--- canonical_name: Green Light Therapy alternate_names: Narrow-Band Green Light, NbGL, Green LED Therapy, GLED, Green Light Exposure, 525 nm Light Therapy canonical_topic: Green Light Therapy to Treat Migraine short_topic_lc: green_light_therapy_migraine creation_date: 2026-0824-1946 creator_ai_fullname: Grok 4 ep_keywords: Phototherapy, Light Therapy --- # Green Light Therapy to Treat Migraine

Evidence Review created on 08/24/2026 using [AI4L](https://github.com/forever-healthy/AI4L) / Grok 4 **Also known as:** Narrow-Band Green Light, NbGL, Green LED Therapy, GLED, Green Light Exposure, 525 nm Light Therapy ## Motivation Migraine is a recurrent neurological attack that combines head pain with light sensitivity, nausea, and loss of work, training, and sleep capacity. Green light therapy is time spent in a dim room lit only by a narrow band of green light, with other lamps and screens off. It is not a drug, not a laser, and not ordinary bright-light treatment for winter depression. Interest grew after laboratory work showed that this green band stirs the eye-to-pain pathway less than blue, red, or white light, and after small clinic studies reported fewer headache days and milder pain. Commercial lamps and tinted glasses now sell the same idea, while some clinicians instead use dim red or orange light for light pain. The groups behind the main human studies hold patents or equity in related products. This review examines whether green light therapy reduces migraine burden in health-conscious adults, how the proposed visual and opioid mechanisms sit with conflicting advice about colored light, what is known about harms, and how protocols, devices, and financial interests shape the evidence. **[Benefits](#expected-benefits) - [Risks](#potential-risks--side-effects) - [Protocol](#therapeutic-protocol) - [Conclusion](#conclusion)** ## Recommended Reading High-level interviews, narrative papers, and primary studies that introduce green light therapy and the visual pathways it is meant to spare. - [Going Green (Light) for Migraine and Pain: A Conversation with Mohab Ibrahim](https://migrainecollaborative.org/going-green-light-for-migraine-and-pain-a-conversation-with-mohab-ibrahim) - Freda Kreier Interview with the patent-holding investigator behind the 2021 migraine crossover, covering discovery, animal work, and remaining skepticism. - [Light Therapy in Chronic Migraine](https://pubmed.ncbi.nlm.nih.gov/38865075/) - Hou et al., 2024 Independent narrative review of light-based treatments for chronic migraine, placing green light beside other phototherapies and stressing small samples. - [Migraine photophobia originating in cone-driven retinal pathways](https://pubmed.ncbi.nlm.nih.gov/27190022/) - Noseda et al., 2016 2016 human and animal study: narrow-band green was the least aggravating color during attacks. A co-author later took equity in a green lamp. - [Evaluating the Potential of Green Light Exposure on Nociception-A Mini Review](https://pubmed.ncbi.nlm.nih.gov/37221686/) - Qaiser et al., 2024 Short review of green light for nociception (pain sensing), including migraine, summarizing animal and early human work without pooling new trials. - [Spectral Composition of Ambient White Light Modulates Photophobia and Headache Intensity During Migraine Attacks](https://pubmed.ncbi.nlm.nih.gov/42323789/) - Sharp et al., 2026 Attack-time crossover: green-boosted white light, acting on the same cone-driven photophobia (painful light sensitivity) pathways as narrow-band green, produced the least visual discomfort among tailored whites. Priority-expert platforms (FoundMyFitness, Peter Attia, Huberman Lab, Chris Kresser, Life Extension, Lifespan.io) were searched; none had a dedicated green-light-therapy-for-migraine overview. Huberman Lab discusses photophobia (painful light sensitivity) but favors dim red or orange light rather than green. ## Grokipedia - [Green light therapy for migraine](https://grokipedia.com/page/Green_light_therapy_for_migraine) Dedicated Grokipedia entry covering wavelength, Burstein and Ibrahim lines of work, commercial lamps, and the limits of unblinded human data. ## Examine No Examine.com article on green light therapy was found. ## ConsumerLab No ConsumerLab article on green light therapy was