This issue on the diagnosis of Alzheimer's disease (AD) is the product of a special meeting of the International Psychogeriatric Association with the cosponsorship of Alzheimer's Disease International, the European Federation of Neurological Societies, the World Health Organization, and the World Psychiatric Association. The meeting was held in Geneva, Switzerland, from November 10 to 12, 1996. Participants included many of the leading experts on the various aspects of AD diagnosis as well as clinical experts, general experts in the field of AD, organizational representatives, and outstanding clinician-scientists who served as facilitators and in other capacities.

Current knowledge with respect to the diagnosis of Alzheimer's disease (AD) is reviewed. There is agreement that AD is a characteristic clinicopathologic entity that is amenable to diagnosis. The diagnosis of AD should no longer be considered one of exclusion. Rather, the diagnostic process is one of recognition of the characteristic features of AD and of conditions that can have an impact on presentation or mimic aspects of the clinicopathologic picture. The present availability of improved prognosis, management, and treatment strategies makes the proper, and state-of-the-art, diagnosis of AD a clinical imperative in all medical settings. Concurrently, information regarding the relevance and applicability of current diagnostic procedures in diverse cultural settings must continue to accrue.

Alzheimer's disease (AD) is an example of cortical dementia, where most of the pathological changes lie in cortical association areas and the predominant clinical features are symptomatic of cortical dysfunction. The differential diagnosis of AD is complex and is usually made in two stages: diagnosis of the dementia syndrome and diagnosis of the underlying etiology. Each of these stages is complicated by the existence of conditions that mimic dementia in presentation and by the risk of inappropriately diagnosing AD where other conditions are responsible for the dementia state. Despite the fact that the concept of dementia, in terms of a decline in cognitive function and behavioral change, has existed for many years, there is still a general lack of agreement about the components of the syndrome. This article reviews the issues surrounding the differential diagnosis of the dementia syndrome and explores the points of differentiation between AD and some similar etiologies encountered when making a differential diagnosis.

Vascular dementia (VaD) is the second most common cause of dementia. In addition, cerebrovascular diseases (CVD) coexist with other causes of dementia. Cognitive impairment related to CVD may be preventable and these patients could benefit from therapy, which emphasizes the importance of early detection and accurate diagnosis of VaD. The conventional concept of VaD is that of multi-infarct dementia (MID). However, VaD is not only MID; it relates to different vascular mechanisms and different changes in the brain, and has different clinical manifestations with different causes. Critical conceptual questions include the cognitive syndrome and the vascular causes. It is unclear whether the conventional concept of dementia is appropriate or should be substituted with a milder and broader definition, such as vascular cognitive impairment. Furthermore, there is confusion about the causes, especially the role of lesion characteristics and the noninfarct factors. The current diagnostic criteria for VaD are based on the infarct concept. The NINDS-AIREN criteria include dementia, CVD, and a relationship between these two disorders. The criteria define the CVD and the relationship between dementia and CVD, and list supporting clinical features for the diagnosis of VaD, as well as features that make the diagnosis uncertain. The interrater reliability of these criteria is moderate to substantial (κ .46 to .72). The expected antemortem accuracy using these criteria approaches 90%. The challenge is to correctly diagnose combined cases with both vascular and Alzheimer-type pathology. Main tools in the diagnosis include a detailed medical history, neurological examination, clinical or neuropsychological mental status examination, and basic laboratory examinations. Brain imaging should always be performed, preferably using magnetic resonance imaging. The diagnosis, especially in early cases, is usually made by a neurologist. The challenge for the primary care physician is to identify the early cases needing further examinations and organizing treatment and follow-up after diagnosis.

Dementia and depression are the two most common mental illnesses in late life, and it is probable that they will coexist in many patients. This coexistence is complicated by the fact that both illnesses can be mistaken for each other, so that many patients with Alzheimer's disease (AD) may initially be diagnosed as depressed, whereas depression is a recognized cause of cognitive impairment. It is important to correctly differentiate between these two diagnoses; if depression is the cause of the cognitive impairment, full recovery is possible. It is also important to recognize depression in patients with AD, because depression represents a treatable source of additional disability. For patients with AD living in the community, functional limitation can determine whether the patient remains at home or is institutionalized, so treatment that can improve functional ability should be strongly considered. The newer antidepressants are better tolerated without increase in cost, so these drugs may provide a therapeutic advantage over the older tricyclic drugs. This article focuses primarily on techniques for establishing a differential diagnosis, with particular emphasis on patients in the primary care setting, and briefly considers the value and impact of treatment.

