Green Light Therapy to Treat Migraine
Evidence Review created on 11/09/2026 using AI4L + evidence-mcp / Opus 5.5
Also known as: Green Light Exposure, Green Light-Emitting Diode Therapy, GLED, Green LED Therapy, Narrow-Band Green Light, NbGL, Green Light Phototherapy, Allay Lamp
Motivation
Green light therapy means spending an hour or two in a dim room lit only by a lamp that gives off a narrow band of green light. It draws interest in migraine because sensitivity to light is one of the condition’s most disabling features, and researchers have asked whether one color of light might be tolerated, or even soothing, rather than painful. The proposed mechanism is that this color sends weaker signals from the eye to the brain areas that handle head pain.
Migraine is among the most common and disabling nerve-related conditions of working age, and many people stop preventive drugs because of side effects or cost. Interest in green light grew from laboratory work at Harvard on how light worsens migraine pain, followed by small clinical studies in the United States and Pakistan, research on green-tinted room lighting, and a lamp sold directly to consumers.
This review examines the human evidence on green light for preventing migraine and for easing attacks, the proposed biology, the risks, the protocols used in studies, and the financial interests of the researchers who produced most of the data.
Benefits - Risks - Protocol - Conclusion
Recommended Reading
Expert commentary and narrative reviews giving a high-level overview of green light therapy for migraine.
-
Green Light Therapy and Migraines: Will Research Translate to Real-World Use? - Heidi Moawad
A neurologist’s appraisal of the early Harvard studies, explaining why green light aggravates migraine less than other colors and what practical uses might follow.
-
How to Stop Headaches Using Science-Based Approaches - Andrew Huberman
A neurobiologist’s headache episode; its photophobia (painful light sensitivity) segment explains how melanopsin (a light-sensing eye pigment) retinal cells relay blue and green light to thalamic (sensory relay) pain pathways, favoring dim red light.
-
Shining a light on migraine relief - Harvard Health Publishing
Summarizes the 2016 color-and-migraine study (Noseda 2016) by the group of Allay stock-option holder Rami Burstein (Lipton 2023), quoting neurologist Carolyn Bernstein and noting that green light did not cure the headache.
-
Light Therapy in Chronic Migraine - Hou et al., 2024
Narrative review of green light and other light-based treatments, such as infrared light and low-level laser, for chronic migraine, covering efficacy, sleep, anxiety and safety.
-
Neurobiology of Photophobia - Burstein et al., 2019
Mechanistic review from Burstein’s group, which developed narrow-band green light (Burstein holds Allay stock options), tracing how retinal signals reach thalamic pain neurons and why color matters in migraine light sensitivity.
No content on green light for migraine was found from Rhonda Patrick, Peter Attia, Chris Kresser, Life Extension Magazine or Lifespan.io; their sites and web searches returned only general light-and-sleep material or unrelated items.
Grokipedia
Green light therapy for migraine
Summarizes the Harvard color studies, the 2023 uncontrolled lamp study by Allay-linked authors (Lipton 2023) and home devices, and notes that the lack of controlled trials hiding the treatment from patients limits conclusions.
Examine
No Examine article on green light therapy exists. Examine.com’s search for green light therapy returns only green tea, green coffee and other unrelated entries.
ConsumerLab
No ConsumerLab article on green light therapy exists. ConsumerLab mainly tests supplements and foods; its search returns product reviews and answers on other light therapies (light boxes, red and near-infrared light), none on green light.
Systematic Reviews
Systematic reviews that include the human studies of green light therapy in migraine.
-
Green Light Therapy for Pain and Pain-Related Psychosocial Outcomes: A Systematic Review of Human Studies - Bain et al., 2026
Thirteen human studies (668 participants), including migraine trials; certainty rated very low for every outcome; no pooled analysis because studies differed too much.
-
Interventions for Migraine and Sleep: A Systematic Review Exploring Their Bidirectional Association - Sforza et al., 2026
Reviews migraine treatments affecting sleep; includes the green light trial by patent applicants (Martin 2021), summarizing fewer headache days and improved sleep onset and maintenance.
