Post-COVID brain fog is real and we still don't have a clean blood test for it, which means most of what gets done with these patients is ruling out the…
Sections
Post-COVID brain fog is real and we still don’t have a clean blood test for it, which means most of what gets done with these patients is ruling out the other stuff that looks exactly like it before landing on the diagnosis by exclusion. That’s not a satisfying answer if you came in wanting a tidy explanation, but it’s the honest one. The fog is real, it’s measurable on neuropsych testing (the multi-hour battery of paper-and-pencil tests that score how your attention, memory, and processing speed are actually performing) in a real chunk of patients, and it usually gets better over twelve to eighteen months without anyone doing anything heroic. The boring nature of that timeline is part of why people don’t believe it… nobody wants to hear “wait a year, also keep showing up for your appointments,” especially after a year of already feeling like garbage.
The complaint is always some version of the same thing. Can’t find words mid-sentence. Walk into a room and forget why. Take three times as long to write the same email you used to bang out in five minutes, and feel completely wiped after a normal day of work that didn’t used to wipe you out at all. Guys describe feeling out of it in a way they didn’t before they got sick, like there’s a thin layer of static between them and the thing they’re trying to think about. They’ll usually have tried the obvious stuff before they show up… extra sleep, less screen time, more coffee, and one of those nootropic stacks their cousin who knows a guy swears by. None of it has moved the needle, which is what got them to actually make the appointment.
What gets ruled out first
Half the patients who come in convinced they have long COVID brain fog have something else, or something else on top of it, that’s much easier to fix. So before signing off on the diagnosis the workup hits thyroid, B12, ferritin, vitamin D, and a metabolic panel. Asking about sleep, because untreated sleep apnea (the condition where your throat collapses during sleep, you stop breathing for short stretches, and your brain spends the whole night low on oxygen without your knowing it) makes you feel exactly like long COVID does, and you can fix it with a CPAP. Asking about how much you’re drinking, because if you’ve been doing four beers a night since the pandemic started to take the edge off, that’s the fog, that’s not COVID.
Depression screen comes next, which presents as cognitive slowing in middle-aged guys way more often than as crying or sadness. If your concentration tanked, you stopped enjoying things you used to, you’re sleeping like hell, and you’ve lost interest in stuff, that’s a depression workup, not a long COVID workup, and those need different things. Same surface story, totally different mechanism, and putting the wrong label on it means a year of treating the wrong problem.
What the data actually shows
Here’s what the actual scans show. Oxford researchers ran brain MRIs on hundreds of UK Biobank adults before and after they caught COVID, with uninfected people scanned the same way for comparison. The infected group lost measurably more grey matter in the memory and smell-related parts of the brain, somewhere between 0.2 and 2 percent extra tissue on top of normal aging, and their whole brain shrank a bit more too (Douaud et al. 2022). They slipped on a timed thinking test too, and here’s the part that should get your attention: it held up even after they threw out everyone who’d been hospitalized. So this isn’t just the people who nearly died, it’s regular mild infections leaving a mark you can see on a scan. The thinking hit shows up in the testing. A study of over 110,000 people in England found the ones whose symptoms dragged on were running about six IQ points behind people who never got infected, and the ones who landed in intensive care were down around nine (Hampshire et al. 2024). Pool the studies on the people whose symptoms actually persist and the cognitive deficit lands in the medium-to-large range, not a rounding error (Sobrino-Relaño et al. 2023). The risk goes up with how sick you got and with each reinfection, which is the case for not catching it over and over if you can help it. Not to be Chicken Little about it, but if you’ve already had it three times and you’re hoping a fourth round is just a regular cold, the math isn’t on your side.
And the mechanism is clearer than it used to be. In both mice and people, even a mild respiratory case set off an inflammatory reaction in the brain’s white matter, killed off the cells that build myelin, and choked the birth of new neurons in the hippocampus, your memory center, the same pattern you see in chemo brain (Fernandez-Castaneda et al. 2022). Layer the older suspects on top, microvascular damage in the tiny blood vessels, persistent low-grade inflammation, and effects on the vagus nerve, and you’ve got a real biological story, not a vibe. What we still don’t have is a clean target to treat, which is why most of the proposed treatments are crap. If somebody’s selling you a supplement that “targets the inflammatory cascade in long COVID,” they’re either a damn liar or talking themselves into something they want to believe.
What recovery actually looks like
There’s a real recovery story here too, and it’s worth saying because the rest of this post leans grim. A lot of people, especially the ones who weren’t hit that hard to begin with, do get noticeably better over a year or two, and some get most of the way back. So if your case is mild, the odds are decent you climb out of it. But it’s not a clean “most people are fine and a tiny few stay stuck.” The people who got hammered, the hospitalized and the ICU crowd, are the ones whose cognition was still measurably down two and three years out, and in that group the depression, anxiety, and fatigue were actually worse at two and three years than they’d been at six months (Taquet et al. 2024). So the honest version is split: milder cases mostly recover on a slow annoying timeline, and the people who were sickest can stay impaired or even slide further. One more reason not to keep rolling the dice on reinfections.
What actually helps
Boring stuff helps. Sleep, real sleep, eight hours, no phone in bed, same time every night. Aerobic exercise, but pacing yourself because post-exertional malaise (the thing where you push too hard one day and pay for it for three) is a real piece of this, and overdoing it sets people back. Cutting alcohol way down or out. Treating any underlying depression or anxiety, because nobody recovers cognitively while they’re depressed, the depression eats whatever cognitive bandwidth would have been doing the healing.