found. ## Systematic Reviews PubMed systematic reviews and meta-analyses that evaluate green light therapy or include it among migraine interventions. - [Green Light Therapy for Pain and Pain-Related Psychosocial Outcomes: A Systematic Review of Human Studies](https://pubmed.ncbi.nlm.nih.gov/42513623/) - Bain et al., 2026 Thirteen human studies (nine randomized; 668 people), including migraine; favorable pain signals, certainty graded very low. - [Interventions for Migraine and Sleep: A Systematic Review Exploring Their Bidirectional Association](https://pubmed.ncbi.nlm.nih.gov/41874004/) - Sforza et al., 2026 Sleep-and-migraine review that lists green light among non-drug methods reported to improve both headache and sleep. No systematic review dedicated to harms of green light therapy was found; Bain et al. address safety only in passing. ## Mechanism of Action Green light therapy acts through the eyes. In rats, opaque contact lenses abolish analgesia, and a [colorblind patient with chronic headache still improved](https://pubmed.ncbi.nlm.nih.gov/36159182/), so image-forming cones are not required. Two human-facing accounts compete. The first, from recordings during migraine attacks, is that a narrow green band (about 530 nm) drives the smallest signals in cone pathways, in thalamic neurons that also receive meningeal pain input, and in the cortex. That weaker drive is offered as the reason green exacerbates photophobia less than blue, red, amber, or white, and why low-intensity green can slightly reduce pain. Intrinsically photosensitive retinal ganglion cells (ipRGCs, brightness-sensing retinal neurons using the pigment melanopsin, most sensitive near 480 nm) sit on a related pathway; some lens products try to block ipRGC-stimulating wavelengths rather than add green. The second account is active analgesia. [Rodent work](https://pubmed.ncbi.nlm.nih.gov/34157406/) shows green light raises spinal β-endorphin and enkephalin (the body's opioid peptides), needs mu- and delta-opioid receptors (the main opioid docking sites in pain circuits), and is reversed by naloxone (an opioid-blocking drug). Reduced N-type calcium-channel (CaV2.2) activity and lower neuroinflammation are additional animal findings. These accounts are not reconciled. The intervention is not a drug: it has no plasma half-life, CYP enzyme (drug-metabolizing liver enzyme) metabolism, or tissue-distribution profile. ## Historical Context & Evolution Colored light as a migraine tool is recent. Photophobia has been a defining migraine feature for centuries, and people have long retreated to dark rooms. In 2016, Rodrigo Noseda, Rami Burstein, and colleagues at Beth Israel Deaconess Medical Center [reported](https://pubmed.ncbi.nlm.nih.gov/27190022/) that, during attacks, a narrow green band was the least aggravating color and, at low intensity, could reduce pain; they traced smaller retinal and thalamic signals. Burstein later co-founded a company selling a lamp built around that band and holds stock options in it. Independently, Mohab Ibrahim at the University of Arizona followed a personal observation—headache easing outdoors under trees—into [rodent experiments published in 2017](https://pubmed.ncbi.nlm.nih.gov/28092651/) showing lasting pain reduction from 525 nm light via the visual system and opioid circuits. His group then ran a 2021 human crossover in people with migraine who had failed many drugs. Ibrahim, Amol Patwardhan, Rajesh Khanna, and Frank Porreca report patents on the method and ownership in a start-up that licensed the University of Arizona technology. Opinion shifted from jokes about "pharmacologists playing with light," as Ibrahim later described, to a small clinical literature and commercial lamps and glasses. Replication remains limited, and the same investigators who generated the core human data have financial stakes in the products those data support. Current use is still an unapproved consumer practice, not a settled medical standard. ## Expected Benefits ### High 🟩 🟩 🟩 #### Less photophobia under dim narrow-band green Photophobia is the painful sense that light makes an attack worse. In attack-time color tests, a narrow green band intensified head pain in a smaller share of people than other colors, and low-intensity green sometimes