals who have a particular disease in the most efficient manner. Efficiency includes using instruments that are accurate and brief (see below). One way to increase the efficiency of instruments is to screen individuals who are at risk. In Alzheimer’s disease, the greatest risk factor is age. As such, screening older groups of individuals will have a higher yield. Some practitioners now routinely screen all patients over the age of 65 on an annual basis. The incidence of Alzheimer’s disease in this age group is approximately 10–15%. By increasing this age to 75, the incidence becomes 20%, and it may be as high as 48% in those over 85 (Evans et al., 1989). Other strategies to “enrich” the screening population are to screen those with a family history, or those with cognitive complaints, either by them or by family members.
e8
Online Only Appendices
SELECTING A SCREENING INSTRUMENT Over the past 15 years, there has been a proliferation of screening instruments for Alzheimer’s disease. In general, screening instruments for Alzheimer’s disease fall into two general categories: clinician-administered neuropsychological instruments or questionnaires that are completed by the patient and/or an informant (Jorm, 1997). Clinician-administered instruments vary in scope, but most contain questions regarding orientation for time and place, recent memory, language, and visuospatial function. They may also contain questions to evaluate attention and concentration, executive function, and semantic memory. Clinician-administered instruments far outnumber informant-based screens. Table B1 lists some of the desirable criteria for a clinician-administered instrument to be accurate and useful in the primary care setting. Informant-based questionnaires for Alzheimer’s disease consist of a series of questions that are completed by a knowledgeable informant, typically a spouse, child, or close friend. These questions may inquire about cognition, mood and behavior, and function. Informant-based instruments are becoming more prevalent. This may be because they have a number of advantages over clinician-administered screens, including (1) they require no time from medical professionals to administer and minimal time to score, (2) they do not require the cooperation of
the patient, (3) they can potentially be completed via telephone, mail, or Internet, and (4) they can be completed by the informant confidentially. In summary, a screening instrument with high sensitivity and specificity, requiring minimal time to administer, and that can be administered and scored by a variety of people in the office with minimal training may be appropriate for the office of the primary care provider. ADMINISTERING THE SCREENING INSTRUMENT Once the criteria for which patients to be screened are selected by the provider (e.g., annually for all patients over the age of 65), an individual in the clinic generally assumes responsibility for screening. If the instrument is clinician-based, the individual may also administer and score the screen. If the instrument is informant-based, the individual would assure that the instrument is administered and scored. Armed with the results of the screen prior to meeting with the patient, the physician has the opportunity to discuss the findings with the patient and family during their visit. In this sense, the screen for Alzheimer’s disease is handled much like the results of a screen for cholesterol or blood pressure: the data are collected by office staff and provided to the physician before the meeting with the patient so that they can be considered and subsequently discussed with the patient and family.
TABLE B1 Desirable Characteristics for a Clinician-Administered Screening Instrument Characteristic
Comment
High sensitivity (sensitivity is the percentage of individuals who have the disease who screen positive)
Low sensitivity leads to false negatives. False negatives lead to the failure to diagnose and treat patients who have the disease
High specificity (specificity is the percentage who do not have the disease who screen negative)
Low specificity leads to false positives that will cause people who do not have the disease to undergo time-consuming and expensive evaluations. False positives can also cause unnecessary anxiety in patients and families
Short administration time
Clinician and staff time in primary care practices is limited. Instruments that require longer periods of clinician/staff time are less likely to be used
Training of the screen administrator
Instruments that require minimal training to administer and/or have clear instructions, including training videos, are more likely to be used
Simplicity of rules for scoring
Some screening instruments are scored by clear rules that are unambiguous. Other instruments have questions for which clinicians have difficulty agreeing on the scoring criteria
B Screening for Memory Loss, Alzheimer’s Disease, and Dementia
PROVIDING FEEDBACK TO THE PATIENT/FAMILY The ultimate goal of screening is to identify, diagnose, and treat people with Alzheimer’s disease and other cognitive disorders. Positive screens provide the opportunity for the physician to indicate to the patient that his or her memory should be more fully evaluated. The provider either may conduct this evaluation or may wish to refer to a specialty memory clinic or to a neurologist or psychiatrist. Much like a positive screen for breast or prostate cancer, patients need to be reassured that the positive screen does not mean that they have Alzheimer’s disease, but simply indicates that a more thorough evaluation is warranted. Negative screens provide the opportunity to reassure patients that, although they may be experiencing memory changes, these changes are consistent with the normal, healthy aging process and are not indicative of Alzheimer’s disease or other dementing illness. On occasion, a family member, close friend, or even the patient will note that, despite a negative screen, they continue to be concerned. In these cases, it is our practice that the view of the individual takes precedence over the results of the screen and we recommend a full evaluation.
