The study included 717 Medicare recipients aged 65 years and older (average age 80.1) in Manhattan. They or their caregivers completed a questionnaire about stroke history, including whether they'd ever had symptoms of stroke or been told by a doctor that they had a stroke.

On the questionnaire, 85 participants (11.9 percent) reported a history of stroke, but MRI scans found evidence of stroke in 225 participants (31.4 percent).

The sensitivity of stroke -- number of people who reported having a stroke divided by the total number of those with stroke detected on MRI -- was 32.4 percent. The specificity -- number of people who reported having no history of stroke divided by the total number of those with no evidence of stroke detected on MRI -- was 78.9 percent.

"Lower functioning memory, cognitive or language ability or presence of hypertension [high blood pressure] or myocardial infarction [heart attack] were associated with an increased frequency of false-negative reports," study author Dr. Christiane Reitz, of Columbia University Medical Center in New York City, wrote in a news release from the journal.

"Our results indicate that sensitivity and specificity of stroke self-report are low when using MRI scans as validation," the researchers concluded. "In stroke research, sensitive neuroimaging techniques rather than stroke self-report should be used to determine stroke history."

The study appears online and in the July print issue of the Archives of Neurology.

More information

The U.S. National Institute of Neurological Disorders and Stroke has more about stroke.