Should Older Adults be Screened for Dementia



March 12, 2007 - 4:30pm PDT

SHOULD OLDER ADULTS BE SCREENED FOR DEMENTIA?

SCREENING FOR EVIDENCE OF DEMENTIA IS IMPORTANT!

J. Wesson Ashford - Stanford / VA Alzheimer Center

Soo Borson - University of Washington

Ruth O’Hara - Stanford University

Paul Dash - Johns Hopkins University

Lori Frank - Center for Health Outcomes Research, United BioSource Corporation

Philippe Robert - University of Nice

William R. Shankle, Cognitive Sciences, U. C. Irvine

Mary C. Tierney - Sunnybrook and Women's College Health Sciences Centre, University of Toronto

Henry Brodaty - University of New South Wales

Frederick A. Schmitt - University of Kentucky

Helena C. Kraemer - Stanford University

Herman Buschke - Albert Einstein College of Medicine

E-mail addresses

washford@; soob@u.washington.edu; roh@stanford.edu; DrDneuro@; frank@; Lori.Frank@; philippe.robert15@wanadoo.fr; rshankle@; Mary.Tierney@sw.ca; h.brodaty@unsw.edu.au; fascom@email.uky.edu; hck@stanford.edu; buschke@aecom.yu.edu; Ashford@stanford.edu

Corresponding author:

J. Wesson Ashford, M.D., Ph.D.

Senior Research Scientist, Stanford / VA Alzheimer’s Center

Palo Alto VA Hospital, 151-Y

3801 Miranda Ave.

Palo Alto, CA 94304-1207

Tel: (cell) (415) 722-0187 (office) (650) 852-3287 (fax) (650) 852-3297

Acknowledgements:

Drs. Ashford, O’Hara, and Kraemer are, supported by grant P30 AG 17824 from the National Institutes of Health, by the Medical Research Service of the Veterans Affairs Palo Alto Health Care System, and by the Department of Veterans Affairs Sierra-Pacific Mental Illness Research, Education, and Clinical Center (MIRECC). Dr. O’Hara is also supported by NIH grants AG18784 and MH070886.

Dr. Borson is supported by National Institute on Aging Grants P50 AG05136 and R01 AG025515.

Dr. Buschke is supported by National Institute on Aging Grant AG03949, NICHD grant HD-01799

Dr. Schmitt is supported by National Institute on Aging grants P50 AG05144 and PO1 AG19241

Dr. Shankle is a Research Fellow, Cognitive Sciences, U. C. Irvine

Disclosures: All of the authors have been associated with the development of screening tools for dementia and Alzheimer’s disease

ABSTRACT

Ashford et al. (2006) presented numerous arguments for considering routine dementia screening. At this time, dementia diagnoses are widely unrecognized, and dementia patients are missing important clinical care and treatment interventions, clearly indicating the need for dementia screening. By distinction, the problems of defining, diagnosing, and treating Mild Cognitive Impairment (MCI) are not yet resolved, and MCI is not ready for screening consideration. Dementia screening approaches, including cognitive testing and functional assessment, must be evaluated on their scientific merits, including sensitivity and specificity for recognizing affected individuals in at-risk populations. Screening tests must be “cost-worthy”, with the benefits of true-positive test results justifying the costs of testing and resolving false-positive cases, with due consideration for proper diagnostic evaluation and potential harms. With the tremendous number of new cases projected and the expected emergence beneficial therapies, considerably more research is needed to develop more efficient screening systems for the near future.

Key words: dementia, screening, Alzheimer, diagnosis, mild cognitive impairment, MCI, cost-benefit analysis, harms, benefits, memory, cognition

A. The clinical evidence justifies screening for dementia

Ashford et al. (Alzheimer’s &Dementia Perspective, 2006) presented the consensus of a group of clinicians and scientists that screening at risk populations for evidence of dementia was an important matter to consider (A&D Consensus Group). The A&D Consensus Group addressed the severe, well-documented, and widely recognized problem of inadequate recognition of dementia in clinical practice (Callahan et al., 1995; Ross et al. 1997; Sternberg et al., 2000; Valcour et al., 2000; Finkel, 2003). Freund (2006) estimated that the missed diagnoses are over 25% of the dementia cases and may be as high as 90%. Dementia exerts a substantial burden on patients’ lives and the lives of those close to them (Frank et al., 2006). The A&D Consensus Group reviewed the responses of numerous national and international organizations to this worsening crisis and noted that none recommended screening for dementia, though essentially all of the reviewed organizations did recommend a diagnostic evaluation when memory problems or dementia was suspected. It is commonly accepted that most dementia patients are cared for in the primary care setting, and clinicians working in this setting do not have adequate time for in depth consideration of unrecognized cognitive difficulties that their patients might have. Further, there are a variety of reasons that neither the clinicians, nor the patients, nor those close to the patients express concerns about the presence of dementia when it is first noticed. Multi-year delays from first symptom occurrence to clinical assessment have been documented and attributed mostly to uncertainty about the severity of the cognitive deficit (47%) and thinking that the observed changes were normal aging (37%) (Knopman et al., 2000 JAGS 48:300).