e9
PROVIDER INVOLVEMENT IN COMMUNITY SCREENING Community-based screening is the point of contention between the Alzheimer’s Association and the groups who advocate widespread screening. Yet community screening may be the ideal way to most efficiently reach the largest group of elders. Over the past 15 years, our clinic has actively organized and participated in community-based screening. While we recognize the concerns of the US Preventive Services Task Force and the Alzheimer’s Association, we also recognize the potential benefits of early detection and diagnosis of Alzheimer’s disease, especially with the potential emergence of disease-modifying drugs. Community-based screening may be most beneficial when conducted in collaboration with either individual or groups of primary care providers. Our clinic has worked with, among others, primary care practices, church groups, councils on aging, and assisted living facilities. It has been our impression that, when conducted properly, community-based screening initiatives can provide a valuable service to large numbers of people, while maintaining scientific accuracy and clinical compassion. Table B2 summarizes the primary
TABLE B2 Concerns and Possible Solutions to Community-Based Screening Criticism
Response
Insufficient accuracy for screening tests
A number of validated tests now show sensitivity and specificity of 90% or more. This is better than screening instruments for many other diseases for which screening is routine
Insufficient accuracy for screening tests for mild disease
A number of validated tests now show sensitivity and specificity of 90% or more for mild disease
Biases for age, education, language, and/or ethnicity
There are now screening instruments that are not significantly affected by age or ethnicity. Many screening instruments take education into account. Several screening tests have been translated and validated in multiple languages
Lack of randomized studies showing the benefits of screening
Randomized studies of screening need to be conducted. The Alzheimer’s Association or the National Institute on Aging should consider sponsoring these studies
Training of screen administrators
The increasing number of screening instruments that require minimal training and clinical judgment to administer and score and the increasing clarity of instructions and inclusion of training videos should minimize these concerns
Follow-up to a positive screen
Follow-up should be conducted by clinicians with sufficient expertise to evaluate and treat Alzheimer’s disease
e10
Online Only Appendices
criticism of community-based screening and suggests strategies to overcome these concerns. In summary, there is now an emerging consensus that the three keys to the eventual successful treatment of Alzheimer’s disease are (1) early detection, (2) using medication to provide symptomatic treatment, and (3) developing medications to slow and eventually halt disease progression. If early detection is the first step in successful treatment and management, criticisms notwithstanding, screening would seem to be an important component. REFERENCES Boustani, M., Peterson, B., Hanson, L., et al., US Preventive Services Task Force, 2003. Screening for dementia in primary care: a summary of the evidence for the US Preventive Services Task Force. Ann. Intern. Med. 138, 927–937. Cooper, P.J., Tallman, K., Tuokko, H., et al., 1993. Vehicle crash involvement and cognitive decline in older adults. J. Safety Res. 24, 9–17. Cummings, J.L., Jeste, D.V., 1999. Alzheimer’s disease and its management in the year 2010. Psychiatr. Serv. 50, 1173–1177. Duncan, B.A., Siegal, A.P., 1998. Early diagnosis and management of Alzheimer’s disease. J. Clin. Psychiatry 59 (Suppl. 9), 15–21. Evans, D.A., Funkenstein, H.H., Albert, M.S., et al., 1989. Prevalence of Alzheimer’s disease in a community population of older persons. Higher than previously reported. JAMA 262, 2551–2556. Gifford, D.R., Cummings, J.L., 1999. Evaluating dementia screening tests: methodologic standards to rate their performance. Neurology 52, 224–227. Irwig, L., Glasziou, P., 2000. Informed consent for screening by community sampling. Eff. Clin. Pract. 3, 47–50.