reduced pain. Booth tests last minutes, not weeks. Rami Burstein, an author on the 2016 color work, took stock options in a commercial green lamp. **Magnitude:** At higher intensities, green still worsened headache in about 40% of people versus about 80% for other colors ([Nir et al., 2018](https://pubmed.ncbi.nlm.nih.gov/29905657/)); at low intensity, green reduced pain by about 15% ([Noseda et al., 2016](https://pubmed.ncbi.nlm.nih.gov/27190022/)). #### Lower attack-phase pain with repeated exposure Daily 525 nm use for 10 weeks lowered attack-phase pain in an unblinded one-way crossover of people with hard-to-treat migraine. A 2025 randomized comparison also reported lower pain scores in a small unblinded sample that excluded photophobic people. Authors of the 2021 crossover hold patents and a company stake in the method. **Magnitude:** After 10 weeks of daily 525 nm light, mean attack pain fell from about 8.0 to 3.2 ([Martin et al., 2021](https://pubmed.ncbi.nlm.nih.gov/32903062/)); the 2025 trial reported lower scores without a published point estimate ([Mahmood et al., 2025](https://pubmed.ncbi.nlm.nih.gov/41300216/)). #### Fewer monthly headache days with daily exposure A 10-week one-way crossover in 29 people with episodic or chronic migraine who had failed multiple drugs used 525 nm light-emitting diode (LED) strips at 4–100 lux (dim indoor light) for 1–2 hours daily in a dark room. Headache days fell sharply after green, not after the preceding white-light weeks. A 2025 trial also reported fewer attacks with green light but excluded photophobic people. Authors of the 2021 crossover hold patents and a company stake in the method. **Magnitude:** About a 60% drop in monthly headache days versus baseline in the 2021 cohort (86% of episodic and 63% of chronic participants had a ≥50% reduction) ([Martin et al., 2021](https://pubmed.ncbi.nlm.nih.gov/32903062/)). #### Better day-to-day function and quality of life The same 2021 crossover reported large improvements on the Headache Impact Test-6 (HIT-6, a six-item disability scale), the Short-Form McGill Pain Questionnaire, and EuroQol-5D (EQ-5D, a five-dimension health-status scale), plus better self-rated sleep, chores, exercise, and work. A 2025 randomized comparison found green light improved migraine-specific quality of life, with a slightly larger quality-of-life shift than transcranial direct current stimulation (tDCS, weak scalp electrical stimulation). These are secondary outcomes in small unblinded studies. **Magnitude:** HIT-6 and EQ-5D improved with green versus white light in the 2021 chronic-migraine cohort, and the Migraine-Specific Quality of Life questionnaire improved in the 2025 trial; those papers report significance without a published point estimate for the scale scores ([Martin et al., 2021](https://pubmed.ncbi.nlm.nih.gov/32903062/); [Mahmood et al., 2025](https://pubmed.ncbi.nlm.nih.gov/41300216/)). ### Medium 🟩 🟩 #### Shorter headache-phase duration with daily exposure The 2021 one-way crossover asked people to rate whether the headache phase felt shorter after weeks of 525 nm light. Authors reported improvement in that secondary duration rating, and noted that a shorter attack could also turn a diary "headache day" into a non-day. No stopwatch or hour-count was published. Unblinded self-ratings in one small cohort are the only human duration data. **Magnitude:** Duration was reported as improved after green LED versus the preceding white-light weeks; the paper gives no mean hours-per-attack figure ([Martin et al., 2021](https://pubmed.ncbi.nlm.nih.gov/32903062/)). ### Low 🟩 #### Better sleep during or after exposure People in the 2021 preventive crossover and in an open-label lamp study during attacks reported easier sleep. Diaries were unblinded, and sleep gain may simply follow less pain. No sleep-lab (polysomnography) data exist for this protocol. **Magnitude:** Sleep was rated improved in 49% of 3,232 treated attacks among purchasers of a commercial 520 nm lamp; a co-author was employed by the manufacturer and another held stock options ([Lipton et al., 2023](https://pubmed.ncbi.nlm.nih.gov/37859647/)). #### Less anxiety during attacks The same open-label lamp study, run among purchasers of a commercial 520 nm lamp with a co-author employed