A Surprising Link Between Mild Cognitive Impairment and Long-COVID
Researchers at New York University have discovered that patients with lingering symptoms of SARS-CoV-2 infection, a condition commonly referred to as Long-COVID, are significantly more likely to develop mild cognitive impairment (MCI) than people who weren’t infected or made full recoveries from the virus. The surprising finding, reported recently in Alzheimer’s & Dementia, came from a study of NACC participants at the university’s Alzheimer’s Disease Research Center (ADRC).
The project began in the darkest days of the COVID-19 pandemic, when New York City was renting refrigerated trucks to store the dead bodies overflowing from local morgues. “Like many physicians, one of our coping strategies is intellectualization, so in the face of all this insane amount of death and horror … we started collecting data,” says Jennifer Frontera, professor of neurology at NYU and lead author on the new paper.
That work soon revealed that many survivors of the disease had a slow recovery, often suffering from “brain fog” that affected their memory, attention, and reasoning ability. “We had a couple of grants early on working with Dr. Thomas Wisniewski, who is the head of our ADRC at NYU … to look at these people prospectively,” says Frontera. The established ADRC already had pre-pandemic data on a large population of people, many of whom later contracted COVID-19. “So we had a group of people pre- and post-COVID that we could look at their differences,” says Frontera.
Other scientists have also been trying to understand Long-COVID, but the NYU team had a unique advantage: their subjects had undergone complete evaluations by trained neurologists using the Uniform Data Set (UDS) criteria. “In COVID, the problem is there’s many things that can cause abnormal neuropsychiatric testing and cognitive impairment, systemic issues like hypothyroidism, B12 deficiency, substance abuse, sleep disorders … that are mechanistically different and require different therapeutic approaches,” says Frontera. She adds that the NACC protocol, which entails having panels of fellowship-trained neurologists evaluating each participant and reaching consensus diagnoses, “is really the gold standard.”
For the new project, the ADRC neurologists were not permitted to see patients’ COVID-19 histories, to avoid biasing their evaluations. Those doctors were also quite skeptical of the hypothesis that the virus could predispose someone to MCI, which is often a precursor to Alzheimer’s disease. Frontera says that “The behavioral neurologists who worked on this project, meaning the people that were interviewing these patients and doing the normal NACC ADRC protocols that they do, were actually very surprised” at the study’s results.
Among the 260 subjects analyzed in the new paper, the incidence of MCI was 27% among those with Long-COVID, 5% in those who’d recovered fully from the virus, and 1% in those who had no evidence of having been infected, just over four years after the pandemic. The team is still following these participants and expects to release a steady stream of results. “This paper is just the beginning, looking at the initial comparisons; we still have repeated measures and other types of data to look at trajectories over time,” says Frontera.
Emerging data support the idea that Long-COVID could result from an overactive immune response, leaky small vessels in the brain, or both. "We're still farther than I would like to be away from understanding what's driving this pathology,” says Frontera.
“Like many physicians, one of our coping strategies is intellectualization, so in the face of all this insane amount of death and horror … we started collecting data”
In their ongoing work, she and her colleagues have already begun trying to figure out how Long-COVID might trigger MCI. Using the same NACC-assessed patient group, they’ve found that phospho-tau 217, a new biomarker for Alzheimer’s disease, is significantly higher in Long-COVID patients, especially those with brain fog. The team is also taking advantage of the long-running nature of the ADRC Program and NACC, to see how their cohort’s condition evolves over time. “Do people recover from this MCI diagnosis, or do they progress towards a dementia diagnosis? If for some reason we see trends towards improvement, which would be great, it would suggest that this is not the same neurodegenerative pathology as [Alzheimer’s disease],” says Frontera.
Frontera JA, Masurkar AV, Betensky RA, Alvarez Z, Boutajangout A, Chodosh J, Hammam S, Hunter J, Jiang L, Li M, Links J, Marsh K, Pang H, Silva F, Thawani S, Vasilchenko D, Vedvyas A, Yakubov A, Ge Y, Wisniewski T. Increased incidence of mild cognitive impairment in long COVID patients. Alzheimers Dement. 2026 Mar;22(3):e71237. doi: 10.1002/alz.71237. PMID: 41772376; PMCID: PMC12953049.