Screening for Cognitive Impairment in the Elderly

Overview
Cognitive impairment is a common finding in older people, as the prevalence of dementia increases with age. The most common cause of dementia is Alzheimer’s disease, a slowly progressive primary dementing disorder. Intercurrent illnesses, infections, metabolic disturbances and drug intoxications may all cause or exacerbate mental confusion. Depression may worsen and occasionally mimic dementia. Identification of dementia in the early stages offers the potential to plan to deal with subsequent deterioration, organize community supports, and anticipate later incompetence, by measures such as advance directives and power of attorney. A large number of drugs have been studied for their effect on improving the cognitive and behavioural aspects of Alzheimer’s disease. While beneficial effects on cognitive performance have been documented, these are rarely of sufficient magnitude to be of clinical importance. The potential harm of labelling an individual as demented must be weighed against possible benefits. There is insufficient evidence to recommend for or against measures to detect asymptomatic cognitive impairment. The prudent physician is advised to remain alert for clues that suggest deteriorating cognitive function, and then to pursue an appropriate diagnostic course of action.

Burden of Suffering
Prevalence studies in Europe, the United States and Canada reveal relatively consistent findings. While methods of ascertainment vary from study to study, the prevalence of severe dementia in people aged 65 and over residing in the community is between 2.5 and 5%.<1-3> For mild degrees of dementia the prevalence is age-dependent, with rates less than 5% below 75 years, to 40% or higher above the age of 80.<4> The incidence has been estimated at 1% in persons over 65 and up to 2.5% in those over the age of 80. Projected figures for Canada are 225,000 new cases of dementia per year.<3> In addition to the cognitive deficits produced by dementia, behavioural abnormalities are common. These frequently lead to excessive caregiver stress, and may precipitate hospital or institutional admission. Behaviours such as restlessness, wandering, aggression, failure to recognize relatives and locations, and inappropriate sexual behaviour are particularly troublesome. The presentation of physical disease may be altered or obscured. People with dementia have reduced survival.

Maneuver
Dementia is readily recognizable in its advanced stages. In the early stages it often goes undetected. Conventional medical histories and examinations frequently fail to identify cognitive impairment or to distinguish it from hearing impairment, depression, aphasia, bradykinesia, etc. Criteria have been established for the diagnosis of dementia.<5,6> While the "complete mental state" examination is well described in standard texts, attempts have been made to develop short mental status questionnaires to screen for cognitive impairment. The Mini Mental State examination (MMS)<7> is the most frequently used and has the most clearly defined test characteristics. Others include the Short Portable Mental Status Questionnaire (SPMSQ)<8> and the Clock Drawing test.<9> The MMS requires no special equipment and can be completed within 5-10 minutes. Little training is required, and a standardized version has been developed.<10> The sensitivity of this instrument to detect moderate dementia approaches 90% with a cut-off point of 24 out of 30. Corresponding specificity is about 80%. The test is valid and reproducible, particularly in its standardized form.<10> The SPMSQ has had similar sensitivity in published series,<11> but has been less well studied. It is a less comprehensive instrument than the MMS, as it examines principally orientation and memory, and does not cover areas of language or motor tasks. The clock drawing test, originally developed for examination of parietal lobe function, is an extremely quick test. Despite its simplicity, it offers excellent sensitivity (92%) and specificity (97%) for the detection of moderate to severe dementia.<12>

An alternative approach to the use of mental status questionnaires is screening using Instrumental Activities of Daily Living (IADL). Sixty-nine percent of a random sample of 2,792 community dwellers aged 65 years and over were subjected to a two-phase screening procedure. The first phase included a functional assessment using an IADL scale and the MMS. Subjects who fulfilled Diagnostic and Statistical Manual of Mental Disorders criteria for dementia, were evaluated by a neurologist using National Institute of Neurological Diseases and Stroke – Alzheimer Disease and Related Disease Association criteria for dementia. The prevalence of dementia in this sample was 2.4%. Subjects experiencing difficulty in telephone use, use of public transportation, responsibility for medication use and handling finances had a 12 times greater probability of being diagnosed with dementia.<13> The MMS score is correlated with the ability to perform daily activities in cognitively impaired individuals.<14>

