Red Light Therapy for Parkinson's Disease: Scientific Evidence

Researchers in Sydney, Australia, treated 7 people with Parkinson's disease using near-infrared light applied only to the abdomen and neck, in a 2022 observational study. Each received 894 to 914 nm light in contact with the skin at 30 to 120 mW, 47 mW/cm², 11.3 J/cm², and 50 Hz for 60 seconds to 2 minutes per session, 3 times a week for 12 weeks in clinic and then at home, across 315 days in total. Mobility, thinking, dynamic balance, handwriting, and sense of smell all improved, though the pattern differed from person to person. Some of those gains were lost during the home treatment period, which overlapped with COVID-19 lockdowns, while others were still present at 45 weeks.

A waitlist design gave 6 of 12 participants light therapy straight away and 6 after a 14-week delay, in a 2021 randomized controlled trial at a university in Sydney, Australia. All received 810 and 904 nm light to the head, the nose, the neck and the abdomen at 15, 39.6 and 240 J across 12 weeks, then continued treating themselves at home for up to a year. Mobility, thinking, dynamic balance and fine hand movement all improved significantly with treatment, and many of the individual gains passed the level judged meaningful for the people taking part. Those gains continued for as long as treatment continued, in a disease where decline is normally expected, and no side effects were seen.

18 people with Parkinson's disease aged 30 to 80 at a hospital in New Taipei City, Taiwan, took part in a 2021 clinical trial of light therapy given together with hydrogen water. Each received 930 to 950 nm light close to the skin at the back of the neck at 240 mW, 5.4 to 6.6 mW/cm², and 42 J for 30 minutes per session, every day across 14 days. Scores on a standard disease-severity rating scale began to decrease significantly in the first week and remained lower through the end of treatment. A week after treatment stopped, scores rose slightly but remained significantly better than before treatment, and no adverse events were recorded.

A 2020 review from a university in Singapore examined near-infrared light as a way to protect the nerve cells lost in Parkinson's disease. It reported that light between 620 and 880 nm acts on the energy-producing parts of cells, with 670 and 830 nm producing the strongest response in nerve cells, and that no acute adverse effects had been reported at doses of 4 to 30 J/cm² and 5 to 50 mW/cm². The review stressed that this effect runs in both directions, so a dose below that range does nothing and a dose above it becomes harmful. Light was found to lose 65% of its strength for every millimeter it travels into brain tissue, which the review identified as the main obstacle to reaching the deep region where the disease begins. It pointed to 2 small human studies of light applied outside the head, 1 reporting better movement scores after 9 weeks and 1 reporting relief of movement and other symptoms in 6 people across 24 months.

This 2025 randomized controlled trial in Toronto, Canada, was the largest test of light therapy for Parkinson's disease so far. 63 participants were split into 32 given active treatment and 31 given a sham device, applying 635, 810, and 904 nm light to the head, the back of the neck, and the abdomen for 24 minutes per session, 3 times a week across 56 days, and everyone kept up vigorous exercise throughout. Neither the first 8-week blinded stage nor a second 8-week stage of active treatment for everyone produced a significant difference between the groups, apart from a small gain in a thinking-skills score in the sham group in the first stage. In a third stage lasting up to 48 weeks, participants chose whether to carry on, and those who did improved significantly on the main timed movement test compared with those who stopped, along with anxiety and day-to-day movement scores.

Whether combining exercise with light therapy would add to what either does alone was the question behind a 2025 randomized controlled trial run from a university in Oviedo. 72 people with Parkinson's disease were assigned to 6 months of aerobic exercise, 808 nm light therapy, both together, or neither, and 53 completed enough assessments to be analyzed. Exercise improved motor scores and walking speed over 10 meters, while light therapy alone improved measures of how steadily people stood. Neither approach changed grip strength or arm function, and putting the 2 together produced no extra benefit on any measure.

Across pre-clinical work and a small number of clinical trials covering 199 participants, a 2024 review from a research institute in Sydney, Australia, examined photobiomodulation for Parkinson's disease. It covered wavelengths from 635 to 1068 nm delivered to the temples, the head, the nose, the neck and the abdomen, at 10 to 60 mW and 2 to 4 J/cm². The review reported a strong safety record for light applied both to the head and to other parts of the body, in contrast to the side effects of standard medication. It concluded that the small trials so far point to benefit in both movement and non-movement symptoms, while stressing that larger and longer controlled trials are needed.

Total motor scores improved 23% with active treatment and 24% with sham in a 2023 triple-blind randomized controlled trial of a light therapy helmet for Parkinson's disease. 40 people in Australia were randomly assigned, and 37 were analyzed, with the active helmet delivering 12 minutes of 635 nm red light followed by 12 minutes of 810 nm infrared light, totaling 1137 J per session, 6 days a week across 84 days. The difference between the groups was not significant, which the trial attributed to a strong placebo response. Looking at parts of the scale rather than the total, facial and lower-limb scores improved significantly with active treatment, and about 70% of those given active treatment improved on every part of the scale, while sham responders improved only on lower-limb scores.

Re-analyzing stool samples from an earlier trial, a 2022 observational study examined whether light therapy changes the mix of bacteria in the gut of people with Parkinson's disease. The same 12 participants received 904 nm light at 30 mW and 3.6 to 240 J to the abdomen, neck, head, and nose for 30 minutes per session across 84 days. The balance between the 2 most common groups of gut bacteria moved from 4.60 to 1.58 after treatment, with 9 of the 12 showing a lower ratio. None of the differences reached statistical significance, and the overall variety of bacteria was largely unchanged, so the study described the pattern as a trend rather than a proven effect.

In 2023, a double-blind randomized controlled trial in Sydney, Australia, tested a light therapy helmet against a sham helmet in 40 people with Parkinson's disease. Participants aged 59 to 85 treated themselves at home for 24 minutes a day, 6 days a week, receiving 72 sessions of 635 and 810 nm light at 27 and 52 mW across 84 days, monitored by video call. Average motor scores decreased from 26.8 to 20.4 in the sham group and from 21.3 to 16.5 in the active group over the first 12 weeks, with no significant difference between them at any assessment. The sham group then received 12 weeks of active treatment and reached 12.2, while the active group stopped treatment and finished at 15.3. Treatment was well tolerated, with 2 minor reactions possibly linked to the device: 1 of temporary leg weakness and 1 of reduced fine hand movement.

1, 2 or 3 sessions a week were compared in a 2021 placebo-controlled randomized controlled trial of laser light for Parkinson's disease in Brisbane, Australia. 22 participants received either sham or active 904 nm light at 60 mW per diode and 50 Hz, delivering 2 J for 33 seconds to each of 21 points on the head and inside the mouth, across 28 days, with a second treatment phase after a 4-week break. None of the physical or thinking measures changed significantly, and a placebo effect showed up in the results. A hand-dexterity test did move by an amount the study called clinically meaningful, and measures of writing and stepping shifted in a positive direction with regular treatment. The study concluded that this combination needs to be given at least 2 to 3 times a week for at least 4 weeks before any improvement becomes visible.