by the manufacturer and another holding stock options, reported less attack-time anxiety. There was no contemporaneous control, so expectation is a plausible driver. **Magnitude:** Anxiety was rated improved in 34% of recorded attacks ([Lipton et al., 2023](https://pubmed.ncbi.nlm.nih.gov/37859647/)). ### Speculative 🟨 #### Lasting analgesia through spinal opioid peptides Rodent work shows days-long pain reduction after green LED, reversed by naloxone and needing mu- and delta-opioid receptors. No human migraine trial has measured this; the basis is animal-only. ## Benefit-Modifying Factors - **Attack phase versus interval:** Color preference is stronger during attacks than between them; green is still less likely than other colors to *start* a headache in the interval, but the soothing signal is mainly ictal (during the attack). - **Intensity and spectrum:** Benefit in the color studies appears at low intensity with a narrow 515–535 nm band. Bright office-level green, wide-band "green-looking" bulbs, and screens do not reproduce the lab stimulus. - **Baseline severity and prior treatment failure:** The 2021 crossover enrolled people with high baseline pain who had failed oral drugs and botulinum toxin; that is a refractory (hard-to-treat) group, not mild infrequent migraine. - **Episodic versus chronic pattern:** Both patterns improved in 2021; the absolute drop in headache days was larger in chronic migraine because the starting count was higher. - **Sex:** Migraine is several-fold more common in women of reproductive age. Green-light trials have not reported sex-stratified efficacy, so a sex-specific modifier remains unmeasured. - **Age:** Trials enrolled adults, typically 18–80. Older adults with cataracts or macular disease may receive a different retinal dose; that has not been titrated. - **Cone function:** A colorblind patient still improved, which argues that cone mosaics are not the main modifier; ipRGC and thalamic routing may matter more. - **Genetics:** No pharmacogenetic (gene–dose) variants are established for this intervention. - **Baseline biomarkers:** No circulating lab marker of green-light engagement is known; expected gain tracks baseline headache-day count and HIT-6, not a blood or urine value. - **Concurrent drugs:** People stayed on existing preventives in 2021; whether CGRP (calcitonin gene-related peptide, a migraine-pain messenger) monoclonal antibodies blunt or add to the light signal is unknown. ## Potential Risks & Side Effects ### High 🟥 🟥 🟥 #### Headache or photophobia intensification when the light is too bright or not narrow-band The same visual pathway that makes dim green relatively tolerable still transmits pain when intensity rises or when blue and red leak into the beam. In color-by-intensity testing, green remained capable of worsening headache, just less often than other colors. Home "green" bulbs, RGB (red-green-blue mixed) strips, and bright decorative lamps are not the 520–530 nm, 4–100 lux laboratory stimulus. This is a human symptom endpoint shown in more than one psychophysical study. **Magnitude:** Green intensified headache in about 40% of people during attacks and triggered headache in about 3% between attacks, versus about 80% and 16–19% for other colors ([Nir et al., 2018](https://pubmed.ncbi.nlm.nih.gov/29905657/); [Noseda et al., 2016](https://pubmed.ncbi.nlm.nih.gov/27190022/)). ### Medium 🟥 🟥 No other risk reaches Medium: replicated human harms besides intensity-dependent photophobia are not reported; trial adverse-event logs are small and unblinded rather than a second clinical-harm endpoint. ### Low 🟥 ### Speculative 🟨 #### Evening circadian stimulation Melanopsin is most sensitive near 480 nm, not 525 nm, but green is not circadian-inert. No migraine green-light trial has measured melatonin or clock-timed rest–activity; the concern is mechanistic only. #### Flicker-triggered seizure in photosensitive epilepsy Pattern-sensitive epilepsy can be provoked by flashing light. No seizure has been reported in the migraine series; flicker quality of consumer strips is unregulated, so the concern is mechanistic only. ## Risk-Modifying Factors - **Intensity:** Risk of intensification rises with lux; the 2021 protocol stayed in a dim 4–100 lux band, whereas office-level