When an older person is discovered to have cognitive impairment a search is usually made for illnesses causing cognitive impairment which may be modifiable, in the hope that the condition will be improved or reversed. Although earlier literature suggested that up to one third of cases of apparent dementia were caused by illnesses whose treatment could lead to improvement, a recent overview analysis of the subject concluded that only 11% of dementing illnesses in older people resolved during follow-up (8% partially, 3% completely). The most common underlying remediable factors were drug intoxication, depression and metabolic abnormalities.<15> Two recent large community studies have been carried out to examine the results of screening and subsequent investigations. In a three-phase study in Eastern Baltimore, Md, 78% of 3,481 subjects completed the National Institute of Mental Health Interview Survey questionnaire together with a version of the MMS. Eighty percent of a random sample of these subjects (n=1,806) were examined by psychiatrists. Thirty-six of the 44 diagnosed by a psychiatrist as having definite or probable dementia were subjected to full neurological investigations. The prevalence of dementia was 6.1% in this population and no cases of reversible dementia were found.<16> In a second large community study from East Boston, 3,624 subjects over the age of 65 were examined with a screening procedure based on detection of immediate and delayed memory. Four hundred and seventy-two who appeared to have cognitive impairment were identified. Of these, 83.5% were found to have a clinical diagnosis of probable Alzheimer’s disease.<4> The vast majority of older community subjects discovered by screening to have cognitive impairment are suffering from Alzheimer’s disease and do not have a correctable or even potentially correctable dementing illness. While there are theoretical reasons to identify people with dementia for early treatment, early intervention has not been shown to modify the course of the illness. Theoretically, in those who have vascular dementia, correction of risk factors (e.g. treatment of hypertension, or anticoagulation for atrial fibrillation) could delay the progress of dementia. A wide variety of agents have been tested in Alzheimer’s disease. Drugs presently showing most promise increase the central levels of acetyl choline. Tacrine (tetrahydroaminoacridine), has been approved for use in the United States and is available in Canada. Modest but definite improvements in cognitive performance have been documented in some<17-19> but not all<20,21> studies. Drugs which promote enhanced cerebral metabolism have also shown some benefit, although drugs such as Hydergine have largely been abandoned in the face of recent studies which have shown no significant effect.<22> Chelation therapy with desferrioxamine has shown some promise and may delay disability in Alzheimer’s disease.<23> There are no published trials examining the effects of treatment on subjects who have been discovered by community screening to suffer from cognitive impairment.

One potential benefit of early identification is the ability to plan for the anticipated further cognitive decline. For example, the assignment of a sustaining power of attorney can be made at a time before mental incompetence occurs, obviating more complex maneuvers to handle an individual’s estate at a later date. The ability to discuss advance directives with an individual is another potential benefit. Planning and consideration of timely relocation to a more protected environment may also be beneficial and early involvement with caregiver support groups may assist individuals in dealing with ultimate disability. None of these theoretical advantages has been subjected to appropriate study.

Potential negative consequences of early identification of cognitive impairment clearly exist. Labelling an individual as demented may affect his or her ability to obtain life or health insurance, and may influence attitudes towards the individual by health care professionals and others. The label of Alzheimer’s disease may cause prejudice and difficulty in gaining admission to some long-term facilities. The negative effects of labelling an older person as demented have not been studied systematically, although a small body of social science literature explores this important area.<24> Negative attitudes have been identified among professionals and lay people.

Recommendations of Others
The U.S. Preventive Services Task Force recommended against screening for cognitive impairment in 1989.<25>

Conclusions and Recommendations
Despite the theoretical advantages of identifying individuals with cognitive impairment, there is no evidence to indicate whether this leads to a net benefit or risk to the individual. Although pharmaceutical agents are able to produce measurable changes in cognitive performance in people with Alzheimer’s disease, none has been shown to result consistently in clinically significant improvement. The high cost of investigation to exclude reversible causes of dementia, and the negative effects of labelling are examples of potential harm. Identification of asymptomatic cognitively impaired individuals by the use of short mental status tests or by any other means has not been demonstrated to produce benefit. Thus there is insufficient evidence to recommend for or against screening (C Recommendation). The prudent physician should be alert for any reports or behaviour which may indicate cognitive impairment (e.g. forgetting appointments, poor medication compliance), and then pursue appropriate strategies for further investigation and treatment.<26>

Unanswered Questions (Research Agenda)

1. Although two of the brief mental status instruments reviewed appear satisfactory for case finding in primary care, they are not ideal, and more sensitive and specific instruments are desirable.
2. The search for effective treatments for Alzheimer’s disease should incorporate outcome measures including physical functioning, behaviour measures of caregiver burden and ability to delay or prevent institutional care.
3. Trials of screening are necessary to examine the impact of detecting cognitive impairment, its subsequent investigation and treatment.
4. Studies should be directed towards discovering any negative effects from attaching the label of Alzheimer’s disease or cognitive impairment to a person.

Full Citation
Patterson C. Screening for cognitive impairment in the elderly. In: Canadian Task Force on the Periodic Health Examination. Canadian Guide to Clinical Preventive Health Care. Ottawa: Health Canada, 1994; 902-9.