According to the latest data, more than a third of the world’s COVID survivors – 36% – struggle with long COVID. And the mental fog that comes with it can be just as debilitating as the initial infection. Trouble concentrating, memory lapses, sluggish thinking remain the most common cognitive complaints that can haunt patients for years, threatening their ability to work, maintain relationships and live independently.
So when the National Institutes of Health launched RECOVER, its large research program on long COVID, one question kept cropping up. Can anything help?
Now, a major new clinical trial suggests the answer (at least for now) is: not yet.
A First-of-Its-Kind National Test
In a five-arm randomized study published in JAMA Neurology, researchers report that three of the most promising rehabilitative approaches – computerized cognitive training, a structured cognitive-behavioral rehabilitation program, and at-home brain stimulation – failed to outperform simple online puzzles and games. All of the participants improved modestly over time, regardless of the treatment they received. But, frustratingly, no specific intervention stood out above the rest.
The trial, part of the RECOVER-NEURO initiative, spanned nearly two dozen clinical sites across the United States and enrolled 328 adults with cognitive symptoms that persisted at least 12 weeks after a SARS-CoV-2 infection. Women made up most of the participants (74%). And the median age was 48. All of the participants reported daily cognitive difficulties despite no significant preexisting neurological disorders.
Researchers evaluated three active treatments:
- BrainHQ: A commercial “interactive, online brain training program that targets memory, attention, and brain speed.”
- PASC-Cognitive Recovery (PASC-CoRE): A structured rehabilitation program tailored to long COVID patients.
- Transcranial Direct Current Stimulation (tDCS): A noninvasive “brain stimulation method … used to modulate cortical excitability, producing facilitatory or inhibitory effects upon a variety of behaviors.”
The research team compared each of these against one of two controls: unstructured online puzzles or sham brain stimulation. The team administered all interventions remotely, five days a week for 10 weeks. The authors reported strong adherence. More than 90% of participants completed at least 80% of the assigned sessions.
No Standout Treatment. But Some Improvements.
The researchers established a relatively straightforward primary outcome. Participants rated their own cognitive functioning using the modified Everyday Cognition Scale (ECog2).
Across all treatment groups, including the puzzle-based control, scores improved by roughly the same amount after 10 weeks.
- BrainHQ vs. puzzles: No difference.
- PASC-CoRE and BrainHQ vs. puzzles: 0.1 difference.
- Active tDCS and BrainHQ vs. sham tDCS and BrainHQ: No difference.
- PASC-CoRE and BrainHQ vs. BrainHQ alone: 0.1 difference.
Obviously, none of these differences appeared to be statistically significant. Secondary outcomes, such as anxiety, depression, and fatigue, followed the same pattern: modest improvement across the board, but no clear winner. Objective neuropsychological testing told a similar story.
Notably, the length of time someone had been living with long COVID mattered more than the treatment they received. Across the full sample, people with more recent infections tended to improve the most, regardless of treatment. Those further out from their initial infection enjoyed more modest gains.
What Should Patients Make of This?
For the growing number of survivors desperate for a reliable cognitive treatment, the findings can feel a bit disheartening. But the researchers stress another way of looking at it. It’s still early days. And this trial lays some crucial groundwork.
The results reinforce what we already know about the complexity of long COVID. More than half of participants reported no measurable cognitive deficits on standardized tests despite reporting substantial daily problems. It’s a mismatch that shows up repeatedly in long COVID studies.
And the authors suggest that these cognitive symptoms might stem from intertwined biological, psychological, and fatigue-related factors that don’t respond predictably to traditional cognitive training.
The Road Ahead
The researchers also insist that this study doesn’t dismiss rehabilitation entirely. It simply clarifies what doesn’t meaningfully move the needle – at least not on its own. And not for the population that these researchers considered.
Other trials are ongoing, including pharmacologic studies and more tailored behavioral interventions. The authors highlight that the active comparator – a schedule of puzzles and games – might itself have offered structure, engagement, and a sense of agency, all things that can help people feel more focused in their daily lives.
For now at least, the takeaway offers a mix of sobriety and hope. Long COVID cognitive symptoms are real, measurable, and remain largely untreatable. But people do manage to improve, and well-designed trials like this one can help pave the way to a long-term solution.
Further Reading
Burnout in Psychiatric Doctors Before and After the COVID-19 Pandemic