green is the condition that still hurts in color tests. - **Bandwidth:** Wide-band green, mixed RGB, and screens reintroduce blue and red that drive larger retinal and cortical signals. - **Baseline photophobia:** People whose attacks are already light-locked may not tolerate even dim green; the 2025 tDCS-comparison trial excluded photophobic individuals entirely. - **Ocular disease:** Cataracts, macular degeneration, recent eye surgery, and uncorrected refractive error change retinal dose; older adults are more likely to sit in this group. - **Sex:** No sex-specific adverse-event split has been published. - **Genetics:** No variants (for example OPN4, the melanopsin gene) are known to modify harm. - **Baseline biomarkers:** No blood or urine marker predicts intensification; a higher baseline photophobia rating is the practical harm modifier. - **Photosensitizing drugs:** Systemic drugs that raise light injury risk (some tetracyclines, amiodarone, certain antipsychotics) have not been studied with this lamp protocol. - **Photosensitive epilepsy and migraine-with-aura visual symptoms:** Flicker and high contrast are the relevant modifiers, not the green hue as such. ## Key Interactions & Contraindications - **Prescription migraine preventives (caution, additive uncertainty):** The 2021 crossover kept existing drugs, including prior CGRP-pathway failures. No pharmacokinetic clash is expected; additive benefit is untested. - **Acute migraine drugs (monitor):** Triptans (serotonin-receptor acute migraine drugs), gepants (CGRP-receptor acute migraine drugs), and NSAIDs (non-steroidal anti-inflammatory drugs) were not restricted. Light does not replace a time-critical acute dose when an attack is escalating. - **Photosensitizing drugs (caution, phototoxic / light-injury risk):** Doxycycline, demeclocycline, amiodarone, chlorpromazine, and similar agents raise light-injury risk in bright sunlight. The therapeutic lamp is dim; extra-bright improvised setups are the concern. - **Opioid analgesics (theoretical, likely minor):** Animal analgesia is naloxone-reversible. No human interaction with prescribed opioids is documented; the consequence would be a change in pain scores, not respiratory depression from the light. - **Over-the-counter (OTC) sleep aids and melatonin (monitor if evening use):** Combined evening green light plus melatonin has not been studied; competing circadian cues are the issue. - **Magnesium, riboflavin, coenzyme Q10 (additive, untested):** These nutraceutical preventives do not share a visual mechanism; stacking is common in this audience and has no known harm signal with the lamp. - **tDCS and other neuromodulation (additive in one small trial):** A 2025 comparison found both tDCS and green light active; a recruiting combination trial is testing synergy. Skin burns are a tDCS risk, not a light risk. - **Tinted ipRGC-blocking glasses versus adding green (competing, not synergistic by design):** Precision tints that cut melanopsin-stimulating wavelengths are a different product class; wearing them under a green lamp has no tested protocol. **Populations who should avoid Green Light Therapy:** - Uncontrolled photosensitive epilepsy, especially with a history of flicker- or pattern-triggered seizures. - Active untreated retinal disease, recent intraocular surgery, or ophthalmologist-imposed light restriction. - People who already find dim narrow-band green painful on a brief test exposure (intensification, not soothing). - None identified as an absolute systemic contraindication in pregnancy; the intervention is optical, but pregnancy-specific trials do not exist. ## Risk Mitigation Strategies - **Stay in the dim, narrow band:** Keep 515–535 nm and about 4–100 lux in a dark room; this is the setting that avoided adverse events in 2021 and is what blunts intensification. - **Ban competing light:** Phones, televisions, and overhead LEDs reintroduce blue/red and undo the spectral filter; they are the usual way people accidentally recreate the High-intensity risk. - **Eyes open, no sleep:** The 2021 protocol required the strip in the visual field without falling asleep, so the visual-pathway dose is not lost and an unattended strip cannot recreate the High-level intensification risk. - **Stop-rule for