Mild Cognitive Impairment in Clinical Practice: A Survey of AAN Members

August 13, 2010


J. Scott Roberts, PhD, from the Department of Health Behavior and Health Education, University of Michigan School of Public Health, discusses the paper "Mild cognitive impairment (MCI) in clinical care: A survey of American Academy of Neurology members" that was recently published in Neurology (2010;75:425-1). He spoke with José Merino, MD, Science Editor of Please briefly describe the methods and results of your study?

Roberts: The survey was a self-report questionnaire, administered via the Internet. Respondents were asked about their preferred clinical terms for describing cognitive symptoms of mild severity, their usual practices when treating patients with such symptoms, and their views on the benefits and drawbacks of the MCI term. The sample was not representative of all US neurologists, as only AAN members with specified practice interests were eligible for the survey. The AAN assisted with survey development and was responsible for study recruitment, implementation, and data analysis. Based on the results of your survey, how are neurologists using the diagnosis of MCI? 

Roberts: Our results suggest that neurologists who see patients with cognitive symptoms of mild severity are frequently using the MCI diagnostic category (including its amnestic subtype) for classification. The ICD-9 code for MCI is 331.83. How prevalent is MCI? Can we identify which individuals with MCI may progress to Alzheimer's disease (AD)?

Roberts: Population-based studies in the US have suggested that approximately 1 in 5 older adults can be classified as having some type of MCI. Research studies have identified numerous factors that predict progression from MCI to dementia, including APOE genotype, performance on certain neuropsychological tests, and biomarkers identifiable through neuroimaging and cerebrospinal fluid. However, these tools have not yet been tested in clinical practice. What did survey respondents feel are major benefits and drawbacks of using the diagnostic category of MCI?

Roberts: A majority of respondents endorsed several potential benefits of a diagnosis of MCI, such as involving the patient in planning for the future, motivating possible risk reduction activities, and helping with financial planning. A minority of respondents noted drawbacks, including difficulty of diagnosis, insufficient attention to etiology, and possible unnecessary worry for the patient and family members. There is no evidence to support the use cholinesterase inhibitors in patients with MCI, yet your survey reveals that these medications are widely used. Can you comment on this phenomenon?

Roberts: Our results do not directly address neurologists' rationale for prescribing particular medications for MCI patients. It may be that the clinician, based on the individual patient's symptoms and medical history, believes that the etiology of MCI is indeed Alzheimer's disease, making cholinesterase inhibitors relevant. The neurologist, patient, and family members may also feel pressure to "do something" in the face of this condition, with medications representing a concrete action to avoid feeling at the mercy of a potentially progressive, severe disease. Why do you think there were differences in terms of diagnosis and management of MCI between academic and private practice neurologists?

Roberts: Academic neurologists were less likely than private practice neurologists to report prescribing cholinesterase inhibitors for MCI, perhaps reflecting a greater awareness of treatment trials in this population that have failed to show significant benefits from these medications. What are the strengths and limitations of your survey-based studies? 

Roberts: The strength of this survey design is that it can provide a "snapshot" of the views and typical practices of a broad sample of practicing neurologists. A main limitation is that neurologists' self-reported behaviors may not always accurately reflect their true practices. Challenges of these types of surveys include identifying representative populations and crafting survey questions that are free of biases and potential for misinterpretation. What is the relevance of your findings for researchers and clinicians?

Roberts: The survey findings raise several issues that could be explored in future clinical research. Future studies should seek more objective measures of outcomes such as billing codes used and medications prescribed. "In vivo" studies of patient–provider communication and treatment decision-making during clinical encounters may also be warranted. Surveys of this type could also be extended beyond neurologists to other medical professionals (e.g., geriatricians, internists, allied health professionals) who often encounter MCI. As for clinicians, findings may be helpful in prompting neurologists to examine (and improve where necessary) their own practices and decision-making processes as it applies to the diagnosis, treatment, and education of this patient population. What are the implications of your findings for the AAN and other organizations?  Did you identify specific areas that should be the focus of CME activities?

Roberts: These findings could help inform AAN's recently revised practice parameter for MCI, currently under development. Specific areas of focus for CME activities might include a review of findings from treatment trials in MCI populations, as well as patient and family education about MCI.

Author Disclosure
Dr. Roberts has received support from the following: National Institutes of Health; Alzheimer's Association; University of Michigan; Veterans Administration. These funds have supported various projects related to health education in Alzheimer's disease and ethical and practice issues in genetic susceptibility testing for common diseases.

Dr. Merino performed a one-time consultation with staff from Bell, Falla and Associates.