The disappearing subject: exclusion of people with cognitive impairment and dementia from geriatrics research.

OBJECTIVES: To evaluate exclusion of persons with cognitive impairment from research in geriatrics by determining its frequency, method, and rationale and treatment in the resulting publications.
DESIGN: All original research articles published in 2008 and 2009 in the Journal of the American Geriatrics Society (n = 434) were reviewed using a structured data collection tool.
SETTING: The Journal of the American Geriatrics Society. PARTICIPANTS: There were no participants in this study. MEASUREMENTS: Data captured included recruitment method, explicit criterion for exclusion of persons with cognitive impairment, justification of exclusion criterion, reason given for exclusion, percentage of individuals excluded, and mention of exclusion as a possible limitation.
RESULTS: Of 434 articles examined, 16% used recruitment methods likely to reduce participation by persons with cognitive impairment. At least 29% of studies (n = 127) employed explicit exclusion criteria. Half used the Folstein Mini-Mental State Examination (MMSE), with variable cut points (10, 12, 17, 18, 23, 26), and 19% excluded individuals for "having dementia" without specifying how this was determined. Few (6%) provided any justification for exclusion criteria used, only 43% gave any reason for exclusion, and only 14% discussed exclusion as a possible limitation.
CONCLUSION: Persons with cognitive impairment are frequently excluded from research, often without rationale or mention of exclusion as a limitation or any discussion of its potential effect on the evidence base in geriatrics. When necessary, exclusion should be done thoughtfully and with awareness that this may reduce the clinical utility of study findings.