Jorm, A.F., 1997. Methods of screening for dementia: a metaanalysis of studies comparing an informant questionnaire with a brief cognitive test. Alzheimer Dis. Assoc. Disord. 11 (3), 158–162. Larson, E.B., Reifler, B.V., Featherstone, H.J., et al., 1984. Dementia in elderly outpatients: a prospective study. Ann. Intern. Med. 100, 417–423. Lawrence, J.M., Davidoff, D.A., Katt-Lioyd, D., et al., 2003. Is large-scale community memory screening feasible? Experience from a regional memory-screening day. J. Am. Geriatr. Soc. 51, 1072–1078. New York Times. Available from: Olafsdottir, M., Skoog, I., Marcusson, J., 2000. Detection of dementia in primary care: the Linkoping Study. Dement. Geriatr. Cogn. Disord. 11, 223–229. Resnick, N., 1994. Geriatric medicine and the elderly patient. In: Tierney, L.M., McPhee, S.J., Papadakis, M.A. (Eds.), Current Medical Diagnosis and Treatment, third ed. Appleton and Lange, Norwalk, CT, pp. 557–558. Solomon, P.R., 2008. Should primary care physicians provide dementia screening. J. Clin. Psychiatry 36, 10. Solomon, P.R., Murphy, C.A., 2008. Comment on: Brayne C, Fox C, Boustani M. Dementia screening in primary care: is it time? (2007). JAMA 298 (20), 2409–2411, Alzheimer Research Forum. Available at: (January, 3, 2008.). Solomon, P.R., Brush, M., Calvo, V., et al., 2000. Identifying dementia in the primary care practice. Int. Psychogeriatr. 12, 483–493. Stoppe, G., Sandholzer, H., Staedt, J., et al., 1994. Diagnosis of dementia in primary care: results of a representative survey in lower Saxony, Germany. Eur. Arch. Psychiatry Clin. Neurosci. 244, 278–283. Valcour, V.G., Masaki, K.H., Curb, J.D., et al., 2000. The detection of dementia in the primary care setting. Arch. Intern. Med. 160, 2964–2968.
Appendix B
Screening for Memory Loss, Alzheimer’s Disease, and Dementia Primary care physicians and other healthcare practi-
tioners routinely screen for many diseases, including hypercholesterolemia, hypertension, and prostate and breast cancer, to mention a few. Yet Alzheimer’s disease, one of the most common disorders in older adults, is rarely screened for (Solomon, 2008; Solomon & Murphy, 2008). Interestingly, not only do most patients readily agree to memory screening, but also many are enthusiastic about having their memory treated as seriously as other functions of the body. In a study that we conducted several years ago (Solomon et al., 2000), all patients over the age of 60 visiting a primary care practice over an eight-week period were asked if they were willing to undergo a brief (less than 10 minutes) screen of their memory. Although both our group and the primary care physician who collaborated on the study had some initial concern regarding the patients’ reaction to being approached about the memory screen, we were all relieved when the typical response of patients was being pleased that their physician was concerned about their memory. Of the 145 patients aged over 60 who visited the practice during this period, 139 agreed to be screened. The few who did not agree to being screened at that time typically did so because of time constraints, and most indicated that they would be willing to be screened at a later time. The results of the study indicated that, of the 139 screened, 13 screened positive (requiring follow-up regarding their memory function); this result was about what we would expect given an incidence of dementia of approximately 10% in individuals over the age of 60. Ten of these patients were later diagnosed with Alzheimer’s disease, two with a mixed dementia, and one had mild cognitive impairment. This study showed both that screening in primary care can identify patients with Alzheimer’s disease and that patients are overwhelmingly in favor of having their memory screened.
