Baricitinib - Reddit anecdote analysis
I used an API to search 3.3 million posts and comments on 14 LC/ME subreddits for every mention of baricitinib, and ran each anecdote through Claude Code with a structured prompt.
Of 38 reports:
- 64% reported a benefit (very effective 10, somewhat 13)
- 25% saw no effect (9)
- 11% got worse (somewhat 1, much 3)
Keep in mind there is a MASSIVE selection bias here, as neutral and even negative responders are less likely to report. The actual outcomes are not this good.
Time to effect
Of 23 positive responders, 11 gave a specific onset time from first dose: 6 within 2 weeks, 5 at 2-6 weeks, 12 didnt say. Four felt something in the first few days, a couple even on day one.
Testing
A chunk of people ran inflammatory labs before and/or after: cytokine panels, interferon, IncellDx immune panels, IL-6/IL-8, ANA.
- The 10 who tested before responded better: 77% got better and zero reported no effect, vs 59% better / 33% no-effect among those who didn't mention testing.
- But testing did not prevent bad reactions. 2 of the 4 who got worse had tested beforehand, one even guided by labs. Though the serious reactions (a hospitalization with electrolyte/kidney issues MCAS flare, a Lyme reactivation, early flares) looked idiosyncratic, not something the panels would have predicted.
Dose
4 mg/day was by far the most common target. Several titrated up from 1-2 mg, usually for tolerability rather than because more worked better. A few held 2 mg (sometimes dose-reduced for kidney function) and still improved. Dose did not appear to affect onset.
What improved (for responders)?
Energy/fatigue 43%, brain fog 39%, PEM 30%, pain 30%, overall/back toward baseline 17%, exercise tolerance 13%, POTS 9%, and sleep/mood/MCAS/gut around 4% each.
Side effects mentioned
Infections (URIs, UTIs, cold sores) 7%, acne/rash 5%, GI upset 5%, headache 5%, MCAS flare 2%. Only 4 of 36 worsened - see the hospitalization, Lyme reactivation, and hard early flares above.
Caveats
- This is self-reported, uncontrolled, selection-biased data. People who improve are really motivated to post, especially about an exotic drug, and Long COVID fluctuates on its own, so some positive responses are noise.
- Many were on concurrent treatments (SCIG, other immune meds, big supplement stacks), so attribution to baricitinib alone is tougher with a non-immediate onset.
Thinking about what to analyze next: maraviroc? Stellate ganglion block? Open to requests!