In recent years, patients who are at high risk for developing Alzheimer's disease (AD) have become a focus of study. Several research groups have identified these patients, developed diagnostic criteria, and followed the patients longitudinally. These patients therefore constitute a clinical entity that is suitable for therapeutic interventions. In this article, we report our 5-year experience at the Mayo Clinic in characterizing a group of patients with mild cognitive impairment. These subjects were recruited from community-dwelling individuals who were receiving general medical care in the Department of Internal Medicine. They received the diagnosis of mild cognitive impairment if they met the following criteria: (a) complaint of defective memory, (b) normal activities of daily living, (c) normal general cognitive function, (d) abnormal memory function for age, and (e) absence of dementia. In following more than 75 of these patients, some of whom have been studied for as long as 5 years, it appears that approximately 10% to 15% of the subjects evolve to AD each year. We therefore evaluated the cognitive profiles of these patients at the time of their initial diagnosis in an attempt to predict who would remain stable and who would evolve to AD. Certain features of learning and memory performance indicated patients who were more likely to progress, but the strongest predictor was their apolipoprotein E status. The patients who possessed an ε4 allele were more likely to convert to AD at a more rapid rate then those who were not carriers. Our results and similar data from other study groups indicate that diagnostic criteria can be defined for patients who are likely to convert to AD; the natural history of these patients is becoming apparent, and variables that predict ultimate conversion can be defined. Consequently, patients with mild cognitive impairment constitute an important group for study, particularly from the aspect of therapeutic interventions.

The role of the physician is critical in the evaluation of cognitively impaired patients for two reasons: (a) The longitudinal relationship between the patient and physician provides a unique opportunity to observe changes over time, generating confidence that may encourage the patient and family to report any concerns; (b) All research in this field is geared toward the earliest possible intervention, a process made possible by early diagnosis. Preparations for interview and the interview process itself require special knowledge of dementia on the part of the evaluating physician, who must understand the potentials and problems of both the content and the process of taking a history from the demented patient and his/her caregivers.

Burns: It has been said that the Hachinski score is actually more an infarct score than an ischemic score. Dr. Erkinjuntti, what do you think?

We were asked to consider differential diagnosis. We decided that it really is different from the other diagnostic modalities and cannot be treated in the same way as, for example, global or neuropsychological measures. In the context of this meeting, we believed it would be appropriate to consider clinical criteria. Thus, clinical criteria for the dementia syndrome as well as for specific diseases were discussed. We recognized that in the future, an increasing number of patients will present with cognitive impairment who do not fulfill the clinical criteria for dementia. These patients may alternatively have an isolated memory impairment syndrome. Nevertheless, a diagnosis will still need to be made in these patients, at least in terms of the underlying molecular pathology, in order to implement potential treatments.

The Mini-Mental State Examination (MMSE) is a widely used screening test for cognitive impairment in older adults. Because the guidelines for its application are brief, the administration and scoring of the test can vary between different individuals. This can diminish its reliability. Furthermore, some of the items must be changed to accommodate different settings, such as the clinic, home, or hospital. Because there are no time limits, it is not clear how long one should wait for a reply to a question. It is also not clear how one deals with answers that are “near misses.” The goal of the Standardized Mini-Mental State Examination (SMMSE) was to impose strict guidelines for administration and scoring to improve the reliability of the instrument. The reliability of the MMSE was compared with the reliability of the SMMSE in 48 older adults who had the tests administered by university students on three different occasions to assess the interrater and intrarater reliability of the tests. The SMMSE had significantly better interrater and intrarater reliability compared with the MMSE: The interrater variance was reduced by 76% and the intrarater variance was reduced by 86%. It took less time to administer the SMMSE compared with the MMSE (average 10.5 minutes and 13.4 minutes, respectively). The intraclass correlation for the MMSE was .69, and .9 for the SMMSE. Administering and scoring the SMMSE on a task-by-task basis are discussed.

