Ten hours of cognitive rehab lifted daily function in long COVID, where no treatment is proven

The phrase people use is brain fog, which sounds mild and is not. In long COVID it means losing the thread of a sentence you started, reading a paragraph four times, and finding that the job you did for fifteen years has become unmanageable.

Cognitive impairment is common in long COVID and severely affects daily life, with no proven treatments to date. That last clause is the state of play Vanova and colleagues set out to change, writing in JAMA Network Open, with a trial across three sites in England.

Their intervention was ten hours. Goal attainment was significantly greater in the CR compared with the TAU group at three months, by an amount that would be unusual for any behavioral treatment. Until now there had been nothing with evidence behind it.

What was actually delivered

Cognitive rehabilitation is easy to picture wrongly. It is not brain training, and there were no puzzles.

The program consisted of 10 individual 1-hour sessions conducted once per week with a trained researcher, applying evidence-based strategies to 3 individually selected, personally meaningful functional goals. The participant names the problem, which might be following a work meeting, or cooking a meal without abandoning it halfway, and the sessions build workarounds for that specific thing: external memory aids, restructured routines, pacing, breaking a task into pieces that survive an interruption.

The premise is compensation rather than repair. Nobody claims the injured processing is restored. The claim is that a person can be taught to get the day done anyway.

Reading a personalized outcome honestly

The primary outcome was goal attainment on the Bangor Goal-Setting Interview, and its strengths and weaknesses are the same fact.

Because the goals are the participant’s own, the measure tracks what actually matters to them, which a standard cognitive test does not. Because the goals are the participant’s own, one person’s score is not the same quantity as another’s, and the measure is self-reported by someone who knows which group they are in.

The trial was single-blind: the assessors were masked, the participants could not be. For a subjective primary outcome in a condition with no treatment and a great deal of hope attached, that is the central limitation, and it is not a small one.

Why the effect size invites caution rather than excitement

The three-month difference came with a large and clinically meaningful treatment effect, at a Cohen d of 1.57.

Effect sizes of that magnitude are rare outside pharmacology. By convention 0.8 counts as large, and behavioral interventions in chronic conditions typically land between 0.2 and 0.5. A 1.57 in a trial of 78 people is the shape of result that most often shrinks when a bigger study repeats it.

The six-month figure is the useful corrective. This was sustained at 6 months, with a lower effect size: 1.72, and a Cohen d of 0.91. Smaller, still large, and more believable for being smaller. Benefit that decays partially over three months is what a real compensatory strategy looks like as people stop practising it.

The sample, and what it cost

A sample size of 88 participants was required to detect a conservative effect of 0.7 on the goal attainment score at 3 months. They randomized 78.

Under-recruitment usually means a trial is underpowered to find its effect. Here the effect found was so much larger than the one planned for that the shortfall did not bury it. That is fortunate rather than reassuring: a trial that undershoots its target and then reports an effect twice the size it was designed to detect is exactly the profile that replication tends to trim.

The comparison arm deserves credit, though. Treatment as usual was variable, with most participants having access to specialist memory clinics, so this was not measured against nothing.

What it means if you have this

Long COVID may affect multiple organs in the body, such as the lungs, kidneys, heart, brain, and skin, and the cognitive symptoms are among the ones people most often describe as having taken their life away.

The finding here is narrow and real: ten weekly sessions of goal-focused strategy work produced large, partially sustained improvement in what people could get done. It says nothing about whether the underlying impairment healed, and the trial did not test that.

It is also not the ordinary forgetting of getting older. Older adults may worry about their memory and other thinking abilities, such as taking longer to learn something new, and these changes are usually signs of mild forgetfulness that are often a normal part of aging. Everyone in this trial had measured impairment, at least one standard deviation below their age norm across two or more domains. Whether services exist to deliver ten hours of anything is the question this result now puts to health systems.