intensification:** If headache or photophobia rises within a session, end the session and drop intensity or abandon that device — this directly limits the High-level harm. - **Flicker-aware hardware:** Prefer specified narrow-band lamps or strips rated flicker-free if photosensitive epilepsy or aura with visual features is in the history. - **Evening timing cap:** Finish sessions well before intended sleep if sleep onset is fragile, to limit the circadian concern. ## Therapeutic Protocol - **University of Arizona preventive:** 525 ± 10 nm LED strip, 4–100 lux, 1–2 hours daily in a dark room for 10 weeks ([Martin et al., 2021](https://pubmed.ncbi.nlm.nih.gov/32903062/)). - **Burstein / Allay acute (during attacks):** 520 ± 10 nm at 5–10 candela/m² (a very dim surface-brightness unit), about 2 hours during the attack; 61% were ≥50% attack responders in an [open-label purchaser study](https://pubmed.ncbi.nlm.nih.gov/37859647/). - **Green-enriched white (Sharp / Schwedt, competing):** Spectrally tuned white, green-boosted, used as ordinary room light during attacks; more practical than monochromatic green, still an experimental indoor spectrum. - **ipRGC-blocking tints (Avulux, competing):** Spectacles that cut melanopsin-stimulating wavelengths and transmit more green; a [2023 randomized trial](https://pubmed.ncbi.nlm.nih.gov/37150129/) missed its primary pain endpoint. First author was an Avulux employee. - **Dim red or orange light (Huberman-style, competing):** A photophobia strategy that treats short-wavelength light, including green, as a trigger rather than a treatment; it is the opposite spectral bet. - **Time of day:** Trials did not lock clock time. Afternoon sessions avoid the evening circadian concern; attack-time use follows symptoms, not the clock. - **Not a split-dose drug:** There is no plasma half-life. "Dose" is minutes of visual exposure; one continuous 1–2 hour block is what the preventive and acute protocols used. - **Genetics:** No OPN4 or opioid-receptor genotype is used to pick a protocol. - **Sex:** No sex-specific duration or lux has been validated; women of reproductive age dominate migraine samples but not protocol math. - **Age:** Adult protocols (18–80 in the Arizona pain/light trial). Cataracts in older adults may warrant a brighter strip only under eye-care input, which trials did not test. - **Baseline pain:** Arizona migraine entry required high average pain (about ≥5/10) and prior treatment failure; milder, infrequent migraine is an extrapolation. - **Pre-existing photophobia:** Severe light pain is both the target symptom and a reason some trials excluded people; a short test session is the practical filter. ## Discontinuation & Cycling - **Duration of use:** Preventive protocols were 10 weeks of daily exposure, not a declared lifelong drug. 28 of 29 participants in 2021 chose to keep the strip when the study ended, which is an adherence signal, not a taper rule. - **Washout:** Animal analgesia lasted about 4 days after green LED stopped. Human washout was a 2-week white-to-green gap in 2021; headache-day rebound timing was not the primary endpoint. - **Withdrawal:** No autonomic withdrawal syndrome is described. Stopping removes a possible analgesic cue; it does not create opioid-withdrawal physiology. - **Taper:** Not applicable. There is no receptor downregulation protocol; people stop or skip days without a milligram ladder. - **Cycling:** No tolerance was reported in the 10-week human run or in the rodent work that explicitly looked for it. Cycling is not part of published protocols. - **Attack-only use:** The Allay-style 2-hour ictal session is inherently intermittent; it does not require a preventive calendar. ## Sourcing and Quality - **Wavelength, not "looks green":** Target peak 515–535 nm (studies used 520–525 ± 10 nm). Decorative bulbs and RGB strips that appear green usually leak blue and red and recreate the High-level intensification risk. - **Intensity control:** Dimmable output in the 4–100 lux (preventive) or very low candela/m² (acute lamp) range. A cheap lux meter is more informative than marketing "brightness" claims. - **Narrow band and flicker:** Prefer specified narrow-band LEDs and flicker-free drivers. ConsumerLab does not test these lamps; third-party spectral plots from