Neuropsychological assessment involves the observation and measurement of an individual's behavior in relation to a given stimulus that has been selected for its likelihood to provoke an abnormal response in the presence of neurological damage. This approach incorporates a medical and psychological approach to diagnosis in that both pathology-specific signs and deviations from normal distributions are examined. In the case of Alzheimer's disease (AD), this methodology has assumed considerable importance. This is because behavioral indicators remain the principal basis for provisional diagnosis when there are no biological markers. Observations of elderly persons with and without senile dementia have led to the development of stimuli, or tests, that are able to assess the functioning of specific cognitive processes affected in AD as opposed to more common age-associated changes. Many of these tests have been found to be culturally portable. Diagnostic difficulties arise principally from the fact that many neuropsychological tests cannot be used in isolation, from uncertainty as to the limits of “normal” functioning and from the psychometric problems related to pathological thresholds and item-weighting. The usefulness of neuropsychological testing in the future could undoubtedly be enhanced by the use of measures derived from indices of decline in functioning rather than normative data and the development of functional testing in collaboration with cerebral imaging.

The goal of this article is (a) to review a series of measurement standards for group and individual-patient applications of cognitive tests for Alzheimer's disease (AD); and (b) to review the literature and to use specific data from the NORMACODEM (a project of standardization of cognitive and functional tools for AD) to evaluate the measurement standards of the Alzheimer's Disease Assessment Scale-Cognitive (ADAS-Cog). Seven measurement standards for group and individual patient use were evaluated: practicality, extent of measurement, depth of measurement, cross-sectional precision, longitudinal precision, validity, and standardization. The results indicated that ADAS-Cog meets practicality standards, that the scoring method is not difficult, but training is needed to administer the test. In comparison with other rating scales, the ADAS-Cog is regarded as more comprehensive, although it is not a substitute for an extensive neuropsychological test. The ADAS-Cog is not a tool that is suitable for primary care physicians, although it is a valuable screening test that has utility in both early detection (diagnosis) and staging of AD, and is useful in the severity range of Global Dementia stages 1 to 6. Interrater reliability coefficients for the ADAS-Cog satisfy recommended standards, and the test shows good internal consistency and reproducibility. Age and education are associated with the ADAS-Cog score. There is, however, a lack of standardization, which restricts the comparability of study results, and test materials remain unpublished.

The SKT is a simple and short test for assessing cognitive impairment of memory and attention. The test consists of nine subtests, each limited to a maximum performance time of 60 seconds. Thus, total testing time normally does not exceed 10 to 15 minutes. Test scoring and auditing are also accomplished within a short time and are easily learned. The test is useful for staging the severity of cognitive deficits and for assessing the benefits of therapy, especially with patients suffering from dementia. The SKT is highly accepted both by patients, who are motivated by the appealing character of the tasks and their play-like nature, and hospital clinicians and general practitioners (GPs) because of its brevity and simplicity. The SKT was introduced in Germany two decades ago and has been successfully validated. Test versions of the SKT in English and other languages have confirmed its international applicability and validity, a point of particular importance in light of the current attempts to harmonize methods for assessing the efficacy of pharmaceutical products.

We propose a two-step process for the assessment of dementia using standardized instruments. The family physician performs a screening consisting of taking a medical history, gathering information from relatives and friends of the patient, and administering the combined Mini-Mental State Examination (MMSE) and Clock Drawing Test (CDT). Specialists examine patients with suspected dementia to confirm the diagnosis of dementia and, after a thorough differential diagnostic process, provide the family physician with recommendations for treatment. Specialists should perform neurological and psychiatric examinations, imaging (computer-assisted tomography [CT], magnetic resonance imaging [MRI]), and laboratory work-up. The Consortium to Establish a Registry for Alzheimer's Disease (CERAD) core neuropsychological battery is proposed to serve as a minimal data set that is internationally compatible. In addition, we recommend the Nurses' Observation Scale for Geriatric Patients (NOSGER) as a standard tool for functional assessment.