Accordingly, the A&D Consensus Group recommended that the process for suspecting and recognizing possible early dementia be operationalized. The process required to detect unrecognized or unacknowledged disease is commonly referred to as “screening”. Given the abundance of adequate tests for recognizing mild dementia, the numerous benefits in doing so, the slight costs associated with such testing, and the minimal nature of the potential harms from such investigation (see Table), this group recommended the consideration of implementation of systems to screen for dementia. The perspective that it is reasonable to recommend screening for dementia has only recently developed and been championed independently by other groups (Solomon & Murphy, 2005; Fillit et al., 2006; Chopra et al., 2007).

B. Defining the dementia-related conditions for screening consideration

There is a long-term problem in the field of dementia of defining the basic diagnostic issues and symptom constellations. The American Psychiatric Association (APA), over the prolonged period of development of the Diagnostic Manual, version III and IV (DSM-III; DSM-IV), has established a diagnostic spectrum of dementia, including “Dementia of the Alzheimer Type”, “Vascular Dementia”, and “Dementia Due to Other General Medical Conditions” (APA, 1994). The core description of dementia includes development of multiple cognitive deficits, including memory and other disturbances, causing impairment in social or occupational functioning. Generally, dementia does not have a defined onset, and the rate of progression varies extensively. The dementia course may be “characterized by gradual onset” that progresses insidiously, as is typical with Alzheimer’s disease, or it may begin suddenly and progress in discrete increments, as may be seen with vascular dementia. The uncertainty of the point of dementia onset is a basic reason that a screening system is needed; with a variable course, early dementia is difficult for clinicians to notice. There are now many widely acceptable management interventions that are not properly applied because the presence of the disease is missed. It is because of the difficulty in recognizing this problem that many well-meaning scientists and clinicians have sought to develop screening tests for this difficult problem.

Recently, there have been increasing discussions and extensive considerations of what follows normal function but precedes dementia, a concept now widely referred to as Mild Cognitive Impairment (MCI, Petersen, 2004). While MCI has received a considerable amount of research attention, it has not been formally defined as a diagnostic entity for routine clinical purposes, and screening tests have not yet been developed to recognize this problem (Schmitt et al., 2006).

C. Reviewing the screening principles

Since dementia has been such a difficult syndrome to recognize, it is important to use the best available screening principles to decide how to evaluate a subject for this problem. It is the contention of the A&D Consensus Group that all of the criteria for conducting a screen for dementia (as opposed to MCI) are met. To briefly review those principles (as also noted by Dale & Sachs) for dementia:

1) “it must be common” - dementia is admittedly very common, but it must also be noted that the prevalence increases steeply with age (more so than mortality, Raber et al., 2004). As yearly incidence increases with age, the imperative to screen increases proportionally. Depending on ancillary considerations, the threshold for recommending routine screening to the population may be reached by age 75 years.

2) “It must have sensitive and specific tests available for its detection” - there are an abundance of tests available for dementia screening whose sensitivities and specificities both exceed 0.8, and several exceed 0.9 (e.g., Solomon & Murphy, 2005; Schmitt et al., 2006).

3) “It must have efficacious treatment” - there are 5 FDA approved medications for Alzheimer’s disease as well as highly-recommended treatments for several other types of dementia. There are a few groups that have questioned the efficacy of the cholinesterase inhibitors for Alzheimer treatment (the first was Ashford et al., 1981; the most prominent has been the statement of the National Institute for Health and Clinical Excellence (NICE) in Great Britain in response to data from Courtney et al., 2004; the NICE appraisal was revised 11/2006 to include a recommendation for use of cholinesterase inhibitors for moderately demented patients with Alzheimer’s disease; NICE, 2006). However, the preponderance of the studies have shown that the approved dementia medications have substantive and prolonged benefit (see Tinklenberg et al., in press; Fillet et al., 2006; Geldmacher et al., 2006). Beyond specific treatments for Alzheimer’s disease, there are efficacious biological, psychological, and social interventions that should be made as soon as possible in the disease course for Alzheimer’s disease and the numerous other types of dementia associated with other etiologies.

4) “If treatment exists, treated patients must have better outcomes than untreated patients”

As noted for #3, there has been some debate about this point, but many studies have shown the benefits of the treatments, both biological (brain scans), psychological (cognitive testing and behavioral testing), and social (ADL measurements) (see Fillit et al., 2006; Geldmacher et al., 2006 for reviews). Further, there are many types of dementia beyond Alzheimer’s disease where a specific early intervention is definitely warranted. Additionally, it is unclear why there is any argument to the preponderance of recommendations to provide general supportive care to dementia patients and their families at the earliest time possible.

5) “The benefits from screening must outweigh the harms” -

There are numerous benefits and relatively minor harms associated with screening for dementia (see Table). The issue of harmful side-effects from treatment is not directly related to screening. Rather, the cost-benefit of implementing treatments is a decision made based on the diagnostic evaluation, independent of the rationale for initiating the diagnostic evaluation.

