O'Brien & McDougall 2026 — Green light reduces knee OA pain (human crossover)
One-paragraph summary
Human crossover clinical trial of ambient green-light therapy in 19 patients with moderate-to-severe knee osteoarthritis pain. Crossover design: dim white light first (6.57 ± 1.00 lux, 1-2 hours/day × 10 weeks), then a 2-week washout, then dim green light (525 nm wavelength, 6.82 ± 0.78 lux, 1-2 hours/day × 10 weeks). Primary outcome: arthritis disability score via the Western Ontario and McMaster Universities Arthritis Index (WOMAC) questionnaire. Secondary outcomes: pain intensity, pain interference, patient global satisfaction. Result: WOMAC scores decreased significantly with green light (44.1 ± 17.5 baseline → 32.5 ± 16.2); white light produced no significant effect on WOMAC (39.6 ± 15.3). Both white and green light reduced pain intensity, but green light's analgesic effect was significantly greater. Pain interference was reduced with green light only. Patient Global Impression of Change improved with both interventions but more strongly with green. For the project, this is direct human-trial evidence for green-light analgesia in a peripheral-pain condition with rigorous within-patient comparison to a wavelength-matched white-light control. The lead sentence of the abstract is significant for FM specifically: "Regular viewing of dim green light has been shown to reduce the pain associated with migraine, fibromyalgia, and post-surgery" — positioning green light therapy as already-established for FM in the published literature. Closes the v0.3.1 §12.7 green-light arm framing gap, which had been anchored only on the Ferrari 2026 Dahl SS rat finding.
Claims as triples
green_light_exposure — modulates → widespread_pain[evidence: WOMAC reduction 44.1→32.5 in knee OA crossover trial; confidence: emerging]green_light_exposure — bridges → fm_central_only[evidence: abstract's lead sentence confirms FM as established green-light-responsive condition; confidence: bridging]green_light_exposure — modulates → central_sensitization[evidence: pain-interference reduction beyond pain-intensity reduction suggests CNS-modulation component; confidence: emerging]
Methods note
Open-label crossover trial. N=19 with moderate-to-severe knee OA pain. Sequence: 10 weeks dim white light → 2 weeks washout → 10 weeks dim green light. Light specifications: white 6.57 ± 1.00 lux (4000K neutral); green 525 nm wavelength, 6.82 ± 0.78 lux. Daily exposure: 1-2 hours, ambient conditions (room illumination). Primary endpoint: WOMAC arthritis index. Secondary: pain-intensity numerical rating, pain-interference scale, Patient Global Impression of Change. Statistical analysis: within-patient paired comparisons.
Limitations
- Open-label, no blinding — green-light vs. white-light is visually distinguishable; patient expectation effects cannot be excluded.
- N=19 is small for definitive efficacy claim though sufficient for pilot-level effect-size estimation in a crossover design.
- No control group with neither light — both arms received an intervention; the contrast is between two active interventions rather than against no-treatment baseline.
- Knee OA, not FM — translation to FM is by mechanism analogy through green-light's documented FM-responsive effect, which the abstract references but the trial itself doesn't measure.
- Order of crossover not randomized in described design — white light always preceded green light. Carryover or temporal-trend effects cannot be fully excluded despite the 2-week washout.
Open questions raised
- Does green-light analgesia in FM patients show a similar effect size to the WOMAC 11.6-point reduction in this knee OA cohort? An FM-specific replication trial would close the FM-direct evidence gap.
- Does green-light analgesia work synergistically with metformin (per Ferrari 2026 Dahl SS rat finding — both metformin and green light reduce FAM scores) or are they redundant? Combined-arm design could test.
- What's the optimal wavelength + lux + duration regimen for sustained green-light analgesia? The 525 nm / 6-7 lux / 1-2 hr/day regimen here is one point in a parameter space that hasn't been mapped.
Triangulation notes
- Closes the v0.3.1 §12.7 green-light arm framing gap. The arm had been anchored on a single Ferrari 2026 Dahl SS rat finding (no human evidence). This paper provides human-trial evidence in a peripheral-pain condition that is analogous-but-distinct from FM, validating the broader mechanism. The abstract's FM-direct framing inherits established-published-precedent status.
- Compatible with O'Brien et al 2025 PAIN (companion paper, OA rat endocannabinoid mechanism — ingested simultaneously). Both papers from the McDougall lab; together they provide a paired human + rat dossier with mechanism resolution.
- Compatible with the v0.3.1 §12.7 resveratrol arm — both green light and resveratrol are low-barrier, low-cost adjunct interventions. They could be combined in stratified pilot designs.
- Bridges Hypothesis 2's network-FC framework (Hou 2026 thalamocortical decoupling) to a non-pharmacological intervention axis. The Tan et al 2025 visual cortex → LP thalamic circuit paper (ingested simultaneously) provides the candidate neural-circuit mechanism for how visual input modulates pain.
Bridges
- B21 candidate — green light therapy ↔ FM via human peripheral-pain efficacy + endocannabinoid mechanism (paired with the O'Brien 2025 PAIN mechanism paper).
- Reinforces v0.3.1 §12.7 — green light becomes part of the chain-agnostic intervention strategy for the HERV-mitochondrial-inflammation loop framework's downstream symptoms, alongside resveratrol.