the maker, or a handheld spectrometer, are the available checks. - **Commercial lamps versus strips:** The Allay lamp (Burstein-linked; historically about $90–$200) is built around 520 nm. The 2021 trial used commercially sourced 525 nm LED strips, which are cheaper if the spectrum is real. - **Glasses are a different product:** ipRGC-blocking tints are not equivalent to sitting in a green field; they are a competing optical filter with mixed trial results and manufacturer authorship. - **Not an FDA (U.S. Food and Drug Administration)-approved migraine drug or device:** Allay has been sold as a consumer lamp, not a cleared medical device for migraine. "Wellness" labeling does not certify spectrum. ## Practical Considerations - **Time to effect:** Preventive headache-day change was measured over 10 weeks. Acute color and lamp studies report minutes-to-hours during a single attack. This is not a same-morning preventive. - **Common pitfalls:** Using any green bulb; leaving screens on; falling asleep; sitting too close to a bright strip; treating the lamp as a replacement abortive; expecting Allay-type results from a $10 decorative light. - **Regulatory status:** Unapproved for migraine by the U.S. Food and Drug Administration. Sold as consumer lighting. The 2021 trial was registered as supportive care, not as a licensed indication. - **Cost and time:** A verified 525 nm strip can be inexpensive; branded lamps cost more. The scarce resource is 1–2 hours daily in a dark room without phones — a real protocol cost for people who already stack training, work, and sleep. - **Payer incentives:** Insurers typically reimburse migraine drugs, not consumer lamps, which can tilt guideline attention and research funding toward pharmaceuticals over a cheap optical protocol. ## Interaction with Foundational Habits - **Sleep:** Direction is mixed-potentiating. Unblinded diaries report better sleep after exposure, likely via less pain; evening sessions could still delay melatonin because green is not circadian-zero. Practical point: schedule preventive sessions in the afternoon. - **Nutrition:** None identified. No nutrient depletion, no food-timing interaction, no dietary co-factor in published protocols. - **Exercise:** Indirectly potentiating. The 2021 cohort reported better ability to exercise as headache days fell; some protocols allow reading or light movement in the green-lit room. It does not replace training, and it does not sit on a hypertrophy pathway. - **Stress management:** Indirectly potentiating. Attack-time anxiety fell in the open-label lamp study, and a forced dim-room hour is a structured pause. That is rest and reduced sensory load, not a measured cortisol (stress-hormone) protocol. ## Monitoring Protocol & Defining Success Baseline is a 4-week headache diary plus HIT-6 and a 0–10 numeric pain rating, taken before the first green session, with usual preventives held stable. That package is what the 2021 crossover and open-label lamp study actually used; no blood, urine, or imaging marker of green-light engagement exists in people with migraine. Ongoing checks repeat the diary and HIT-6 at 2 weeks (tolerability and intensification stop-rule), at 4 weeks, at 10 weeks (the length of the Arizona exposure block), then every 3 months if daily use continues. Success is a ≥50% drop in monthly headache days or a HIT-6 move of at least 6 points (a commonly cited minimum important difference), without new photophobia from the device. Failure to move those numbers by 10 weeks is a stopping cue, not a reason to raise lux into the intensification range. | Biomarker | Optimal Functional Range | Why Measure It? | Context/Notes | |-----------|--------------------------|-----------------|---------------| | Monthly headache days | ≥50% reduction from the person's own baseline; many longevity-oriented users track toward <4 days/month | Primary endpoint in the 2021 crossover | Conventional "success" in migraine trials is often a 50% responder rate, not a population mean. Diary daily; do not rely on memory. | | HIT-6 score | ≥6-point drop; functional target often <50 (little-to-no impact) | Captures disability, not just day counts | Conventional bands: ≤49 little impact, 50–55 some, 56–59 substantial, ≥60 severe. Not a blood test; complete