For Alzheimer’s disease, the side-effects from the available treatments are usually of minimal clinical significance. The concern that there may be an increase of mortality with the use of cholinesterase inhibitor treatment in patients with MCI is not directly germane to the issue of treating patients with diagnosed dementia of the Alzheimer type. However, the question of harms of treatment does introduce the issue that screening tests need to be followed by appropriate diagnostic evaluation before treatments are initiated. Cholinesterase inhibitors even have some side-effects that are useful, including less constipation, slower heart rate and improved behavior.

D. The need to distinguish the concepts of dementia and MCI

An important distinction must be drawn, which is confused in the paper by Dale & Sachs, between dementia and MCI. MCI is a prodromal condition to dementia in many cases, but it is not a diagnosis of dementia. Screening for dementia is addressing a problem that has already been recognized by numerous official entities as requiring a diagnosis. In spite of popular interest in screening for MCI (Dale et al., 2006) and wide-spread concern about MCI in primary care practice, diagnostic criteria for MCI still need clarification, as was noted in the original A&D Consensus Group paper. Further research is needed on screening methodology and treatment benefits before population screening for MCI will be ready for consideration. Several of the comments from the Dale & Sachs discussion apply to the concern about MCI screening, not to dementia screening as discussed in the A&D Consensus Group paper.

E. Delineation of screening-related risks from adverse consequences of diagnosis

It is important to distinguish the risks involved in a screening test from the results that may occur as the direct outcome of clinical care, including potential adverse consequences of diagnostic tests and treatments. In the “Guidelines and Recommendations about Screening” (Barratt et al., 1999, referenced by Dale & Sachs), there are diagrams describing systems to implement screening tests – this paper does not apply to dementia screening as discussed by Ashford et al. because it presents a direct advance from screenings tests to treatment. What is discussed in the Ashford et al. paper is screening tests to determine when diagnostic tests should be considered. It is an adverse development if the results of any of the dementia screening tests are interpreted wrongly as a diagnosis of “Alzheimer’s disease” without proceeding through the “Standard of Care” diagnostic procedures for dementia diagnosis. Further, the routine dementia diagnostic tests, including progressively: history and physical examination, blood tests, focused neuropsychological assessment, and a brain scan, are generally safe. However, it is reductionistic to apply indiscriminately the numerous potential adverse consequences of complex clinical interventions to a brief check for early dementia signs.

After clarifying the separation of screening and diagnostic procedures, the correct point about screening tests needs to be reiterated, “The only negative impact of a false positive could be at most a brief secondary assessment”! In this circumstance, a screening test only leads to a recommendation of a second step in assessment. Such a test should be considered to be of no greater consequence than the commonly used screening question, “Do you have a cough”, for which the positive response should lead to auscultation of the lungs, not diagnosis and treatment of lung cancer. Thus, a dementia screening test itself should carry no significant social, psychological, or ethical concerns.

There is a related issue expressed by Dale & Sachs that a screening test can result in “labeling”. Labeling cannot occur without a diagnosis. Since screening does not provide a diagnosis, there is no reason for labeling to occur. There have been appropriate concerns raised about psychological and social consequences of making a diagnosis of Alzheimer’s disease. This concern is appropriate and needs to be addressed as part of the refinement of dementia diagnosis. However, this is not a problem that should be related to appropriate screening.

Another concern raised by Dale & Sachs is that individuals participating in a screening test would make inappropriate decisions about the recommendations resulting from the testing. Failure of a subject with a positive screen to get further diagnostic assessment is a concern, but compliance with medical recommendations is a widespread problem, not just related to screening tests. Compliance problems represent one of the points where modern medicine needs broad based improvement that is difficult to address within an over-burdened health care system. The opposite problem, that an individual with a negative screen may see this result as permanent freedom from worry about dementia, is also a concern and a misunderstanding. The negative screening result only suggests that the concern about dementia can be reduced for a limited period of time, for example, one year. Such inappropriate patient responses to screening test results should not be consider harms of screening but areas for attention where the quality of the whole screening system could be progressively improved and population education can lead to an overall improvement in the health of society.

F. Financial costs associated with screening

Dale & Sachs note that the monetary costs incurred by screening and the resulting increase of care burden. Analyses of this issue should formally address the “cost-worthiness” of the medical test (Kraemer, 1992). Such analyses have been reported and published (see Fillit et al., 2006), and looking at limited health economic data, the results tend to support screening. However, the A&D Consensus Group further argued that the numerous benefits from having information about mental dysfunction can help the patient’s support system function more efficiently and with less stress and plan for more effective management of the issues that will develop as the patient deteriorates.

CONCLUSIONS:

1) In the development of cognitive and memory tests, a clear distinction must be drawn between screening for dementia and screening for MCI. Tests are currently available that should be considered “cost-worthy” for screening for dementia. The bases for screening for MCI are not yet established.

2) The decision to screen for dementia should be based on sound scientific consideration of all relevant issues and public health benefits, not political manipulations. Further attention to dementia screening is clearly warranted, though implementation will require careful development of practice guidelines and may vary considerably across different settings.

