Related Experiment Video
Updated: Mar 25, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Routine cognitive screening in a neurology practice: Effect on physician behavior
Michael Rosenbloom1, Soo Borson1, Terry Barclay1
1HealthPartners Center for Memory and Aging (MR, TB, LRH), St. Paul; Department of Neurology (MR, JM), University of Minnesota Medical School, Minneapolis; Geriatric Research, Education and Clinical Center (GRECC) (JM), Minneapolis VA Health Care System, MN; the University of Washington School of Nursing (SB), University of Washington School of Medicine, Seattle; and HealthPartners Institute for Research and Education (LRH, AW, LS), Bloomington, MN.
Background:
Alzheimer disease is one of the most prevalent and costly neurologic disorders. American Academy of Neurology guidelines call for diagnosis and treatment when dementia is present, but provide no specific instruction relating to cognitive screening.
Methods:
Our center piloted a cognitive screening initiative using the Mini-Cog, which was administered to all neurology patients aged ≥70 years without a history of a cognitive disorder.
Results:
There was a 37.4% screen positive rate on the Mini-Cog. The percentage of patients with subjective memory complaints did not differ between patients screening positive vs negative on the Mini-Cog. Prospective analysis over an 18-month postscreening period showed that individuals screening positive for cognitive impairment were 10 times more likely to have follow-up cognitive assessment by the provider (p < 0.0001), almost 3 times more likely to be referred for neuropsychological testing (p = 0.003), and 3 times more likely to receive a diagnosis of cognitive impairment or dementia (p < 0.0001) compared to those screening negative. Diagnosis of a cognitive disorder, referral to a cognitive specialty clinician, and prescription of cognitive-enhancing medications were no more frequent than was observed in a randomized trial of screening in primary care, and evidence of neurologists' actions relevant to cognitive impairment was found in a minority of individuals screening positive.
Conclusion:
Further studies are needed to better understand factors influencing neurologist actions in the evaluation and treatment of cognitive impairment.