To Screen or Not to Screen? To an increasing number of clinicians, the argument that we screen our patients for memory disorders seems obvious. But screening for cognitive function is not without its critics. The criticisms come from both practical and theoretical concerns (Box B1). TO SCREEN There are a number of reasons to screen our patients in the context of day-to-day clinical practice. Below we discuss a number of the more important reasons to screen patients for memory loss. Underdiagnosis of Alzheimer’s Disease
One of our colleagues who has a very busy primary care practice tells a story about a 72-year-old woman who came in for a check-up for her type 2 diabetes. For reasons our colleague did not understand, her medications did not seem to be controlling her insulin levels. She reported taking the medications as prescribed, monitoring her diet, and even exercising daily. He was puzzled. At the end of the visit she talked to him about a trip she had taken with her sister last month. She conveyed the details of the trip without difficulty. They then walked out of the examination room to the waiting room together, where they met the patient’s sister. Our colleague mentioned the trip the two sisters had recently taken only to be confronted with a blank look. It turns out that the two sisters had taken the trip, but it was six years earlier. This interaction led to a memory screen for the patient and the eventual diagnosis of Alzheimer’s disease. It also explained the patient’s poor control of her diabetes: it seems she was not reliably taking her medication nor was she reliably controlling her diet or exercising. Perhaps because the symptoms of early-stage Alzheimer’s disease are subtle and not readily detectable and because patients with early Alzheimer’s e5
e6
Online Only Appendices
BOX B1 BARRIERS TO EARLY DETECTION OF ALZHEIMER’S DISEASE • Patient may be unaware; denies or minimizes symptoms • Evaluation may be time-consuming and not compensated financially • Belief that memory loss may be part of normal aging • Belief that Alzheimer’s disease is not treatable by some health care professionals and the general public Reproduced from Solomon PR, Murphy CA. Geriatrics. 2005 Nov;60(11):26–31.
disease may appear entirely appropriate in the context of a brief office visit (Resnick, 1994; Stoppe et al., 1994; Solomon et al., 2000), cases of Alzheimer’s disease in the early stages often go undetected during a primary care visit. By some estimates, up to 95% of cases of mild dementia are not detected by clinicians (Gifford & Cummings, 1999). In one small study, 78.6% of patients with mild dementia, 71.4% with moderate dementia, and 20% with severe dementia had no indication of cognitive deficits in their medical records (Valcour et al., 2000). Other studies in primary care practices have found rates of dementia between 50% and 66% (Olafsdottir et al., 2000; Valcour et al., 2000). In brief, many cases of early-stage Alzheimer’s disease are missed, and, as the example of our diabetic patient reveals, these patients’ memory problems can have potentially dangerous consequences. Benefits of Medication
As discussed in detail in Chapters 16 and 17, there are currently four approved medications for the treatment of Alzheimer’s disease: donepezil (generic and Aricept), rivastigmine (Exelon), galantamine (now generic, formerly under the names Razadyne and Reminyl), and memantine (generic and Namenda). Each of these medications provides “symptomatic benefit” (i.e., facilitates function of remaining neurons/ synapses), but does not protect against subsequent neuronal loss (i.e., no “disease-modifying” properties). The cholinesterase inhibitors are approved for all stages of Alzheimer’s disease (mild, moderate, and severe) and data suggest that they are most beneficial when administered early in the course of disease (Duncan & Siegal, 1998; Cummings & Jeste, 1999). Memantine is approved for moderate to severe Alzheimer’s disease. The emergence of disease-modifying
treatments further emphasizes the need for early detection and diagnosis. There are currently drugs in clinical trials that are being evaluated for their ability to alter amyloid production or aggregation, and others that may clear brain amyloid (see Chapter 19). Early detection and treatment will be the key to the eventual success of disease-modifying drugs. Implementation of Social Support Systems
We saw a patient in clinic who came accompanied by his wife and his son, who recently returned from Colorado. Our patient, who lived in central Vermont, had for many years run a small chain of sporting goods stores. The business had been successful for many years, but not over the past year. The son, who had worked with his father in the business for several years before going to Colorado, had returned to see if he could help. When he examined the records he was struck to find out that his father had cornered the market on scuba gear. And, while he got a very good price, unfortunately there was not much of a market for scuba gear in central Vermont in January. As his son learned more about the recent activities of the business, he discovered that his father had uncharacteristically made other poor financial judgments that collectively now threatened his business. The cause of the poor judgment turned out to be the onset of Alzheimer’s disease. Implementation of social support systems to address issues of care, nutrition, and safety, as well as financial and legal planning, are most beneficial when initiated early in the course of a progressive disease. For example, patients with Alzheimer’s disease have increased risk for automobile accidents (Cooper et al., 1993) and do not accurately report symptoms of comorbid illnesses (Larson et al., 1984). Early screening and diagnosis of Alzheimer’s disease would allow patients and families to make decisions regarding transportation, living arrangements, and other aspects of care when the patient is functioning at the highest possible level. As in the case of our patient, they allow families to make thoughtful plans for transition of businesses and other financial assets. NOT TO SCREEN … The US Preventive Services Task Force Position