3) Dementia screening in clinical settings is clearly appropriate for those at highest risk at this time, for example, over age 75 years. Widespread screening of the whole elderly population also has merit, though systematic recommendations need to be developed. Full public health screening will become justifiable when more substantive therapeutic and/or preventive interventions are developed, and such therapies are under intense testing. Consequently, now is the time to prepare for the future by developing dementia screening systems and memory testing programs that will be able to detect patients with early phases of dementia.

ADDENDUM

Motivations of professionals in the field of dementia screening

Dale & Sachs raise concerns about the whether there are irreconcilable conflicts of interest among those in the field of Alzheimer’s disease. There are financial incentives that operate in all human beings. Financial motivations are most clear for the Pharmaceutical Industry. However, as was clear from problems that certain pharmaceutical companies have had recently, attempts to obtain results falsely lead to the risk of financial ruin. More often, and more profitably, pharmaceutical companies do their best to follow all of the rules very carefully, though at least in part because of the careful scrutiny. Further, even the peer reviewers at the NIH have their own issues and biases, and they are working in a political arena which also is fraught with personal ambitions. The co-authors of the original paper are admittedly those interested in screening test development. That interest has been developed in all of the co-authors because of their interest in providing optimum care for patients. It is likely true that we all have had dreams of professional recognition for our work, but there is no evidence that there is substantial financial reward for cognitive test development or that this is a significant motivation for any of the original authors.

NOTES about Dale and Sachs potential conflicts of interests (stated to be none).

Dale & Sachs have recently published a paper specifically on the popular interest in screening for mild cognitive impairment (Dale et al., 2006). They clearly have revealed a conflict in their interests in the focus of the comment paper on screening for MCI. Addressing MCI screening was clearly not the intent of the Ashford et al. paper.

It should also be noted that Dr. Dale has received funding for his research from various organizations, some of which are opposed to screening. Dr. Sachs has been on the Ethics Panel for the Alzheimer’s Association and has had his research project, “Palliative Care Project, PACE” funded by the Alzheimer’s Association. The Alzheimer’s Association has taken a position against screening for Alzheimer’s disease. Dr. Dale’s and Dr. Sach’s involvement with this organization and their support for the position of the recommendation of that organization represents a conflict of interest which is no less than any of the authors of the A&D Consensus Group.

REFERENCES

American Psychiatric Association. Diagnostic and Statistical Manual, IV.1994.

J. Wesson Ashford, Soo Borson, Ruth O’Hara, Paul Dash, Lori Frank, Philippe Robert, William R. Shankle, Mary C. Tierney, Henry Brodaty, Frederick A. Schmitt, Helena C. Kraemer, Herman Buschke. Should older adults be screened for dementia? Alzheimer’s & Dementia 2 (2006) 76–85.

Ashford JW, Soldinger S, Schaeffer J, Cochran L, Jarvik LF. Physostigmine and its effect on six patients with dementia. Am J Psychiatry. 1981 Jun;138(6):829-30.

Barratt A, Irwig L, Glasziou P, Cumming RG, Raffle A, Hicks N, Gray JA, Guyatt GH. Users' guides to the medical literature: XVII. How to use guidelines and recommendations about screening. Evidence-Based Medicine Working Group. JAMA. 1999 Jun 2;281(21):2029-34.

Callahan CM, Hendrie HC, Tierney WM. Documentation and evaluation of cognitive impairment in elderly primary care patients. Ann Intern Med. 1995 Mar 15;122(6):422-9.

Chopra A, Cavalieri TA, Libon DJ. Dementia screening tools for the primary care physician. Clinical Geriatrics. 2007;15:38-45.

Courtney C, Farrell D, Gray R, Hills R, Lynch L, Sellwood E, Edwards S, Hardyman W, Raftery J, Crome P, Lendon C, Shaw H, Bentham P; AD2000 Collaborative Group. Long-term donepezil treatment in 565 patients with Alzheimer's disease (AD2000): randomised double-blind trial. Lancet. 2004 Jun 26;363(9427):2105-15.

Dale W, Hougham GW, Hill EK, Sachs GA. High interest in screening and treatment for mild cognitive impairment in older adults: A pilot study. J Am Geriatr Soc. 2006 Sep;54(9):1388-94.

Fillit HM, Smith Doody R, Binaso K, Crooks GM, Ferris SH, Farlow MR, Leifer B, Mills C, Minkoff N, Orland B, Reichman WE, Salloway S. Recommendations for best practices in the treatment of Alzheimer's disease in managed care. Am J Geriatr Pharmacother. 2006;4 Suppl 1:S9-S24.

Finkel SI. Cognitive screening in the primary care setting. The role of physicians at the first point of entry. Geriatrics. 2003 Jun;58(6):43-4.

Frank L, Lloyd A, Flynn JA, Kleinman L, Matza LS, Margolis MK, Bowman L, Bullock R. Impact of cognitive impairment on mild dementia patients and mild cognitive impairment patients and their informants. Int Psychogeriatr. 2006 Mar;18(1):151-62.