The supplement industry has gone insane on this. Patients show up with twenty-bottle stacks. None of it has data. Save your money. Drink water like you actually like it, you’re gonna need it for the headaches the dehydration is contributing to that nobody pointed out.
Half the patients who come in convinced they have long COVID brain fog have something else, or something else on top of it, that’s much easier to fix.
What it usually turns out to be
The version of this story that comes up the most, picture a guy in his late thirties who had a moderate COVID infection a year or two back and never quite bounced back. Foreman is on him for missing details on jobs he used to nail. He thinks he’s losing his mind and shows up convinced he’s heading toward early dementia. Workup pulls up two things almost every time. Ferritin (the iron storage marker) is low enough to make a healthy guy feel like garbage but not low enough to flag at primary care, and the sleep study he never had shows moderate apnea with an AHI (apnea-hypopnea index, the number of breathing pauses per hour, where 5-15 is mild and 30+ is severe) somewhere in the high teens or twenties. Start iron, start CPAP, lose the four beers a night that built up during the lockdown stretch. Six months later he’s back to baseline, and we never had to call it long COVID at all. Was probably all three things compounding, with COVID as the trigger that tipped him over the edge.
That’s most of these cases. It’s rarely just one thing, and the thing it gets blamed on is rarely the thing that’s actually the biggest piece.
When it really is just long COVID
Some patients, you rule out everything and they still have it. For those guys the focus is on rehab, not cure. Cognitive rehab through occupational therapy (working with an OT on real-world cognitive tasks like organizing a workday, breaking down complex projects into smaller pieces, building external memory aids), graded return to activity (do a little more than yesterday but not enough to crash tomorrow), treating any anxiety that’s piggybacking on the cognitive stuff, and patience. Most patients are better at eighteen months than they were at six. That’s not a fast recovery, but it’s one.
The piece nobody likes is that the treatments that actually have data are the boring ones. Sleep, exercise, alcohol off, treating the mood piece, slow pacing. The interventions that get marketed (HBOT, NAD+ infusions, peptides) don’t have the data to back the price tag. If somebody’s charging you four figures a month and there’s no randomized trial behind it, ask why.
The cognitive piece is real, but so is the depression that gets layered on top
One thing that’s worth saying because most patients aren’t ready to hear it. A year of feeling like your brain doesn’t work makes most people depressed even if they weren’t to start with. The cognitive piece is real, and the secondary depression that comes from watching your job slip, your patience with your kids tank, and your sense of being a competent adult erode is also real, and it makes the cognitive piece worse. Treating just the cognitive piece without addressing the depression piggybacking on it’s why some patients don’t get traction. The standard approach is an SSRI (selective serotonin reuptake inhibitor, the most common class of antidepressants, Lexapro and Zoloft and that crew) at a normal dose, plus the sleep and activity work, and most people start to feel some daylight within a couple months.
The four big mimics
Sleep apnea, depression, alcohol use, and low ferritin or B12. Treating any of these often gets you back to baseline without needing a long COVID label at all. If your wife is sleeping in the guest room because of your snoring, that’s where the workup starts.
Boring stuff, slow recovery
Sleep, pacing, exercise without crashing, alcohol off, treating mood symptoms. Twelve to eighteen months is the typical recovery window. No supplement stack has data. Save the money.
The marketed stuff
HBOT (hyperbaric oxygen), NAD+ infusions, peptide protocols, mega-dose vitamin stacks, and most of what you’ll find on social media. If a long COVID clinic is charging four figures a month and pointing to “improvement reports” instead of randomized trials, ask why.
Bottom line
If you think you have long COVID brain fog, get the full workup before deciding that’s what it is. The patients who actually have it deserve a real diagnosis and not a label that got slapped on because nobody looked hard enough. And if you’re one of the guys who actually has it, the boring stuff is what’s going to get you back to baseline, not the supplements your buddy at the gym is selling. The timeline is slow, the prognosis is mostly good, and the path forward is the unsatisfying one of sleep, pacing, alcohol off, and not chasing the next promised cure. Show up to your own life and let the year do its work.
Sources
- Fanshawe JB, Sargent BF, Badenoch JB, et al. Cognitive domains affected post-COVID-19; a systematic review and meta-analysis. Eur J Neurol. 2025;32(1):e16181. PMID 38375608.
- Taquet M, Skorniewska Z, De Deyn T, et al. Cognitive and psychiatric symptom trajectories 2-3 years after hospital admission for COVID-19: a longitudinal, prospective cohort study in the UK. Lancet Psychiatry. 2024;11(9):696-708. PMID 39096931.
- Davis HE, Assaf GS, McCorkell L, et al. Characterizing long COVID in an international cohort: 7 months of symptoms and their impact. EClinicalMedicine. 2021;38:101019. PMID 34308300.
- Douaud G, Lee S, Alfaro-Almagro F, et al. SARS-CoV-2 is associated with changes in brain structure in UK Biobank. Nature. 2022;604(7907):697-707. PMID 35255491.
- Hampshire A, Azor A, Atchison C, et al. Cognition and Memory after Covid-19 in a Large Community Sample. N Engl J Med. 2024;390(9):806-818. PMID 38416429.
- Sobrino-Relano S, Balboa-Bandeira Y, Pena J, et al. Neuropsychological deficits in patients with persistent COVID-19 symptoms: a systematic review and meta-analysis. Sci Rep. 2023;13(1):10309. PMID 37365191.
- Fernandez-Castaneda A, Lu P, Geraghty AC, et al. Mild respiratory COVID can cause multi-lineage neural cell and myelin dysregulation. Cell. 2022;185(14):2452-2468.e16. PMID 35768006.