There is increasing awareness of the importance of psychopathological and behavioral changes in dementia and a need for a technique to measure these noncognitive features. Such a schedule should keep screening questions to a minimum, include a severity measure, exclude symptoms resulting from physical illness, be as brief as possible, and not mix domains of psychopathology. To test the reliability, sensitivity, and validity of a newly developed test, 30 carers were interviewed four times during 6 weeks. An obligatory stem question in each category was followed by supplementary questions. The interviewer recorded the presence of each symptom, its severity, when each symptom started, its duration, and whether it was still present. To detect the presence of delusions, the informant was asked about the patient's insight. Satisfactory differences in mean κ values were demonstrated in test-retest and interrater reliability and validity compared with other techniques. This test may be useful to measure the outcome of drug trials, for correlating psychopathological and behavioral changes with autopsy findings and in epidemiological surveys.

The Columbia University Scale for Psychopathology in Alzheimer's disease (CUSPAD) was developed as a short, semistructured instrument that can be administered by a clinician or trained lay interviewer to the relatives or caregivers of patients with probable Alzheimer's disease (AD). The components of the scale focus on symptoms of psychosis, behavioral disturbance, and depression. A simple decision-tree approach is used to identify the presence of specific psychotic features and makes it possible for the instrument to be used by trained lay interviewers. Interrater reliability has been established. Use of this scale in a multicenter longitudinal study of patients with mild to moderate AD revealed that agitation was the most common and persistent symptom, depressed mood with vegetative signs was the least common and least persistent, and paranoid delusions and hallucinations showed moderate persistence over time. Besides its established psychometric properties, the strengths of the CUSPAD include its brevity, clarity in assessing psychotic symptoms, and ease of administration. These features make it useful for cross-sectional and longitudinal evaluation of the symptoms of psychopathology in AD. The main disadvantage is the lack of quantitative ratings of severity for many of the components, making it unsuitable for use in psychopharmacologic trials.

Ham: There seems to be an astonishing range of expectation from this group as to what a general practitioner (GP) will do for his or her patients, ranging from screening only through to being the primary manager. I feel that we are expecting a fairly high level of sophistication from the GPs within an average British, Canadian, or American family practice. I think we need to be aware that, for some time, the core recommendation in the USA has been to assess the patient using the Mini-Mental State Examination (MMSE) or the Clock Test and perhaps two or three other quantitative assessments. Noncognitive features, such as behavior, are harder to quantify and the individual who assesses them must be the patient's primary manager. It will increasingly happen in the USA that as care becomes more managed, a greater emphasis will be put upon primary care.

Our group accomplished two tasks. First, it identified several general issues that were relevant to our discussions and probably also to the discussions of the other groups; and second, it formulated several criteria for the evaluation of instruments for neuropsychological assessment and applied them to generally available instruments.

Inability to perform activities of daily life is inherent to dementia and an essential component of its severity. Examination of this disability is crucial for diagnosis, management of the patient and family, and evaluation of treatment effects. To examine everyday disability in community-dwelling patients with mild dementia, we developed the Interview for Deterioration in Daily living Activities in Dementia (IDDD) at the memory clinic of the Academic Medical Center at Amsterdam. The IDDD is a caregiver-based measure, which consists of 20 concretely worded items, reflecting the initiative to perform and actual performance of self-care and more complex activities. The original version of the IDDD (1988) was an interview measure, but since 1990 a paper-and-pencil version has been used, which has good construct validity and test-retest reliability, as well as good responsiveness to deterioration over 6 months. In the present study, we examined interobserver agreement among 25 caregiver pairs, which consisted of primary and secondary caregivers. Although the reliability of caregiver reports is often questioned, we found substantial to almost perfect agreement at subscale level and acceptable agreement for most of the items. We conclude that the paper-and-pencil version of the IDDD is a suitable instrument for the description and discrimination of disability among patients with mild dementia still living at home. The IDDD may also be a useful method to incorporate in clinical trials and longitudinal studies.