No systematic review or meta-analysis addresses the harms of green light exposure; harm data come only from individual studies.
Mechanism of Action
Green light therapy acts through the eyes, not the skin. Migraine pain travels through the trigeminovascular pathway (the nerve network carrying pain signals from the brain’s coverings) to the thalamus (a relay station for sensory signals), and light can worsen it because retinal cells, mainly intrinsically photosensitive retinal ganglion cells (ipRGCs, light-sensing nerve cells containing the pigment melanopsin), contact these thalamic pain neurons (Noseda 2010).
- Cone pathway: Narrow-band green light (about 520–525 nm, nanometers of wavelength) mainly stimulates medium-wavelength cones (the retina’s green-sensitive color receptors). Green activated cone-driven retinal pathways less than white, blue and red (Noseda 2016), and thalamic neurons in rats were least responsive to green (Noseda 2016).
- Body’s own opioids: In rats, green LED (light-emitting diode) light raised pain thresholds (Ibrahim 2017); opaque contact lenses prevented this (Ibrahim 2017), naloxone (an opioid-blocking drug) reversed it (Ibrahim 2017), and spinal enkephalins (natural opioid molecules) likely mediated it (Ibrahim 2017); lead author Mohab Ibrahim has a related patent pending (Martin 2021).
- Two receptor routes: A color-blind man reported lower headache intensity without fewer headache days, which the authors, including patent holder Ibrahim, read as ipRGCs driving intensity and cones driving frequency (Cheng 2022).
- Competing explanations: Participant expectation and placebo responses may explain the positive open-label (color known to participants) findings (Bain 2026); a systematic review called for sham-controlled trials (comparing against a fake treatment) with blinded outcome assessment (assessors unaware of the treatment) to separate these (Bain 2026). Rats see green poorly, limiting rodent data.
Historical Context & Evolution
Colored light was promoted as “chromotherapy” in the 19th century without controlled testing, and modern light therapy began with bright white light for seasonal depression in the 1980s. Green light for migraine came from photophobia research (light sensitivity that worsens pain). In 2010, Rami Burstein’s Harvard group showed that light worsens headache even in blind people who retain non-image light detection (Noseda 2010). In 2016 the group reported that green light exacerbated migraine headache significantly less than white, blue, amber or red light (Noseda 2016).
Mohab Ibrahim’s University of Arizona group found green LED analgesia (pain relief) in rats (Ibrahim 2017) and then ran a preventive migraine trial (Martin 2021); Ibrahim and a co-author have a patent pending through the University of Arizona for green light therapy in chronic pain (Martin 2021). Burstein holds stock options in AllayLamp, the company selling a narrow-band green lamp (Lipton 2023), and an author of its user study is an Allay employee (Lipton 2023).
Later work added a randomized trial in Pakistan comparing green light with brain stimulation (Mahmood 2025) and Arizona State University studies of green-enriched white lighting (Sharp 2026), two of whose authors have held equity in Allevalux, a lighting company (Sharp 2026). Supporters describe a low-cost, low-risk option; critics point to small, unblinded studies led by inventors. A 2026 systematic review rated the certainty of evidence very low for all outcomes (Bain 2026). Neither position has yet been tested in a sham-controlled trial, so the question remains open.
Expected Benefits
High 🟩 🟩 🟩
No benefit reaches High: the controlled migraine data come from one non-randomized crossover trial (each participant received both lights in turn) by a single group (Martin 2021), and the only randomized preventive trial had no inert light comparator (Mahmood 2025).
Medium 🟩 🟩
Fewer Monthly Headache Days
Used daily for prevention, green light was followed by fewer headache days. In a University of Arizona one-way crossover (10 weeks of white light, then 10 of green; 7 episodic and 22 chronic migraine patients, meaning under 15 or at least 15 headache days a month), 86% of episodic and 63% of chronic patients had over 50% fewer headache days during green exposure (Martin 2021). A Pakistani randomized trial also reported fewer attacks with green light, but compared it only with brain stimulation arms (Mahmood 2025).