Freund B. Office-based evaluation of the older driver. J Am Geriatr Soc. 2006 Dec;54(12):1943-4.

Geldmacher DS, Frolich L, Doody RS, Erkinjuntti T, Vellas B, Jones RW, Banerjee S, Lin P, Sano M. Realistic expectations for treatment success in Alzheimer's disease. J Nutr Health Aging. 2006 Sep-Oct;10(5):417-29.

Knopman D, Donohue JA, Gutterman EM. Patterns of care in the early stages of Alzheimer's disease: impediments to timely diagnosis. J Am Geriatr Soc. 2000 Mar;48(3):300-4.

Molnar FJ, Patel A, Marshall SC, Man-Son-Hing M, Wilson KG. Clinical utility of office-based cognitive predictors of fitness to drive in persons with dementia: A systematic review. J Am Geriatr Soc. 2006 Dec;54(12):1809-24.

National Institute for Health and Clinical Excellence (NICE). NICE technology appraisal guidance 111. Donepezil, galantamine, rivastigmine (review) and memantine for the treatment of Alzheimer’s disease, Includes a review of NICE technology appraisal guidance 19. November 2006

Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med. 2004 Sep;256(3):183-94.

Raber J, Huang Y, Ashford JW. ApoE genotype accounts for the vast majority of AD risk and AD pathology. Neurobiol Aging. 2004 May-Jun;25(5):641-50.

Ross GW, Abbott RD, Petrovitch H, Masaki KH, Murdaugh C, Trockman C, Curb JD, White LR. Frequency and characteristics of silent dementia among elderly Japanese-American men. The Honolulu-Asia Aging Study. JAMA. 1997 Mar 12;277(10):800-5.

Schmitt FA, Mendiondo MS, Kryscio RJ, Ashford JW. A brief Alzheimer’s screen for clinical practice. Res Pract in Alzeimer’s dis. 2006;11:1-4.

Sternberg SA, Wolfson C, Baumgarten M. Undetected dementia in community-dwelling older people: the Canadian Study of Health and Aging. J Am Geriatr Soc. 2000 Nov;48(11):1430-4.

Valcour VG, Masaki KH, Curb JD, Blanchette PL. The detection of dementia in the primary care setting. Arch Intern Med 2000;160:2964–8.

TABLE - BENEFITS and HARMS

BENEFITS

Psychological and social benefits from early dementia recognition

- Early education of caregivers on how to handle the patient

- Advance planning while patient is competent, establishing a will, proxy, power of attorney, advance directives

- Reduce family stress and misunderstanding of demented patient, reduce caregiver burden, blame, denial

- Promote safety in driving, compliance, cooking, etc.

- Patient’s and Family’s right to know especially about genetic risks

- Promote advocacy for research and treatment development

Medical benefits from early dementia recognition

(see Fillet et al., 2006; Geldmacher et al., 2006 - for references)

- Early diagnosis and treatment and appropriate intervention may improve overall course substantially, including lessening disease burden on caregivers and society

- Specific treatments are now available for Alzheimer’s disease (anti-cholinesterases, memantine). These medications have been shown to:

Improve cognition

Improve function (ADLs)

Delay conversion from Mild Cognitive Impairment to AD

Slow underlying disease process, the sooner the treatment the more benefit

Decreased development of behavior problems

Delay nursing home placement, possibly over 20 months

Delay nursing home placement longer if started earlier

Harms from failure to recognize early dementia

- Dangerous behaviors - cooking, operating machinery

- Driving problems (Molnar FJ et al., 2006)

Listing and accounting for the harms of not screening for dementia

(Chopra et al., 2007)

Missed opportunities for:

- the application of available treatments

- participation in research

- advance care planning

- support of caregivers

Listing and accounting for the harms of dementia screening

(Barratt et al., 1999)

Harms that may occur to those with positive screening test result:

- Complications arising from further testing (only additional clinical questions necessary to enquire about the patient’s history should be considered at this point, as is recommended for evaluation of suspicion of dementia, as recommended by the ANA.

- Adverse effects of treatment must be considered with respect to the benefits, on their own merits, based on the clinician that makes the decision for the treatment.

- Anxiety generated by investigation and treatment - such anxiety must be balanced against the already considerable and appropriate anxiety about Alzheimer’s disease in our society. Screening tests and proper diagnosis are appropriate means to manage that anxiety.

- Costs and inconvenience incurred during evaluation - the cost of dementia evaluation needs to be entered in the calculation of whether screening is “cost-worthy”

Harms that may occur to those with negative screening test result:

- false reassurance - a false negative may wrongly diminish concern and motivation to participate in future evaluation. The consequences of incorrect results are factors that can be accounted for in the decision to screen.

- In reviewing the harms of dementia screening, it is difficult to consider any of them as a major factor against the decision to recommend wide spread screening beyond the calculation of cost-worthiness.

Separate letter by Ashford et al.