on a typical day, not only during an attack. | | Pain intensity (0–10 numeric rating) | Drop of ≥2 points or ≥30% from baseline attack/average pain | Secondary in 2021 (mean 8.0 to 3.2) | Rate the same time of day. Office-level green that *raises* this score is a device failure, not a reason to push brighter. | | Photophobia rating (0–10 during a standard indoor light) | Stable or falling versus baseline; a rise during a green session is a stop-rule | Directly tracks the High-level harm | No lab assay. Pair with a note of lux and whether screens were off. | Qualitative markers (logged daily, not lab values): - Ability to work, train, and do chores during the week - Sleep onset and night awakenings after session days - Whether other lights or screens leaked into the session - Rescue-drug counts (not a target to drop blindly) ## Emerging Research - **Combination with tDCS (recruiting):** [NCT06943625](https://clinicaltrials.gov/study/NCT06943625) randomizes about 40 people with migraine to tDCS, green light (520 nm, 100 lux, 2 hours, 3–5 days/week), both, or medication only over 4 weeks; a positive combination arm would support stacking, a null would weaken add-on claims. - **Central mechanism trial (not yet recruiting):** [NCT05569486](https://clinicaltrials.gov/study/NCT05569486) plans PET (positron emission tomography, a brain-inflammation scan) and spinal fluid in fibromyalgia under green versus white light. A null glial-inflammation finding would weaken the neuroinflammation story that is often exported to migraine. - **Light-driven analgesia circuits (recruiting):** [NCT07245303](https://clinicaltrials.gov/study/NCT07245303) tests S-cone-modulating visual stimuli and pain circuits. Results could support or undercut the idea that cone/ipRGC routing is the right therapeutic target. - **Green-enriched white as daily lighting:** [Sharp et al., 2026](https://pubmed.ncbi.nlm.nih.gov/42323789/) and [Fani et al., 2026](https://pubmed.ncbi.nlm.nih.gov/41983992/) found tuned white can cut attack-time discomfort without monochromatic green. Home replication would loosen the dark-room protocol. - **Certainty is still very low:** [Bain et al., 2026](https://pubmed.ncbi.nlm.nih.gov/42513623/) graded green-light pain evidence as very low. A large blinded, sham-spectrum trial could shrink the apparent 60% headache-day effect to placebo-sized, which would weaken preventive use. - **Lens trial miss:** [Posternack et al., 2023](https://pubmed.ncbi.nlm.nih.gov/37150129/) did not meet its primary pain endpoint for ipRGC-blocking spectacles (manufacturer-authored). That already weakens the glasses-as-equivalent-to-green-light claim. ## Conclusion Green light therapy is daily or attack-time exposure to a narrow slice of green light, usually near 520–525 nanometers, at low brightness in a dim room. The method is of interest to people who already treat migraine as a long-horizon health constraint: attacks steal sleep, training, and work capacity, and many preventive drugs fail or are poorly tolerated. The case for benefit rests on reduced light pain, lower attack intensity, fewer headache days, and better function in small human studies. Those signals are not matched by large studies that hid the light color. A systematic review of green light for pain judged that evidence as very weak. Preventive data come from a small study in which people knew they were using green light, led by investigators with patents and a company stake. Acute-lamp data come from purchasers of a commercial lamp whose founder holds stock options in the manufacturer. Harms in the protocol look limited: participants reported no serious adverse events. The main documented downside is the opposite of the intended effect—brighter or off-band light, including some "green" sources, can still worsen head pain and light sensitivity in a large share of people with migraine. Time in a dim room is a real cost. For a health-optimizing adult already willing to run inconvenient protocols, the intervention is a low-systemic-risk add-on whose size of benefit remains uncertain and whose research base is financially entangled with the products under study. **[Top](#top) - [Benefits](#expected-benefits) - [Risks](#potential-risks--side-effects) - [Protocol](#therapeutic-protocol)**