The US Preventive Services Task Force routinely evaluates the risks and benefits of screening for many
B Screening for Memory Loss, Alzheimer’s Disease, and Dementia
diseases and conditions. This group evaluated screening for Alzheimer’s disease in 1996 and 2003. As in the 1996 report, the most recent report (Boustani et al., 2003) did not endorse routine screening for Alzheimer’s disease. Although it raised a number of concerns regarding screening for Alzheimer’s disease, including a low prevalence of the disease, insufficient evidence for the accuracy of screening tests, low accuracy of screening tests for mild dementia, and biases for age, education, and ethnicity, the primary rationale of the Preventive Services Task Force for not endorsing screening is that, until there are studies that demonstrate that screening (and, by implication, earlier diagnosis) provides better outcomes for patients with Alzheimer’s disease, endorsement is premature. Importantly, the Task Force did, however, recognize in 2003 the potential importance of screening to detect dementia at an early stage because only early intervention has the potential to modify an otherwise certain decline. But, rather than endorsing screening, the Task Force endorsed further research in screening for dementia. The Alzheimer’s Association Position
In 2000, the Alzheimer’s Association formed the Work Group on Screening for Cognitive Impairment and Alzheimer’s Disease in an effort to develop guidance for its chapters across the country. The 2000 Working Group listed 21 questions that should be considered before embarking on a public screening program, but it did not endorse community-based screening for Alzheimer’s disease. In the 2005 guidelines to the local chapters produced by the Alzheimer’s Association, the potential benefits of early detection of Alzheimer’s disease were recognized, but again concerns were also raised, including the possible need for informed consent (Irwig & Glasziou, 2000), the challenges of providing adequate follow-up in the case of a positive screen (Lawrence et al., 2003), and the variability in the professional training and ability of those carrying out screening; the concerns of the US Preventive Health Services were also reiterated. The Alzheimer’s Association’s position further distinguishes three levels of screening that are currently operating in the community: level I, memory awareness or pre-screening for memory problems; level II, community assessment of memory problems (using the Mini-Mental State Examination and other such
e7
tests administered in a variety of settings); and level III, comprehensive diagnostic work-up (by a physician). The association states that “there seem to be no major problems with Levels I and III.” It does not, however, endorse level II. In conclusion, the Alzheimer’s Association recommended against community screening by its chapters, but it is fine with physicians screening and diagnosing in their practices. In regard to community screening, the Alzheimer’s Association recommends instead public awareness programs based on “The 10 Warning Signs of Alzheimer’s Disease” as an alternative to most memory screening activities (www.alz.org). It is noteworthy that these symptoms have not been validated and are not being promoted as a screening instrument. Nevertheless, they bear a resemblance to recently developed and validated informant questionnaires (see below) that are now being used to screen. Of some concern is that individuals experiencing cognitive deficits and their families may treat the 10 Warning Signs as a screening instrument, raising exactly the same concerns of psychological distress that the Alzheimer’s Association raises regarding community screening. We have certainly experienced this issue with patients coming to our clinic.
Screening in Primary Care Practice WHO SHOULD BE SCREENED The goal of any screening program is to identify the largest number of individuals who have a particular disease in the most efficient manner. Efficiency includes using instruments that are accurate and brief (see below). One way to increase the efficiency of instruments is to screen individuals who are at risk. In Alzheimer’s disease, the greatest risk factor is age. As such, screening older groups of individuals will have a higher yield. Some practitioners now routinely screen all patients over the age of 65 on an annual basis. The incidence of Alzheimer’s disease in this age group is approximately 10–15%. By increasing this age to 75, the incidence becomes 20%, and it may be as high as 48% in those over 85 (Evans et al., 1989). Other strategies to “enrich” the screening population are to screen those with a family history, or those with cognitive complaints, either by them or by family members.