Statistical and Pharmacoeconomic Issues for Alzheimer Screening

J. Wesson Ashford, M.D., Ph.D., Helena C. Kraemer, Ph.D., Jared R. Tinklenberg, M.D., Ruth O’Hara, Ph.D., Joy L. Taylor, Ph.D., Jerome A. Yesavage, MD.

Stanford / VA Alzheimer Center, VA Palo Alto Health Care System, &

Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Palo Alto, CA, USA.

Corresponding author: J. Wesson Ashford, M.D., Ph.D., Stanford / VA Alzheimer Center, VA Health Care System (151Y-PAD), 3801 Miranda Ave., Palo Alto, CA 94304, USA; Tel.: (650) 852-3287; Fax: (650) 852-3297: email: ashford@stanford.edu

LETTER

The principles for determining if a screening test should be implemented as they are listed by Dale & Sachs are incomplete (see Kraemer, 1992). For determining whether a screening test is appropriate to give, there are several additional considerations. A comprehensive view can be developed from considering a mathematical analysis of the costs and benefits of a clinical evaluation.

A mathematical calculation of the “cost-worthiness” of a test will give a better direct assessment of whether a screening test should be implemented. To calculate cost worthiness, the following factors must be considered:

I = incidence (new occurrences each year, by age)

$T = cost of test, time to take (for Subject and Tester)

Se = sensitivity of test = True positive / I

Sp = specificity of test = True negative / (1-I)

$B = benefit of a true positive diagnosis

$C = cost of a false positive diagnosis

True negative = (real peace of mind) (no money)

False negative = false peace of mind (no price)

$W = ($B x I x Se) – ($C x (1-I) x (1-Sp)) - $T

If $W is greater than zero, then the test is cost-worthy.

As an example of implementing this equation using conservative estimations, consider the following conservative variables:

I = Incidence of Alzheimer’s disease - increase from 1/1000 per year at age 62, doubling every 5 years.

Se = 0.9 ; Sp = 0.9 (tests of less than 5 minutes appear to be able to reach this level)

(contrasted with a perfect test, Se and Sp = 1)

$B = Vary linearly from $25,000 in a 50 year old patient to $0 in a centenarian patient

$C = a false positive would require a $500 clinic visit to disprove the dementia suspicion

It is apparent from the graph (Figure), that even with these conservative estimates, that a cost for screening for dementia of $25 per year is justified from 75 years of age until over 95. Better or less expensive tests could lead to recommendations as low as 55 years of age, and more valuable treatments would similarly reduce the age for recommending broad application of screening tests.

Risk factors can be introduced to the “cost-worthiness” equation as they affect incidence “I”. For example, APOE genotype can greatly affect risk (Raber et al., 2004), and it would be justified for those with a known APOE-epsilon-4 allele to begin annual dementia screening by age 65 years, while those without that allele could fore-go screening until age 80 years, using the same calculations.

Beyond the hard, cold mathematics, there are humanitarian considerations. As Alzheimer’s disease has become a widely feared condition among the elderly, a more effective means of addressing the problem needs to be provided to identify this problem in those at risk. However, the “cost-worthiness” equation provides a clear guideline for assessing the relative implications of costs and benefits of medical testing.

REFERENCES

Kraemer HC. Evaluating medical tests. Newbury Park, CA: Sage Publications, Inc, 1992.

Raber J, Huang Y, Ashford JW. ApoE genotype accounts for the vast majority of AD risk and AD pathology. Neurobiol Aging. 2004 May-Jun;25(5):641-50.

[pic]

Separate letter by Philippe Robert

It is difficult to argue against the Dale & Sachs paper because it fails to address the issue of screening at the same level as the Ashford et al. paper. The original paper concerns a scientific approach regarding the interest in screening (with the positive and negative aspects addressed). The Dale & Sachs paper is devoted to interpretation concerning the reason for the interest in screening. Just because a medical process is in fashion, it is not necessarily adopted officially (particularly in the US). The Ashford et al. paper has attempted to go beyond the popular recommendation (of not screening for dementia) to consider the scientific points of view.

An additional point is that “PHARMA” (the pharmaceutical industry) is not evil. This is an important, carefully scrutinized area of commercial interest that has contributed tremendous positive advances to mankind.

Dale & Sachs do not adequately address the other aspect of the decision to screen, the purpose of screening. Screening should not be considered only to categorize the patient for administrative or research purposes. Screening should be considered most importantly for making an appropriate early diagnosis, in order to help the patient. An additional point is to distinguish the interest in screening for clinical purposes and for prevention, which requires future research.

Separate letter by Lori Frank, PhD

Senior Research Scientist

Executive Director, Center for Health Outcomes Research

United BioSource Corporation

Bethesda, MD

LETTER

Dale and Sachs raise a critical point about dementia screening, ensuring that evidence in favor outweighs evidence against, and the authors do us all a great service by calling attention to the substantial ethical concerns raised by dementia screening. Some of these concerns are raised by dementia diagnosis as well, and discussions of screening are best served by recognizing the overlapping concerns while maintaining clarity about the differing goals of screening and diagnosis, as we indicate in our Perspective statement (Ashford et al., 2006). There are two issues raised by Dale and Sachs that require further attention specifically: that the “evidence” does not meet standards for recommendation of screening, and that private funding sources for some research activities of some of the authors invalidates the perspective expressed. As a point of clarification, we do not re-interpret existing guidelines; we propose that new guidelines be created. Ideally screening is used within a system of care which permits necessary steps for accurate diagnosis following suspicion of the presence of a disorder.