Magnitude: Green phase: 18.4 to 7.4 headache days per month; white phase: 18.2 to 16.5 days; not compared between groups (Martin 2021).
Lower Headache Pain Intensity
Pain intensity of the headache phase, rated on the NPS (numeric pain scale, 0–10), fell markedly during green exposure and only slightly during white exposure in the Arizona crossover (Martin 2021). In the Pakistani trial, the green light group showed a significant decrease in pain severity at every time point (Mahmood 2025), but pain also fell in the sham stimulation group (Mahmood 2025). Only the Arizona trial compared green with another light, and neither concealed the treatment color.
Magnitude: Headache-phase NPS 8.0 to 3.2 during green versus 8.5 to 7.9 during white exposure; not compared between groups (Martin 2021).
Less Migraine-Related Disability
HIT-6 (Headache Impact Test-6, a validated questionnaire on how headaches limit daily life) scores fell significantly during green exposure in all Arizona subgroups, while white exposure did not improve them (Martin 2021). In the Pakistani trial, green light improved migraine-specific quality of life slightly more than active brain stimulation (Mahmood 2025). Neither trial used a blinded, inert light comparator.
Magnitude: HIT-6 fell 12.85 points during green versus 1.14 points during white exposure in the trial’s 55 chronic pain participants, not only those with migraine; not compared between groups (NCT03677206).
Less Light-Triggered Discomfort During Attacks
Green light provokes less photophobia than other colors. In testing by the Harvard group, at the highest intensity nearly 80% of patients had worse headache with every color except green, which affected half that proportion (Noseda 2016). An Arizona State University randomized crossover during untreated attacks found that green-enriched white light caused the least visual discomfort (Sharp 2026); two of its authors have held equity in Allevalux, a lighting company (Sharp 2026). Because that group tested green-enriched white light rather than narrow-band green, replication is indirect.
Magnitude: Visual discomfort 4.78 with green-enriched versus 7.58 with blue-enriched white light on a 0–10 scale during attacks (Sharp 2026).
Better Sleep ⭕️ Not Central to Treat Migraine
This item bears on sleep quality rather than headache. In the Arizona crossover, green exposure significantly improved the ability to fall and stay asleep in all migraine groups, while white exposure improved it only modestly (Martin 2021). Lamp users reported better sleep after about half of treated attacks (Lipton 2023), and a color-blind patient reported better sleep after the Arizona protocol (Cheng 2022). The controlled data come from one trial whose patients knew the light color.
Magnitude: Sleep improved in 49% of treated attacks; no control group (Lipton 2023). The Arizona trial reports no sleep figure in its text or abstract (Martin 2021).
Low 🟩
Acute Relief When Used During Attacks ⚠️ Conflicted
In an uncontrolled online study of lamp buyers, headache improved in just over half of attacks treated with 2 hours of narrow-band green light (Lipton 2023). A randomized crossover found green-enriched white light did not significantly lower headache intensity (Sharp 2026). Net: acute benefit is unconfirmed by controlled testing.
Magnitude: Headache improved in 55% of 3,232 treated attacks; no control group (Lipton 2023).
Less Acute Medication Use
Most Pakistani trial participants used paracetamol at baseline (Mahmood 2025); use fell to none after 4 weeks of green light (Mahmood 2025). It fell with active and sham brain stimulation too (Mahmood 2025), so a color-specific effect is unproven. The Arizona trial could not assess this (Martin 2021).
Magnitude: Paracetamol use fell to no medication after 4 weeks in all three arms, including sham stimulation; the literature reports no outcome figure for the green light arm alone (Mahmood 2025).