e8
Online Only Appendices
SELECTING A SCREENING INSTRUMENT Over the past 15 years, there has been a proliferation of screening instruments for Alzheimer’s disease. In general, screening instruments for Alzheimer’s disease fall into two general categories: clinician-administered neuropsychological instruments or questionnaires that are completed by the patient and/or an informant (Jorm, 1997). Clinician-administered instruments vary in scope, but most contain questions regarding orientation for time and place, recent memory, language, and visuospatial function. They may also contain questions to evaluate attention and concentration, executive function, and semantic memory. Clinician-administered instruments far outnumber informant-based screens. Table B1 lists some of the desirable criteria for a clinician-administered instrument to be accurate and useful in the primary care setting. Informant-based questionnaires for Alzheimer’s disease consist of a series of questions that are completed by a knowledgeable informant, typically a spouse, child, or close friend. These questions may inquire about cognition, mood and behavior, and function. Informant-based instruments are becoming more prevalent. This may be because they have a number of advantages over clinician-administered screens, including (1) they require no time from medical professionals to administer and minimal time to score, (2) they do not require the cooperation of
the patient, (3) they can potentially be completed via telephone, mail, or Internet, and (4) they can be completed by the informant confidentially. In summary, a screening instrument with high sensitivity and specificity, requiring minimal time to administer, and that can be administered and scored by a variety of people in the office with minimal training may be appropriate for the office of the primary care provider. ADMINISTERING THE SCREENING INSTRUMENT Once the criteria for which patients to be screened are selected by the provider (e.g., annually for all patients over the age of 65), an individual in the clinic generally assumes responsibility for screening. If the instrument is clinician-based, the individual may also administer and score the screen. If the instrument is informant-based, the individual would assure that the instrument is administered and scored. Armed with the results of the screen prior to meeting with the patient, the physician has the opportunity to discuss the findings with the patient and family during their visit. In this sense, the screen for Alzheimer’s disease is handled much like the results of a screen for cholesterol or blood pressure: the data are collected by office staff and provided to the physician before the meeting with the patient so that they can be considered and subsequently discussed with the patient and family.
TABLE B1 Desirable Characteristics for a Clinician-Administered Screening Instrument Characteristic
Comment
High sensitivity (sensitivity is the percentage of individuals who have the disease who screen positive)
Low sensitivity leads to false negatives. False negatives lead to the failure to diagnose and treat patients who have the disease
High specificity (specificity is the percentage who do not have the disease who screen negative)
Low specificity leads to false positives that will cause people who do not have the disease to undergo time-consuming and expensive evaluations. False positives can also cause unnecessary anxiety in patients and families
Short administration time
Clinician and staff time in primary care practices is limited. Instruments that require longer periods of clinician/staff time are less likely to be used
Training of the screen administrator
Instruments that require minimal training to administer and/or have clear instructions, including training videos, are more likely to be used
Simplicity of rules for scoring
Some screening instruments are scored by clear rules that are unambiguous. Other instruments have questions for which clinicians have difficulty agreeing on the scoring criteria
B Screening for Memory Loss, Alzheimer’s Disease, and Dementia
PROVIDING FEEDBACK TO THE PATIENT/FAMILY The ultimate goal of screening is to identify, diagnose, and treat people with Alzheimer’s disease and other cognitive disorders. Positive screens provide the opportunity for the physician to indicate to the patient that his or her memory should be more fully evaluated. The provider either may conduct this evaluation or may wish to refer to a specialty memory clinic or to a neurologist or psychiatrist. Much like a positive screen for breast or prostate cancer, patients need to be reassured that the positive screen does not mean that they have Alzheimer’s disease, but simply indicates that a more thorough evaluation is warranted. Negative screens provide the opportunity to reassure patients that, although they may be experiencing memory changes, these changes are consistent with the normal, healthy aging process and are not indicative of Alzheimer’s disease or other dementing illness. On occasion, a family member, close friend, or even the patient will note that, despite a negative screen, they continue to be concerned. In these cases, it is our practice that the view of the individual takes precedence over the results of the screen and we recommend a full evaluation.