Dale and Sachs recommend that the public “be counseled to wait until unbiased professional assessments determine that screening for dementia is supported by the evidence.” I agree that evidence in support of dementia screening is a necessity before proceeding with a screening program. The specific types of evidence required are not universally agreed upon, however. Even within existing standards for screening, definitions of key elements currently vary widely. What constitutes “efficacious treatment”? What are superior “outcomes” for treated patients relative to untreated patients? How should “benefits” and “harms” be characterized and measured? Clearly by some definitions all of the conditions that would favor screening are currently met. The discussion is now best advanced by specific review of the available evidence for each condition with keen attention paid to the varying standards for evidence acceptance. What evidence is required to conform to “evidence-based” standards? “Evidence” is too broad a term and specific levels of evidence, and standards used to evaluate them, must be made explicit. This problem with nonspecific definitions of evidence affects many areas of inquiry beyond screening. For example, while RCTs are the gold standard for many clinical questions by design there are questions RCTs cannot validly address, and for which “weaker” observational designs are required.

My greatest concern with this letter relates to their view of a “confluence of conflicted interests,” specifically “systematic incentives to interpret evidence in support of an unproven hypothesis…” They are correct that, as stated at the end of our Perspective, all authors have been associated with the development of screening tools. Clearly, our participation in drafting the Perspective represents scientific interest in dementia screening by all authors. Do our funding sources invalidate our viewpoints?

It is essential that all readers of scientific literature review funding sources and other potential sources of bias. The peer-review process is a critical albeit imperfect method for screening out unacceptable bias. Clearly the peer-review process has failed in the past to screen out bias related to funding source, as documented by multiple authors (Kassirer, 2005; Angell, 2004; Rettig, 2000). How damaging to the advancement of science is private funding, and how is the potential impact of that bias being managed? Under what conditions can privately funded research, or research and opinions generated by individuals with some private funding, be ethical and produce valid results? The chief condition is when involved scientists are motivated to maintain a reputation for credible work. Clearly reputation for credibility is fundamental to all scientists who therefore have strong motivation – even in the absence of personal ethical standards - to avoid actions that could harm that reputation. Beyond the individual level motivators of which a sense of responsibility may or may not exist, many of us belong to organizations with overt requirements for ethical conduct. Can researchers who accept private funding for some of their research activities produce an unbiased opinion? For empirical findings, a second condition pertains: when privately funded research, or research or opinion generated by individuals with some private funding, meets standards for peer review. Submitting work to the peer-reviewed literature is one strategy to ensure that research questions are important, methods are appropriate, and conclusions are acceptable. Peer opinion is therefore an important element in protecting against bias.

Dale and Sachs’ statement indicates appropriate vigilance to the potential for bias due to funding source. On what basis do they have the evidence to conclude that interests are conflicted – what is the evidence? I appreciate their questioning stance but I reject their implicit conclusion that, given presence of private funding for some of the authors’ research or other activities, the research or the opinion expressed is suspect. Bias is not an inevitable result of private funding. Knowledge of the funding source is necessary but not sufficient for determining if the substance of the opinions is credible.

REFERENCES

Angell M. The truth about drug companies. How they deceive us and what to do about it. New York: Random House. 2004.

Kassirer JP. On the take. How medicine’s complicity wiht big business can endanger your health. New York: Oxford. 2005.

Rettig RA. The industrialization of clinical research. Health Affairs 2000: 19:129-146.

NOTES

Comments from Lori Frank and Rod Shankle on 3/11/07 draft

OK-

Not sure variable course is the basic reason in favor of screening; I suggest removing that text.

Calling harms "trivial" is a strong statement -- is it worth it?

Not sure the Merck example is the best to use -- the general point is important though so I suggest nonspecific language there.

Lori

Dear Wes and others,

I suggest one global change in the manuscript. In referring to the consensus paper we published, refer to the paper as the "ADAD consensus article on screening", which gives it more impact in terms of the fact that this is a well considered consensus opinion.

I have also made two small but perhaps significant changes in wording that are highlighted in yellow in the attached revision of Wes's draft.

These changes are as follows:

1. It is more correct to say that at least half of MCI converts to dementia. The reason is that the mean duration of the MCI stage is 7 to 22 years depending upon whether one is in FAST stage 2 or 3 when MCI criteria are met. I have done calculations from various publications and found that using the annualized rate of conversion to dementia from these studies, virtually 100% of MCi cases convert

to dementia if followed for the mean duration of MCi. This is a fundamental issue that is widely misunderstood.

2. I do not agree that MCI screening is an issue not yet ready for consideration, particularly in primary care practice, where most physicians can not distinguish MCI from mild dementia. I do not expect others to agree with me, but since we are talking consensus, I think it would be better to say "a conservative point of view is that MCI screening is not ready for consideration".

Rod

Notes by Paul Dash, Rod Shankle, Fred Schmitt, and Soo Borson

I've also seen several patients put on Aricept for memory complaints but with no documentation of any cognitive testing, and I suspect this may well be a widespread practice.  If these doctors used a screening tool, for normal patients they might be LESS likely to inappropriately prescribe cholinesterase inhibitors, which is another argument for the use of such tests that I don't recall anybody making yet.  The usual argument is that screening is bad because false positives are treated.

  I think you have an excellent point about the difference between a population-based screening program versus doctors using screening tests as part of a routine examination process.  I agree with you that we're not ready for the former yet unless we have more effective means of therapeutic intervention than what we've got now.

 

Paul

 

In a message dated 2/28/2007 10:48:05 P.M. Eastern Standard Time, soob@u.washington.edu writes:

paul, I agree with you. I believe there is no rationale (and could be a several possible harms) from population screening except for research purposes. For screening as a matter of routine medical care of older adults, the harms come when the results are misinterpreted or a physician responds inappropriately by prescribing a cognitive enhancer without a confirmatory clinical investigation. Most of the concerns about possible cognitive screening harms in general medical care derive, in my opinion, from incorrect thinking.

To digress briefly, I am concerned about what seems to be an increasing number of patients whose doctors put them on Aricept if they complain that they are forgetful; one such patient, who has had two bouts of transient global amnesia but no interval cognitive deficits, was prescribed it without any workup at all.

Paul:

Couldn't have said it better.

fred

 

Rod writes:

I have a question for all of you.

If you were going to implement screening, what steps would you take to address the concerns about potential harms that have been expressed by various authors?

Sincerely

Rod Shankle

On Feb 25, 2007, at 9:47 AM, DrDneuro@ wrote

 

To begin with, I think screening will be done by individual physicians as a routine part of the physical exam in older patients, and that's how I think it should be regarded.  I don't see any special need to, for example, get some kind of special signed informed consent from patients to do it, any more than I would think of asking consent before doing each part of the neurological examination.  The patient gives implied consent to be tested by seeing a doctor in the first place.  There's always the "risk" that a neuro exam may pick up incidental findings, which may be quite significant, outside of the patient's chief complaint; I'm sure we've all had the experience of picking up a sensory neuropathy that stems from undiagnosed diabetes, or detected hitherto unsuspected Parkinson's disease in a patient with masked facial expression on testing motor tone.   However, because I don't think asking mental status questions is something people are used to, and to prevent them from thinking that I am doing it because I think they're "crazy," I do say to patients who I want to screen as a matter of courtesy, "I'd like to ask you a few questions to check your memory, if that's ok."  Occasionally a patient may ask why, in which case I respond that I do it routinely in older people because memory loss is so common.  The only times I can remember somebody refusing was with moderately demented patients who were both paranoid and anosognosic about their cognitive losses. 

  So to sum up, I personally don't think anything special needs to be done about doing a brief mental status test as part of a routine examination.  The only "harm" that can come is if the results are used inappropriately, and that is a risk with all findings from the physical exam.

  Anybody have other thoughts on this issue?

 

Paul Dash

-----------------------------------------------------------------

-----------------------------------------------------------------

My first reading of this manuscript is that they have totally confounded the concepts of MCI and dementia, and that will be the first thing to unravel. That will make the process of responding rather messy. Beyond that problem, I think there are some good points to make, and I will try to get my thoughts about them outlined ASAP.

My first thought is that it is difficult to argue against this paper because the two papers are not at the same level. Our paper concerns a scientific approach of the interest of the screening (with the positive and negative aspect). This new paper is devoted to interpretation concerning the reason for the int

erest in screening. Even if it is in fashion it is not adopted (particularly in the US to have such comment). Further more I do not think that pharma are devil. To come back to the other aspect, the scientific points of view discussed are poor. One point to be discussed concerns the interest of screening and to separate the notion of screening for prevention (in order to help the patient) and not to only categorize the patient.

There is a small but vocal group of nihilistic purists that insist that pharmacologic treatments and other management interventions with dementia patients are of no use - in spite of the preponderance of evidence of benefit. This seems to be the same group that is against screening.

Clearly, Pharma is commercially interested in the treatments for Alzheimer's disease and screening, but they are not evil, and they are carefully scrutinized. Similarly, we have interests in screening tests, but arguably we have more experience with the need for such tests and have not entered this area for any evil purposes. We must appreciate the need for and benefits of debate, as well as the attention to our possible motivations. But the decision to screen must be based on estimations and calculations of the risks and benefits, not an attack on our intentions.

Zaven mentioned one concept which could be hi-lighted: dementia is somewhat different from other disorders in that it has to emerge slowly from a broad range of normal cognition (at a variable rate), and so there is no clear measure of onset. Therefore, validation of studies is a long-term proposition and screening can only give a probabilistic statement until the illness is well situated.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download