Calmer Mood During Attacks ⭕️ Not Central to Treat Migraine
This item bears on anxiety rather than headache. Lamp users reported less anxiety after a third of treated attacks (Lipton 2023). The data come from an uncontrolled online diary, so expectation and rest in a dark room cannot be separated from a color effect.
Magnitude: Anxiety improved in 34% of treated attacks; no control group (Lipton 2023).
Speculative 🟨
Lasting Effect After Exposure Stops
In rats, green LED antinociception (a reduced pain response) persisted for 4 days after the last exposure without tolerance (Ibrahim 2017). The basis is animal data only; no human study has measured carry-over.
Benefit-Modifying Factors
- Color vision genes: Red-green color deficiency, inherited through the X-linked OPN1LW and OPN1MW genes (which encode the red and green cone pigments), may blunt the frequency benefit; a color-blind patient gained pain relief without fewer headache days (Cheng 2022).
- Baseline headache frequency: Responder rates were 86% in episodic and 63% in chronic migraine (Martin 2021); green light improved McGill pain-quality scores in chronic but not episodic patients (Martin 2021).
- Baseline light sensitivity: The Pakistani trial excluded people with photophobia (Mahmood 2025), whereas the acute studies tested it directly (Noseda 2016) or tracked it as an outcome (Lipton 2023), so effects in strongly light-sensitive patients rest mainly on uncontrolled data.
- Biomarkers: No blood or imaging biomarker predicts response; the baseline headache-day count from a diary is the only measured predictor.
- Sex: All but two of the 29 Arizona participants were women (Martin 2021); in rats with knee arthritis, green light analgesia came sooner and was larger in males (Ventura 2024).
- Age: Arizona participants were 24 to 72 years old (Martin 2021); age-related lens yellowing and smaller pupils reduce light reaching the retina, so older adults may need the lamp nearer within the tested range.
- Health conditions: Medication-overuse headache (rebound headache from frequent pain-reliever use), cataract or retinal disease could reduce response by sustaining pain or blocking light; none has been studied.
Potential Risks & Side Effects
High 🟥 🟥 🟥
No risk reaches High: harm data in green light trials consist of statements that no side effects occurred and one registry report of zero adverse events (NCT03677206), and no trial was designed or powered (large enough to detect a real effect) to detect harms.
Medium 🟥 🟥
Headache Worsening or Triggering ⚠️ Conflicted
Green light can itself aggravate migraine. In testing by patent holder Burstein’s Harvard group, green light exacerbated headache in 40% of patients during attacks (Nir 2018) and triggered it in 3% between attacks (Nir 2018); the share rose with intensity (Noseda 2016). Yet no side effects were reported at 4–100 lux (lux measures brightness at the eye) (Martin 2021), and green-enriched white light did not raise headache intensity in a randomized crossover (Sharp 2026). Net: worsening appears tied to brightness and attack state, and low-intensity use has no reported harm.
Magnitude: Headache worsened in 40% of patients with green versus about 80% with white, blue, amber or red light during attacks (Nir 2018); triggered in 3% with green versus 16–19% with other colors between attacks (Nir 2018); adverse events in 0 of 55 participants with green and 0 of 55 with white light through study completion (NCT03677206).
Low 🟥
Evening Body-Clock Disruption ⚠️ Conflicted
Green light at 2,000 lux, far above therapy levels, reduced nighttime melatonin (sleep-timing hormone) output in sleep-deprived adults (Horne 1991), indirect evidence by dose for a circadian (body clock) effect. Migraine users instead reported better sleep (Lipton 2023). Net: disruption is unlikely at therapy brightness.
Magnitude: At 2,000 lux, green light suppressed melatonin output from midnight to 6 am and raised it from 6 am to noon, with no change in 24-hour output; the literature reports no outcome figure, and none at therapy brightness (Horne 1991).
Speculative 🟨
Delayed Effective Treatment
Relying on a lamp could postpone proven preventive or acute drugs while migraine worsens. The basis is mechanistic reasoning only; no study has measured this.
Flicker From Low-Cost Light Strips
Low-cost dimmable LEDs often flicker invisibly, and flicker can provoke migraine or photosensitive seizures (seizures triggered by flashing light). No report links green therapy lamps to such events; the basis is mechanistic.
Retinal Light Injury
Retinal light damage is greatest with blue wavelengths, and green light at 4–100 lux is far dimmer than daylight. No injury has been reported; the concern is mechanistic only.
Risk-Modifying Factors
- Genetic factors: Photosensitive epilepsy, often familial, raises concern with flickering lamps; no gene variant is known to change tolerance of green light, and color vision deficiency has no reported safety effect.
- Baseline state: No laboratory marker predicts harm; the relevant baseline is attack state, since white, blue, amber and red light exacerbated headache in about 80% of patients during attacks (Nir 2018), whereas light initiated headache in only 16–19% between attacks (Nir 2018).
- Sex: No sex differences in harms have been reported; the Arizona trial enrolled almost only women (Martin 2021).
- Health conditions: Epilepsy, bipolar disorder (bright light can trigger mania; inferred at higher intensities), delayed sleep timing and retinal disease warrant caution; none has been studied with green light.
- Age: Older adults face fall risk when moving in a dark room and more often have cataract or retinal disease; the Arizona trial included people up to 72 years old (Martin 2021).
Key Interactions & Contraindications
Documented interactions:
- Verteporfin (Visudyne, a light-activated eye drug): After injection, skin and eyes should be kept from direct sunlight or bright indoor light for 5 days (Visudyne label); the consequence is a light-induced burn. Avoid: postponing light sessions until the window ends removes the risk.
- Porfimer (Photofrin, a light-activated cancer drug): After injection, skin and eyes should be kept from direct sunlight or bright indoor light for at least 30 days (Photofrin label); consequence: a light-induced burn. Avoid: postponing light sessions until the window ends removes the risk.
Theoretical interactions (inferred from mechanism, not studied in humans):
- St. John’s wort (a supplement containing the light-activated pigment hypericin): Inferred from light activation of hypericin; consequence is phototoxicity (light-triggered skin or eye damage). Therapy lamps are dim, but pausing sessions while taking it removes the concern.
- Migraine preventive drugs (propranolol, topiramate, erenumab): An additive reduction in headache days is inferred from different mechanisms; consequence is greater benefit but unclear attribution. Changing one treatment at a time keeps effects interpretable.
- Acute pain relievers (ibuprofen, naproxen, acetaminophen, sumatriptan): No pharmacological interaction is expected; delaying them for a 2-hour session weakens relief because they work best early. Taking rescue medication at the usual time avoids this.
- Melatonin: Evening light is inferred to counter melatonin’s timing signal through melanopsin; consequence is later sleep onset. Ending sessions 2 hours before bed avoids it.
- Migraine supplements (magnesium, riboflavin, coenzyme Q10): An additive preventive effect is inferred from separate mechanisms; consequence is unclear attribution of benefit. Starting them at a different time from the lamp clarifies response.
- Tinted lenses (FL-41 rose tint, blue-blocking glasses): Inferred to filter part of the green band; consequence is a lower effective dose. Removing them during sessions avoids this.
- Transcranial direct current stimulation (tDCS, a weak electrical current through the scalp): An additive effect is inferred from separate pain pathways; consequence is unclear attribution. Combined use is under study (NCT06943625).
Populations who should avoid Green Light Therapy:
- People whose headache worsens under green light during attacks, as occurred in 40% of tested patients (Nir 2018).
- People within 5 days after verteporfin injection, when bright indoor light and prolonged light from light-emitting medical devices are to be avoided (Visudyne label).
- People within at least 30 days after porfimer injection, up to 90 days in some patients (Photofrin label).
- People with photosensitive epilepsy using flickering or color-cycling lamps: inferred from flicker-triggered seizures; no data with green therapy lamps.
- People with a new, sudden or changing headache not yet evaluated: inferred from the risk of masking a secondary cause such as bleeding or tumor; no data.
Risk Mitigation Strategies
Doses and timings below follow common practice unless cited.
- Low brightness: Keeping light at the eye at 4–100 lux, the trial range, by placing the lamp 1–2 m away (Martin 2021) limits the headache worsening seen at higher intensities.
- Stop on worsening: Ending a session if headache, nausea or light discomfort increases, and stopping use after three such sessions, limits green-light-triggered headache.
- Daytime sessions: Scheduling sessions in the daytime or early evening, ending at least 2 hours before bed, limits circadian disruption.
- Flicker-free source: A lamp with a flicker-free driver, checked with a phone camera’s slow-motion mode, prevents flicker-related migraine or seizures.
- Keep proven treatment: Continuing prescribed preventive and acute medication, as the Arizona trial allowed (Martin 2021), prevents delayed effective treatment.
- Safe dark room: Clearing walking paths and keeping a dim phone light nearby prevents falls during 1–2 hour dark-room sessions.
- Headache diary: Recording headache days and acute medication days for 4 weeks before starting and throughout use detects worsening and medication overuse early.
- Photosensitizer check: Reviewing medications and pausing sessions for 5 days after verteporfin (Visudyne label) limits light-induced burns; after porfimer, skin and eyes stay sensitive to bright indoor light for at least 30 days, sometimes 90 or more (Photofrin label).
Therapeutic Protocol
Parameters below without a citation (timing, sessions per week, titration) reflect common practice.
- Preventive protocol (Mohab Ibrahim, University of Arizona): A 525 ± 10 nm green LED strip at 4–100 lux, used 1–2 hours daily in a dark room for 10 weeks (Martin 2021).
- Acute protocol (Rami Burstein, Harvard, Allay Lamp): A 520 ± 10 nm lamp at 5–10 cd/m² (candela per square meter, a measure of surface brightness), used for 2 hours during an attack with all other lights off (Lipton 2023).
- Short daily protocol (Riphah International University): A 525 ± 10 nm lamp for 1 hour a day, 5 days a week, for 4 weeks alongside prescribed medication (Mahmood 2025).
- Alternative delivery: Green-filtering eyeglasses, which allow moving about, were feasible in a fibromyalgia (chronic widespread pain) pilot (Nelli 2023); green-enriched white light at low intensity was the most visually acceptable option for everyday settings (Fani 2026).
- Single session: The Arizona protocol used one session at the same time each day, all at once rather than split, with the light in view, no staring and no sleeping (Martin 2021).
- Time of day: The Pakistani trial used daytime sessions (Mahmood 2025); evening use is common, but ending well before bedtime limits circadian effects.
- Half-life: Light has no half-life; in rats the effect persisted 4 days after exposure ended (Ibrahim 2017), but human carry-over is unmeasured.
- Genetics: A color vision screen (Ishihara plates, a standard color test) sets expectations, since a color-blind patient gained pain relief without fewer headache days (Cheng 2022).
- Sex: No sex-specific dosing exists; the Arizona data come almost entirely from women (Martin 2021).
- Age: Older adults may need the lamp nearer, within the tested brightness range, because age-related lens yellowing and smaller pupils reduce retinal light.
- Baseline biomarkers: No laboratory marker guides dosing; baseline headache days from a 4-week diary define the starting point and the benchmark.
- Health conditions: People with chronic migraine may need the full 10-week course to judge response; people with epilepsy need flicker-free sources.
Discontinuation & Cycling
- Duration: Preventive courses ran 10 weeks (Martin 2021) and 4 weeks (Mahmood 2025); acute use is as needed. Whether lifelong daily use is required is unknown.
- Withdrawal: No withdrawal effects have been reported; trials did not follow patients after stopping.
- Tapering: No taper is needed; reducing session frequency stepwise shows whether benefit persists at a lower time cost.
- Cycling: No evidence supports cycling; in rats the effect developed without tolerance (Ibrahim 2017).
- Rebound check: A 4-week diary after stopping shows whether headache days return toward baseline.
Sourcing and Quality
- Wavelength: Studies used narrow bands peaking at 525 nm (Martin 2021) and 520 nm (Lipton 2023), each ±10 nm; green party bulbs and color-changing strips often emit broader, unverified spectra.
- Brightness control: A dimmer reaching very low output matters; the Allay Lamp emits 1–10 lux (NCT04841083).
- Flicker: Flicker-free drivers are preferable; pulse-width dimming in budget strips can flicker.
- Brands: The Allay Lamp is the main commercial narrow-band device; its study’s senior author holds stock options in it (Lipton 2023). Arizona State University researchers have held equity in Allevalux, a lighting company (Sharp 2026).
- Third-party testing: No independent body tests green therapy lamps; a manufacturer’s spectral data sheet or a spectrometer reading is the main check.
Practical Considerations
- Time to effect: Acute sessions last 2 hours (Lipton 2023); in the Pakistani trial pain fell by 2 weeks (Mahmood 2025), while the Arizona trial judged prevention over 10 weeks (Martin 2021).
- Common pitfalls: Using the lamp in a lit room, choosing too bright a setting, staring at the source, falling asleep, or stopping effective medication.
- Regulatory status: The Allay Lamp is sold as a consumer product, not a medical device, and does not require a 510(k) (the standard U.S. Food and Drug Administration device clearance) (NCT04841083).
- Cost and access: Lamps and LED strips are inexpensive and sold without prescription; the main cost is 1–2 hours a day in a dark room.
- Funding asymmetry: Low-margin light devices draw little industry funding compared with costly migraine drugs. Payers would save if a low-cost lamp replaced costly preventives, yet insurers have not funded trials; this is a potential structural bias in research funding and guideline evidence.
Interaction with Foundational Habits
- Sleep: Potentiating when timed in the daytime; lamp users reported better sleep after attacks (Lipton 2023), but evening light may delay melatonin through melanopsin. Ending sessions 2 hours before bed and keeping regular sleep times, a core migraine-prevention habit, aligns both.
- Nutrition: None direct: light does not affect nutrient status or digestion. Regular meals and hydration remain separate trigger-management habits, and no food needs to be avoided around sessions.
- Exercise: None direct; the Arizona protocol encouraged activities such as exercise during sessions as long as the light stayed in view (Martin 2021). Stationary cycling or stretching fits a 1–2 hour session.
- Stress management: Potentiating; lamp users reported less anxiety after a third of attacks (Lipton 2023). Sessions pair naturally with breathing exercises, meditation or relaxing audio, since screens and other light sources must stay off.
Monitoring Protocol & Defining Success
Before starting, a 4-week headache diary establishes the baseline: headache days, attack pain intensity, acute medication days and light sensitivity. A HIT-6 questionnaire records disability, a color vision screen sets expectations, and a medication review identifies light-activated drugs. A phone lux meter reading at eye level confirms the lamp sits within the tested brightness range.
Ongoing monitoring follows the diary daily, with summary reviews at 2 weeks, 4 weeks and 10 weeks, then every 3 months. The Arizona protocol ran 10 weeks (Martin 2021), making the 10-week point the natural point to judge whether headache days have fallen by half. The lux reading is repeated after any change in lamp position or device.
| Biomarker | Optimal Functional Range | Why Measure It? | Context/Notes |
|---|---|---|---|
| Monthly headache days | No established target; track change from own 4-week baseline | Expected to change (main benefit) | The Arizona trial counted a headache day as moderate to severe pain lasting at least 4 hours and a responder as over 50% fewer days (Martin 2021) |
| Attack pain intensity (NPS, 0–10) | No established target; track average attack score against baseline | Expected to change | Rate at peak pain, using the same scale each time |
| HIT-6 score | No established target; track change from baseline | Expected to change (disability) | Six-item questionnaire; completed monthly |
| Acute medication days per month | No established target; track against baseline | Safety check: rising use signals medication-overuse headache and changes management | Paired with headache days in the same diary |
| Light at the eye (lux) | No established target; record the reading and keep it stable | Safety check: brighter light worsens headache in some patients | The Arizona trial used 4–100 lux (Martin 2021); measured at eye level with a lux meter app |
Qualitative markers:
- Sleep quality: ease of falling and staying asleep after sessions
- Calm during attacks: anxiety and irritability with the lamp on
- Light sensitivity: tolerance of ordinary room and screen light
- Daily function: ability to work and complete planned activities
- Adherence: sessions completed per week and their length
Emerging Research
- Combined green light and tDCS (NCT06943625): Riphah International University randomized trial, 40 planned patients, headache diary and pain outcomes at 4 weeks. Estimated primary completion (March 2026) has passed; the record still lists recruiting and no posted results. A positive result would support green light as an add-on; a null result would weaken it.
- Workplace lighting study (NCT07476053): An Arizona State University single-group study in 20 planned adults with migraine, comparing monthly migraine days under baseline and optimized white workplace lighting; recruiting, completion expected March 2027 (NCT07476053). A positive result would support everyday lighting; a null result would confine interest to dedicated sessions.
- Central mechanism trial (NCT05569486): A University of Arizona randomized study in 70 planned adults with fibromyalgia, measuring brain glial (support cell) activation after green light; not yet recruiting, completion expected December 2029 (NCT05569486). A positive result would anchor a brain mechanism; a null result would leave expectation as a leading explanation.
- Arizona chronic pain crossover (NCT03677206): The migraine trial was registered here (Martin 2021), a white-then-green crossover in 55 chronic pain patients (NCT03677206); active, not recruiting, completion December 2026 (NCT03677206). HIT-6 fell 12.85 points with green, 1.14 with white (NCT03677206), without adverse events (NCT03677206). A positive report would support prevention; a null would weaken it.
- Completed lighting crossover (NCT07898722): An Arizona State University randomized crossover in 40 participants comparing optimized with conventional white light for visual discomfort completed in October 2025, with no results posted (NCT07898722).
- Sham-controlled trials: A 2026 systematic review called for adequately powered, pre-registered (methods published in advance), sham-controlled trials with blinded outcome assessment (Bain 2026); such a trial would be the first to separate a color effect from expectation.
- Evidence that could weaken the case: Lenses blocking the wavelengths that stimulate ipRGCs did not significantly change two-hour pain versus control lenses in 78 subjects (Posternack 2023), and green-enriched light did not lower headache intensity (Sharp 2026).
- Pathway studies: A color-blind patient’s response suggests separate ipRGC and cone roles (Cheng 2022); testing people with defined color deficiencies could show which pathway lowers headache frequency.
Conclusion
Green light therapy for migraine is a low-cost, drug-free approach: sitting for one to two hours in a dark room lit only by a narrow band of green light, either daily to prevent attacks or during an attack to ease it. The biological case is plausible. Green appears to send weaker signals from the eye to the brain’s head-pain relay than other colors, and animal work points to the body’s own pain-relieving chemicals.
The human evidence is encouraging but thin. One small trial where everyone used white light first, and one trial assigning patients by chance without an inactive comparison light, both reported fewer and milder headaches and less disability, and a study of lamp buyers with no comparison group reported relief during many attacks, along with calmer mood and better sleep. A test comparing green-tinted white light with other colors during attacks found less discomfort but no clear fall in pain. Reviewers rate the overall certainty as very low.
Harms appear minor at low brightness. Brighter green light can still worsen headache in a sizeable minority during attacks, and evening sessions could, in principle, disturb sleep timing.
Much of the evidence comes from researchers with a stake in the outcome: a pending patent, stock options in the main commercial lamp, and equity in a lighting company. For a person willing to spend time on a low-cost, low-risk method alongside existing treatment, green light sits as a plausible but unproven option whose true size of effect remains unknown.