e9
PROVIDER INVOLVEMENT IN COMMUNITY SCREENING Community-based screening is the point of contention between the Alzheimer’s Association and the groups who advocate widespread screening. Yet community screening may be the ideal way to most efficiently reach the largest group of elders. Over the past 15 years, our clinic has actively organized and participated in community-based screening. While we recognize the concerns of the US Preventive Services Task Force and the Alzheimer’s Association, we also recognize the potential benefits of early detection and diagnosis of Alzheimer’s disease, especially with the potential emergence of disease-modifying drugs. Community-based screening may be most beneficial when conducted in collaboration with either individual or groups of primary care providers. Our clinic has worked with, among others, primary care practices, church groups, councils on aging, and assisted living facilities. It has been our impression that, when conducted properly, community-based screening initiatives can provide a valuable service to large numbers of people, while maintaining scientific accuracy and clinical compassion. Table B2 summarizes the primary
TABLE B2 Concerns and Possible Solutions to Community-Based Screening Criticism
Response
Insufficient accuracy for screening tests
A number of validated tests now show sensitivity and specificity of 90% or more. This is better than screening instruments for many other diseases for which screening is routine
Insufficient accuracy for screening tests for mild disease
A number of validated tests now show sensitivity and specificity of 90% or more for mild disease
Biases for age, education, language, and/or ethnicity
There are now screening instruments that are not significantly affected by age or ethnicity. Many screening instruments take education into account. Several screening tests have been translated and validated in multiple languages
Lack of randomized studies showing the benefits of screening
Randomized studies of screening need to be conducted. The Alzheimer’s Association or the National Institute on Aging should consider sponsoring these studies
Training of screen administrators
The increasing number of screening instruments that require minimal training and clinical judgment to administer and score and the increasing clarity of instructions and inclusion of training videos should minimize these concerns
Follow-up to a positive screen
Follow-up should be conducted by clinicians with sufficient expertise to evaluate and treat Alzheimer’s disease
e10
Online Only Appendices
criticism of community-based screening and suggests strategies to overcome these concerns. In summary, there is now an emerging consensus that the three keys to the eventual successful treatment of Alzheimer’s disease are (1) early detection, (2) using medication to provide symptomatic treatment, and (3) developing medications to slow and eventually halt disease progression. If early detection is the first step in successful treatment and management, criticisms notwithstanding, screening would seem to be an important component. REFERENCES Boustani, M., Peterson, B., Hanson, L., et al., US Preventive Services Task Force, 2003. Screening for dementia in primary care: a summary of the evidence for the US Preventive Services Task Force. Ann. Intern. Med. 138, 927–937. Cooper, P.J., Tallman, K., Tuokko, H., et al., 1993. Vehicle crash involvement and cognitive decline in older adults. J. Safety Res. 24, 9–17. Cummings, J.L., Jeste, D.V., 1999. Alzheimer’s disease and its management in the year 2010. Psychiatr. Serv. 50, 1173–1177. Duncan, B.A., Siegal, A.P., 1998. Early diagnosis and management of Alzheimer’s disease. J. Clin. Psychiatry 59 (Suppl. 9), 15–21. Evans, D.A., Funkenstein, H.H., Albert, M.S., et al., 1989. Prevalence of Alzheimer’s disease in a community population of older persons. Higher than previously reported. JAMA 262, 2551–2556. Gifford, D.R., Cummings, J.L., 1999. Evaluating dementia screening tests: methodologic standards to rate their performance. Neurology 52, 224–227. Irwig, L., Glasziou, P., 2000. Informed consent for screening by community sampling. Eff. Clin. Pract. 3, 47–50.
Jorm, A.F., 1997. Methods of screening for dementia: a metaanalysis of studies comparing an informant questionnaire with a brief cognitive test. Alzheimer Dis. Assoc. Disord. 11 (3), 158–162. Larson, E.B., Reifler, B.V., Featherstone, H.J., et al., 1984. Dementia in elderly outpatients: a prospective study. Ann. Intern. Med. 100, 417–423. Lawrence, J.M., Davidoff, D.A., Katt-Lioyd, D., et al., 2003. Is large-scale community memory screening feasible? Experience from a regional memory-screening day. J. Am. Geriatr. Soc. 51, 1072–1078. New York Times. Available from: